Energy-saving control method and device for oxygen production molecular sieve

Abstract

The invention discloses an energy-saving control method and device for an oxygen production molecular sieve, and relates to the technical field of intelligent control. The method comprises the following steps: obtaining to-be-treated air of an oxygen generation molecular sieve assembly according to an oxygen generation system; obtaining air detection data, performing control analysis on the oxygen generation molecular sieve assembly, and establishing a molecular sieve control strategy first space; setting a multi-feature oxygen generation target; performing multilateral constraint optimization on the molecular sieve control strategy first space, and establishing a molecular sieve control strategy second space; establishing a multi-target coupling oxygen generation optimization function; performing coupling optimization on the molecular sieve control strategy second space to generate a molecular sieve control optimization result; and based on the molecular sieve control optimization result and the oxygen production molecular sieve assembly, oxygen production is carried out according to the to-be-treated air. The technical problems that in the prior art, an oxygen generation molecular sieve is high in energy consumption and low in oxygen generation quality are solved, energy-saving control over the oxygen generation process is achieved through intelligent control, and the technical effects of improving the oxygen generation quality and prolonging the service life of the oxygen generation molecular sieve are achieved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent control technology, and in particular to an energy-saving control method and device for an oxygen-producing molecular sieve. Background Art

[0002] In the oxygen production system, the oxygen production molecular sieve is a key core component, and its performance directly affects the oxygen production quality and energy consumption of the oxygen production system. However, in the prior art, the oxygen production system usually only focuses on the optimization of a single goal, such as pursuing high oxygen production or high purity, while ignoring the overall energy efficiency of the system and the management of the molecular sieve life. This single-goal optimization method often leads to energy waste and excessive wear of the molecular sieve, shortening the service life of the equipment. With the increasing requirements of industry for energy conservation and environmental protection, how to reduce energy consumption and extend the service life of the molecular sieve while ensuring the quality of oxygen production has become an important issue that needs to be urgently solved in the field of oxygen production technology. Summary of the invention

[0003] The present application provides an energy-saving control method and device for an oxygen-producing molecular sieve, which solves the technical problems of high energy consumption and low oxygen production quality of the oxygen-producing molecular sieve in the prior art.

[0004] In view of the above problems, the present application provides an energy-saving control method and device for oxygen-producing molecular sieves.

[0005] In a first aspect of the present application, a method for energy-saving control of an oxygen-producing molecular sieve is provided, the method comprising: According to the oxygen production system, the air to be processed of the oxygen production molecular sieve component is obtained, wherein the oxygen production molecular sieve component includes a plurality of oxygen production molecular sieves; the air detection data of the air to be processed is obtained, and the oxygen production molecular sieve component is controlled and analyzed in combination with the molecular sieve control network to establish a first space of the molecular sieve control strategy that meets the capacity constraint of the first strategy; based on the air to be processed, a multi-feature oxygen production target is set, wherein the multi-feature oxygen production target includes a target oxygen production output, a target oxygen production purity, a target oxygen production energy consumption and a molecular sieve life attenuation impact threshold; based on the multi-feature oxygen production target, a multilateral constraint optimization is performed on the first space of the molecular sieve control strategy to establish a second space of the molecular sieve control strategy that meets the capacity constraint of the second strategy; based on the multi-feature oxygen production target, a multi-objective coupled oxygen production optimization function is established; based on the multi-objective coupled oxygen production optimization function, a multi-objective coupled optimization is performed on the second space of the molecular sieve control strategy to generate a molecular sieve control optimization result; based on the molecular sieve control optimization result and the oxygen production molecular sieve component, oxygen is produced according to the air to be processed.

[0006] The second aspect of the present application provides an energy-saving control device for oxygen-producing molecular sieves, the device comprising: To-be-processed air acquisition module, which is used to obtain the to-be-processed air of the oxygen generation molecular sieve assembly according to the oxygen generation system, wherein the oxygen generation molecular sieve assembly includes a plurality of oxygen generation molecular sieves; analysis module, which is used to obtain the air detection data of the to-be-processed air and perform control analysis on the oxygen generation molecular sieve assembly in combination with the molecular sieve control network to establish the first space of the molecular sieve control strategy that meets the first strategy capacity constraint; target setting module, which is used to set multi-feature oxygen generation targets based on the to-be-processed air, wherein the multi-feature oxygen generation targets include target oxygen generation output, target oxygen generation purity, target oxygen generation energy consumption, and molecular sieve life attenuation influence threshold; first optimization module, which is used to perform multi-sided constraint optimization on the first space of the molecular sieve control strategy based on the multi-feature oxygen generation targets to establish the second space of the molecular sieve control strategy that meets the second strategy capacity constraint; function establishment module, which is used to establish a multi-objective coupled oxygen generation optimization function based on the multi-feature oxygen generation targets; second optimization module, which is used to perform multi-objective coupled optimization on the second space of the molecular sieve control strategy according to the multi-objective coupled oxygen generation optimization function to generate a molecular sieve control optimization result; control module, which is used to generate oxygen based on the to-be-processed air based on the molecular sieve control optimization result and the oxygen generation molecular sieve assembly.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages: First, according to the oxygen generation system, obtain the to-be-processed air of the oxygen generation molecular sieve assembly, wherein the oxygen generation molecular sieve assembly includes a plurality of oxygen generation molecular sieves. Then, obtain the air detection data of the to-be-processed air and perform control analysis on the oxygen generation molecular sieve assembly in combination with the molecular sieve control network to establish the first space of the molecular sieve control strategy that meets the first strategy capacity constraint. Further, set multi-feature oxygen generation targets based on the to-be-processed air, wherein the multi-feature oxygen generation targets include target oxygen generation output, target oxygen generation purity, target oxygen generation energy consumption, and molecular sieve life attenuation influence threshold. Then, perform multi-sided constraint optimization on the first space of the molecular sieve control strategy based on the multi-feature oxygen generation targets to establish the second space of the molecular sieve control strategy that meets the second strategy capacity constraint. Next, establish a multi-objective coupled oxygen generation optimization function based on the multi-feature oxygen generation targets; and perform multi-objective coupled optimization on the second space of the molecular sieve control strategy according to the multi-objective coupled oxygen generation optimization function to generate a molecular sieve control optimization result. Finally, generate oxygen based on the to-be-processed air based on the molecular sieve control optimization result and the oxygen generation molecular sieve assembly. This solves the technical problems of high energy consumption and low oxygen generation quality of the oxygen generation molecular sieve in the prior art, realizes energy-saving control of the oxygen generation process through intelligent control, and achieves the technical effects of improving the oxygen generation quality and extending the service life of the oxygen generation molecular sieve. Description of the Drawings

[0008] To more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the accompanying drawings required for the description of the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0009] Figure 1 Schematic flow diagram of the energy-saving control method for the oxygen-making molecular sieve provided by the embodiment of the present application; Figure 2 Schematic structural diagram of the energy-saving control device for the oxygen-making molecular sieve provided by the embodiment of the present application.

[0010] Explanation of reference numerals: the module 11 for obtaining the air to be processed, the parsing module 12, the target setting module 13, the first optimization module 14, the function establishment module 15, the second optimization module 16, and the control module 17. Specific embodiments

[0011] By providing an energy-saving control method and device for an oxygen-making molecular sieve, the present application solves the technical problems of high energy consumption and low oxygen-making quality of the oxygen-making molecular sieve in the prior art.

[0012] The following will clearly and completely describe the technical solutions in the embodiments of the present application in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, rather than all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0013] It should be noted that the terms "including" and "having" are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units does not necessarily have to be limited to those clearly listed steps or units, but may include other steps or modules that are not clearly listed or are inherent to these processes, methods, products, or devices.

[0014] Embodiment 1, as Figure 1 shown, the present application provides an energy-saving control method for an oxygen-making molecular sieve, wherein the method includes: Obtain the air to be processed of the oxygen-making molecular sieve assembly according to the oxygen-making system, wherein the oxygen-making molecular sieve assembly includes a plurality of oxygen-making molecular sieves.

[0015] In the oxygen generation system, the air to be processed is obtained from the oxygen generation molecular sieve assembly. The air to be processed refers to the air that has been compressed and had impurities removed, but has not yet been processed by the oxygen generation molecular sieve assembly. The oxygen generation molecular sieve assembly is composed of multiple oxygen generation molecular sieves, and each molecular sieve undertakes a part of the air separation task respectively.

[0016] Obtain the air detection data of the air to be processed, and combine it with the molecular sieve control network to perform control analysis on the oxygen generation molecular sieve assembly, and establish the first space of the molecular sieve control strategy that meets the first strategy capacity constraint.

[0017] By detecting the air to be processed to obtain the air detection data of the air to be processed, including but not limited to temperature, humidity, pressure, oxygen concentration, impurity content, etc.; input the air detection data into the molecular sieve control network to perform control analysis on the oxygen generation molecular sieve assembly, that is, determine the optimal working state and parameter settings of the molecular sieve assembly according to the current air conditions and oxygen generation targets. Among them, the molecular sieve control network can quickly analyze and process the input data to evaluate the impact of the current air state on the oxygen generation molecular sieve assembly. On the basis of the control analysis, establish the first space of the molecular sieve control strategy that meets the first strategy capacity constraint. The first strategy capacity constraint refers to a series of conditions that the molecular sieve control strategy set by the system must meet according to the current resource and capacity limitations, including the maximum oxygen generation amount, the minimum oxygen generation purity, the maximum energy consumption, and the maximum working load of the molecular sieve. By meeting these constraint conditions, the feasibility and effectiveness of the molecular sieve control strategy in actual operation can be ensured.

[0018] Furthermore, obtaining the air detection data of the air to be processed, and combining it with the molecular sieve control network to perform control analysis on the oxygen generation molecular sieve assembly, and establishing the first space of the molecular sieve control strategy that meets the first strategy capacity constraint, includes: The molecular sieve control network includes multiple molecular sieve control indicators. The multiple molecular sieve control indicators include adsorption pressure, desorption pressure, adsorption time, desorption time, molecular sieve temperature, inlet air flow rate, and molecular sieve regeneration cycle; based on the air detection data, perform control adaptation mining on the multiple oxygen generation molecular sieves according to the molecular sieve control network to obtain multiple molecular sieve control adaptation domains; perform control decision-making on the multiple oxygen generation molecular sieves based on the multiple molecular sieve control adaptation domains to obtain multiple molecular sieve control decision domains. Among them, each molecular sieve control decision domain includes multiple molecular sieve control decisions corresponding to each oxygen generation molecular sieve; based on the multiple oxygen generation molecular sieves, perform control decision-making combination on the multiple molecular sieve control decision domains to obtain the first space of the molecular sieve control strategy. Among them, the first space of the molecular sieve control strategy includes multiple molecular sieve control strategies, and the number of the multiple molecular sieve control strategies meets the first strategy capacity constraint.

[0019] Specifically, the molecular sieve control network includes a series of multivariate indicators for precisely controlling the working state of the molecular sieve. These indicators include, but are not limited to, adsorption pressure, desorption pressure, adsorption time, desorption time, molecular sieve temperature, inlet air flow rate, and the molecular sieve regeneration cycle. Based on the obtained air detection data, the molecular sieve control network is used to conduct control adaptation mining on multiple oxygen-producing molecular sieves, that is, according to the current air conditions, explore the adaptability of each molecular sieve under different control conditions to identify the optimal working range of each molecular sieve under different operating parameters, and form multiple molecular sieve control adaptation domains. After obtaining multiple molecular sieve control adaptation domains, control decisions are made for each oxygen-producing molecular sieve according to these adaptation domains to generate multiple molecular sieve control decision domains. Each molecular sieve control decision domain contains multiple control decisions for the corresponding molecular sieve. The multiple molecular sieve control decision domains are combined to form the first space of the molecular sieve control strategy that meets the first strategy capacity constraint. The first space of the molecular sieve control strategy includes multiple molecular sieve control strategies, and the number of the multiple molecular sieve control strategies meets the first strategy capacity constraint. Each molecular sieve control strategy includes control decisions corresponding to each oxygen-producing molecular sieve of the oxygen-producing molecular sieve assembly.

[0020] Furthermore, based on the air detection data, control adaptation mining is conducted on the multiple oxygen-producing molecular sieves according to the molecular sieve control network to obtain multiple molecular sieve control adaptation domains, including: Collect control constraint parameters of the multiple oxygen-producing molecular sieves according to the molecular sieve control network to obtain multiple molecular sieve control constraint domains; conduct control record retrieval on the multiple oxygen-producing molecular sieves based on the air detection data to obtain multiple molecular sieve control retrieval domains; respectively conduct control trigger feature recognition on the multiple molecular sieve control retrieval domains based on the molecular sieve control network to obtain multiple molecular sieve control trigger domains, where each molecular sieve control trigger domain includes multiple control trigger feature intervals corresponding to each oxygen-producing molecular sieve; based on the multiple oxygen-producing molecular sieves, conduct intersection analysis on the multiple molecular sieve control trigger domains and the multiple molecular sieve control constraint domains to generate the multiple molecular sieve control adaptation domains.

[0021] Specifically, according to the molecular sieve control network, acquisition of control constraint parameters for multiple oxygen-producing molecular sieves is carried out. These constraint parameters are the boundary conditions that the molecular sieve must abide by under normal working conditions, including but not limited to adsorption pressure range, desorption pressure range, temperature limit, flow rate limit, etc. By collecting these parameters, multiple molecular sieve control constraint domains are obtained. The molecular sieve control constraint domain defines the feasible operation space of the molecular sieve under different working conditions; retrieval of historical control records for multiple oxygen-producing molecular sieves is carried out based on air detection data. The historical control records contain the actual working state and performance of the molecular sieve under different air conditions. By retrieving the historical control records, multiple molecular sieve control retrieval domains can be obtained. The molecular sieve control retrieval domain reflects the actual response ability and performance characteristics of the molecular sieve under different conditions; further analysis of multiple molecular sieve control retrieval domains is carried out using the molecular sieve control network to identify control trigger characteristics. The control trigger characteristic refers to the key parameter or condition interval that can trigger the change of the molecular sieve state or performance optimization. By identifying the control trigger characteristics, multiple molecular sieve control trigger domains are obtained. Each molecular sieve control trigger domain contains multiple control trigger characteristic intervals for a specific molecular sieve; based on the actual situation of multiple oxygen-producing molecular sieves, intersection analysis of multiple molecular sieve control trigger domains and multiple molecular sieve control constraint domains is carried out to find the operation space that simultaneously meets the control trigger conditions and constraint conditions, that is, the molecular sieve control adaptation domain; through intersection analysis, multiple molecular sieve control adaptation domains are generated. Each molecular sieve control adaptation domain represents the optimal working state range of the molecular sieve under specific conditions.

[0022] Based on the air to be processed, set multiple characteristic oxygen production targets, where the multiple characteristic oxygen production targets include target oxygen production, target oxygen purity, target oxygen production energy consumption, and the influence threshold of molecular sieve life attenuation.

[0023] Based on the characteristics of the air to be processed, set multiple oxygen production targets, which cover multiple key characteristics that need to be optimized during the oxygen production process. Specifically, the multiple characteristic oxygen production targets include the following aspects: The target oxygen production refers to the amount of oxygen that needs to be achieved within a certain period of time; the target oxygen purity refers to the purity requirement of the produced oxygen to ensure that the finally produced oxygen meets the predetermined quality standard; the target oxygen production energy consumption refers to the energy consumed by the system during the oxygen production process; the influence threshold of molecular sieve life attenuation refers to the degree to which the life of the molecular sieve is affected by attenuation during the oxygen production process. By setting this threshold, the system can control the usage intensity of the molecular sieve while meeting other targets, extend its service life, and avoid excessive wear or premature replacement.

[0024] Based on the multiple characteristic oxygen production targets, perform multi-sided constraint optimization on the first space of the molecular sieve control strategy to establish the second space of the molecular sieve control strategy that meets the second strategy capacity constraint.

[0025] Taking the multi-feature oxygen generation target as a constraint condition, perform multi-sided optimization on the first space of the molecular sieve control strategy, that is, find one or more control strategies that can achieve the best operating state while meeting all feature targets; after multi-sided constraint optimization, a group of control strategies that meet these multi-feature targets will be screened out, and these strategies together constitute the second space of the molecular sieve control strategy.

[0026] Furthermore, perform multi-sided constraint optimization on the first space of the molecular sieve control strategy based on the multi-feature oxygen generation target, and establish a second space of the molecular sieve control strategy that meets the second strategy capacity constraint, including: Extract the first molecular sieve control strategy according to the first space of the molecular sieve control strategy; perform multi-feature prediction on the first molecular sieve control strategy based on the air detection data to obtain the first molecular sieve control prediction result; establish multi-sided constraint optimization rules based on the multi-feature oxygen generation target, where the multi-sided constraint optimization rules include that the predicted oxygen generation output is greater than or equal to the target oxygen generation output, the predicted oxygen generation purity is greater than or equal to the target oxygen generation purity, the predicted oxygen generation energy consumption is less than or equal to the target oxygen generation energy consumption, and the predicted molecular sieve life attenuation influence coefficient is less than or equal to the molecular sieve life attenuation influence threshold; judge whether the first molecular sieve control prediction result meets the multi-sided constraint optimization rules; if the first molecular sieve control prediction result meets the multi-sided constraint optimization rules, record the first molecular sieve control strategy as the first molecular sieve control optimization strategy and add the first molecular sieve control optimization strategy to the second space of the molecular sieve control strategy; continue to perform multi-sided constraint optimization on the first space of the molecular sieve control strategy based on the multi-sided constraint optimization rules and the second strategy capacity constraint to obtain the second space of the molecular sieve control strategy.

[0027] Specifically, a preliminary control strategy is extracted from the first space of the molecular sieve control strategy, called the first molecular sieve control strategy, as the starting point for further optimization; based on the current air detection data, multi-feature prediction is performed on the extracted first molecular sieve control strategy to generate the first molecular sieve control prediction result; according to the multi-feature oxygen production target, a set of multi-sided constraint optimization rules is established. The multi-sided constraint optimization rules include that the predicted oxygen production should be greater than or equal to the target oxygen production, the predicted oxygen purity should be greater than or equal to the target oxygen purity, the predicted oxygen production energy consumption should be less than or equal to the target oxygen production energy consumption, and the predicted molecular sieve life attenuation influence coefficient should be less than or equal to the molecular sieve life attenuation influence threshold; check whether the first molecular sieve control prediction result meets the multi-sided constraint optimization rules. If all conditions are met, it is considered that the strategy has achieved the expected effect on each target. If the first molecular sieve control prediction result conforms to the multi-sided constraint optimization rules, mark this control strategy as the first molecular sieve control optimization strategy and add it to the second space of the molecular sieve control strategy; perform the same multi-sided constraint optimization on other strategies in the first space of the molecular sieve control strategy, verify them one by one according to the above process, and add the strategies that meet the conditions to the second space of the molecular sieve control strategy; when the second space of the molecular sieve control strategy reaches the capacity constraint or no new strategy that meets all constraint conditions can be found anymore, the multi-sided constraint optimization process ends. At this time, the control strategy set stored in the second space of the molecular sieve control strategy is the one that meets the multi-feature oxygen production target and multi-sided constraint conditions.

[0028] Furthermore, based on the air detection data, multi-feature prediction is performed on the first molecular sieve control strategy to obtain the first molecular sieve control prediction result, including Based on the oxygen production molecular sieve component, oxygen production record backtracking is performed to obtain multiple oxygen production record groups. Among them, each oxygen production record group includes historical air detection data, historical molecular sieve control strategy, historical oxygen production, historical oxygen purity, historical oxygen production energy consumption, and historical molecular sieve life attenuation influence coefficient; Based on the multiple oxygen production record groups, a molecular sieve control prediction model that meets the prediction accuracy constraint is trained. The molecular sieve control prediction model includes an input layer, a multi-feature prediction parallel layer, and an output layer. The multi-feature prediction parallel layer includes an oxygen production prediction layer, an oxygen purity prediction layer, an oxygen production energy consumption prediction layer, and a molecular sieve life attenuation influence prediction layer; based on the air detection data and the first molecular sieve control strategy, according to the molecular sieve control prediction model, the first molecular sieve control prediction result is output.

[0029] Specifically, retrospective analysis is carried out based on the historical operation records of the oxygen generation molecular sieve assembly to generate multiple oxygen generation record groups. Each oxygen generation record group contains the following historical data: historical air detection data, historical molecular sieve control strategy, historical oxygen generation output, historical oxygen generation purity, historical oxygen generation energy consumption, and historical molecular sieve life attenuation influence coefficient. Based on these oxygen generation record groups, a molecular sieve control prediction model that meets the prediction accuracy constraint is trained. The molecular sieve control prediction model includes an input layer, a multi-feature prediction parallel layer, and an output layer. Among them, the input layer receives the current air detection data and the molecular sieve control strategy as inputs. The multi-feature prediction parallel layer is responsible for predicting the results of multiple key features respectively, specifically including: an oxygen generation output prediction layer, an oxygen generation purity prediction layer, an oxygen generation energy consumption prediction layer, and a molecular sieve life attenuation influence prediction layer. The output layer refers to generating the final prediction output according to the results of the multi-feature prediction parallel layer. After the model training is completed, the current air detection data and the first molecular sieve control strategy are input into the molecular sieve control prediction model. The model will perform multi-feature prediction based on the input data and output the first molecular sieve control prediction results. These prediction results include multiple key indicators such as oxygen generation output, oxygen generation purity, oxygen generation energy consumption, and molecular sieve life attenuation influence coefficient, which can provide data support for subsequent strategy optimization.

[0030] Based on the multi-feature oxygen generation target, a multi-objective coupled oxygen generation optimization function is established.

[0031] Furthermore, the multi-objective coupled oxygen generation optimization function is: ; where COP represents the coupled oxygen generation optimization coefficient, exp represents the exponential function with the natural constant e as the base, OPX represents the predicted oxygen generation output, OPXO represents the target oxygen generation output, OCI represents the predicted oxygen generation purity, OCIO represents the target oxygen generation purity, OEK represents the predicted oxygen generation energy consumption, OEKO represents the target oxygen generation energy consumption, DIC represents the predicted molecular sieve life attenuation influence coefficient, DICO represents the molecular sieve life attenuation influence threshold, represents the coupled oxygen generation optimization predetermined weight condition.

[0032] According to the multi-feature oxygen generation target, a multi-objective coupled oxygen generation optimization function is constructed: ; The multi-objective coupled oxygen production optimization function comprehensively considers multiple oxygen production objectives to optimize the performance of the oxygen production system; among them, COP represents the coupled oxygen production optimization coefficient, which is used to evaluate the overall quality of the current oxygen production strategy, and the higher the value, the better the strategy; exp represents the exponential function with the natural constant e as the base, which is used to expand or contract the proportion of the result; OPX represents the predicted oxygen production; OPXO represents the target oxygen production; OCI represents the predicted oxygen purity; OCIO represents the target oxygen purity; OEK represents the predicted oxygen production energy consumption; OEKO represents the target oxygen production energy consumption; DIC represents the predicted influence coefficient of molecular sieve life attenuation; DICO represents the molecular sieve life attenuation influence threshold; respectively represent the weights of oxygen production, purity, energy consumption, and the influence of molecular sieve life attenuation, which are set according to different optimization requirements and affect the contribution of each part to the final COP.

[0033] Perform multi-objective coupled optimization on the second space of the molecular sieve control strategy according to the multi-objective coupled oxygen production optimization function to generate the molecular sieve control optimization result.

[0034] Substitute each strategy in the second space of the molecular sieve control strategy into the multi-objective coupled oxygen production optimization function. Through the multi-objective coupled oxygen production optimization function, evaluate the comprehensive performance of each strategy on multiple objectives such as production, purity, energy consumption, and molecular sieve life attenuation; for each strategy, calculate its coupled oxygen production optimization coefficient (COP). The higher the COP value, the better the performance of the strategy in comprehensively optimizing multiple objectives; according to the calculated COP values, sort and screen all strategies, select the strategy with the highest COP value, or appropriately combine and adjust multiple preferred strategies to generate the final optimization result of molecular sieve control. The molecular sieve control optimization result is the optimized control strategy.

[0035] Furthermore, performing multi-objective coupled optimization on the second space of the molecular sieve control strategy according to the multi-objective coupled oxygen production optimization function to generate the molecular sieve control optimization result includes According to the second space of the molecular sieve control strategy, extract the nth molecular sieve control optimization strategy, where n is a positive integer; according to the multi-objective coupled oxygen production optimization function, calculate the nth coupled oxygen production optimization coefficient corresponding to the nth molecular sieve control optimization strategy; according to the second space of the molecular sieve control strategy, extract the (n + 1)th molecular sieve control optimization strategy, and according to the multi-objective coupled oxygen production optimization function, calculate the (n + 1)th coupled oxygen production optimization coefficient; according to the nth coupled oxygen production optimization coefficient and the (n + 1)th coupled oxygen production optimization coefficient, perform a winning analysis on the nth molecular sieve control optimization strategy and the (n + 1)th molecular sieve control optimization strategy to determine the current winning coupled oxygen production optimization coefficient and the current winning molecular sieve control optimization strategy corresponding to the current winning coupled oxygen production optimization coefficient; based on the current winning coupled oxygen production optimization coefficient, perform iterative optimization on the current winning molecular sieve control optimization strategy according to the multi-objective coupled oxygen production optimization function and the second space of the molecular sieve control strategy to determine the molecular sieve control optimization result that meets the predetermined number of iterative optimization times.

[0036] Preferably, extract the nth molecular sieve control optimization strategy (n is a positive integer) from the second space of the molecular sieve control strategy as the current optimization object; use the multi-objective coupled oxygen production optimization function to calculate the coupled oxygen production optimization coefficient (COP) corresponding to the nth molecular sieve control optimization strategy. The coupled oxygen production optimization coefficient represents the comprehensive performance of the strategy in multi-objective optimization; extract the (n + 1)th molecular sieve control optimization strategy from the second space of the molecular sieve control strategy and also use the multi-objective coupled oxygen production optimization function to calculate its coupled oxygen production optimization coefficient (COP); compare the coupled oxygen production optimization coefficients of the nth and (n + 1)th molecular sieve control optimization strategies, and perform a winning analysis based on these two COP values. By comparison, determine the currently better-performing coupled oxygen production optimization coefficient (i.e., the current winning coupled oxygen production optimization coefficient) and the corresponding molecular sieve control optimization strategy (i.e., the current winning molecular sieve control optimization strategy); based on the current winning coupled oxygen production optimization coefficient, continue to perform iterative optimization in the second space of the molecular sieve control strategy, that is, based on the current winning strategy, continue to extract new strategies for comparison and optimization, and repeat the above process; according to the preset number of iterative optimization times, continuously perform the above-mentioned winning analysis and strategy iteration. Finally, determine a molecular sieve control optimization result that has been optimized through multiple rounds of iteration, and this result achieves the best balance among multiple objectives.

[0037] Based on the molecular sieve control optimization result and the oxygen-producing molecular sieve assembly, oxygen is produced according to the air to be processed.

[0038] After generating the final molecular sieve control optimization result, oxygen production will be carried out on the air to be processed based on these optimized control strategies and in combination with the actual operating conditions of the oxygen-producing molecular sieve assembly.

[0039] In summary, the embodiments of the present application have at least the following technical effects: First, according to the oxygen generation system, the air to be processed of the oxygen generation molecular sieve assembly is obtained, wherein the oxygen generation molecular sieve assembly includes a plurality of oxygen generation molecular sieves. Then, the air detection data of the air to be processed is obtained, and the oxygen generation molecular sieve assembly is controlled and analyzed in combination with the molecular sieve control network to establish the first space of the molecular sieve control strategy that meets the first strategy capacity constraint. Further, based on the air to be processed, multi-feature oxygen generation targets are set, where the multi-feature oxygen generation targets include the target oxygen generation output, the target oxygen generation purity, the target oxygen generation energy consumption, and the molecular sieve life attenuation influence threshold. Then, based on the multi-feature oxygen generation targets, multi-sided constraint optimization is performed on the first space of the molecular sieve control strategy to establish the second space of the molecular sieve control strategy that meets the second strategy capacity constraint. Next, based on the multi-feature oxygen generation targets, a multi-objective coupled oxygen generation optimization function is established; and according to the multi-objective coupled oxygen generation optimization function, multi-objective coupled optimization is performed on the second space of the molecular sieve control strategy to generate the molecular sieve control optimization result. Finally, based on the molecular sieve control optimization result and the oxygen generation molecular sieve assembly, oxygen is generated according to the air to be processed. The technical problem of high energy consumption and low oxygen generation quality of the oxygen generation molecular sieve in the prior art is solved, and energy-saving control of the oxygen generation process is realized through intelligent control, achieving the technical effects of improving the oxygen generation quality and extending the service life of the oxygen generation molecular sieve.

[0040] Embodiment 2, based on the same inventive concept as the energy-saving control method of the oxygen generation molecular sieve in the foregoing embodiment, as Figure 2 shown, the present application provides an energy-saving control device for an oxygen generation molecular sieve, wherein the device includes: The to-be-processed air acquisition module 11 is configured to obtain the to-be-processed air of the oxygen generation molecular sieve assembly according to the oxygen generation system, wherein the oxygen generation molecular sieve assembly includes a plurality of oxygen generation molecular sieves; the analysis module 12 is configured to obtain the air detection data of the to-be-processed air, and perform control analysis on the oxygen generation molecular sieve assembly in combination with the molecular sieve control network to establish a first space of the molecular sieve control strategy that meets the first strategy capacity constraint; the target setting module 13 is configured to set a multi-feature oxygen generation target based on the to-be-processed air, wherein the multi-feature oxygen generation target includes a target oxygen generation output, a target oxygen generation purity, a target oxygen generation energy consumption, and a molecular sieve life attenuation influence threshold; the first optimization module 14 is configured to perform multi-sided constraint optimization on the first space of the molecular sieve control strategy based on the multi-feature oxygen generation target to establish a second space of the molecular sieve control strategy that meets the second strategy capacity constraint; the function establishment module 15 is configured to establish a multi-objective coupled oxygen generation optimization function based on the multi-feature oxygen generation target; the second optimization module 16 is configured to perform multi-objective coupled optimization on the second space of the molecular sieve control strategy according to the multi-objective coupled oxygen generation optimization function to generate a molecular sieve control optimization result; the control module 17 is configured to perform oxygen generation according to the to-be-processed air based on the molecular sieve control optimization result and the oxygen generation molecular sieve assembly.

[0041] Further, the analysis module 12 is configured to execute the following method: The molecular sieve control network includes molecular sieve control multiple indicators, and the molecular sieve control multiple indicators include adsorption pressure, desorption pressure, adsorption time, desorption time, molecular sieve temperature, inlet air flow rate, and molecular sieve regeneration period; based on the air detection data, perform control adaptation mining on the plurality of oxygen generation molecular sieves according to the molecular sieve control network to obtain a plurality of molecular sieve control adaptation domains; perform control decision-making on the plurality of oxygen generation molecular sieves based on the plurality of molecular sieve control adaptation domains to obtain a plurality of molecular sieve control decision domains, wherein each molecular sieve control decision domain includes a plurality of molecular sieve control decisions corresponding to each oxygen generation molecular sieve; perform control decision combination on the plurality of molecular sieve control decision domains based on the plurality of oxygen generation molecular sieves to obtain the first space of the molecular sieve control strategy, wherein the first space of the molecular sieve control strategy includes a plurality of molecular sieve control strategies, and the number of the plurality of molecular sieve control strategies meets the first strategy capacity constraint.

[0042] Further, the analysis module 12 is configured to execute the following method: Collect control constraint parameters for the multiple oxygen-producing molecular sieves according to the molecular sieve control network to obtain multiple molecular sieve control constraint domains; retrieve control records for the multiple oxygen-producing molecular sieves based on the air detection data to obtain multiple molecular sieve control retrieval domains; respectively identify control trigger features for the multiple molecular sieve control retrieval domains based on the molecular sieve control network to obtain multiple molecular sieve control trigger domains, where each molecular sieve control trigger domain includes multiple control trigger feature intervals corresponding to each oxygen-producing molecular sieve; perform intersection analysis on the multiple molecular sieve control trigger domains and the multiple molecular sieve control constraint domains based on the multiple oxygen-producing molecular sieves to generate the multiple molecular sieve control adaptation domains.

[0043] Further, the first optimization module 14 is used to execute the following method: Extract the first molecular sieve control strategy according to the first space of the molecular sieve control strategy; perform multi-feature prediction on the first molecular sieve control strategy based on the air detection data to obtain the first molecular sieve control prediction result; establish a multi-sided constraint optimization rule based on the multi-feature oxygen production target, where the multi-sided constraint optimization rule includes that the predicted oxygen production is greater than or equal to the target oxygen production, the predicted oxygen purity is greater than or equal to the target oxygen purity, the predicted oxygen production energy consumption is less than or equal to the target oxygen production energy consumption, and the predicted molecular sieve life attenuation influence coefficient is less than or equal to the molecular sieve life attenuation influence threshold; determine whether the first molecular sieve control prediction result satisfies the multi-sided constraint optimization rule; if the first molecular sieve control prediction result satisfies the multi-sided constraint optimization rule, record the first molecular sieve control strategy as the first molecular sieve control optimization strategy and add the first molecular sieve control optimization strategy to the second space of the molecular sieve control strategy; continue to perform multi-sided constraint optimization on the first space of the molecular sieve control strategy based on the multi-sided constraint optimization rule and the second strategy capacity constraint to obtain the second space of the molecular sieve control strategy.

[0044] Further, the first optimization module 14 is used to execute the following method: Based on the oxygen generation molecular sieve assembly, oxygen generation record backtracking is performed to obtain multiple oxygen generation record groups. Each oxygen generation record group includes historical air detection data, historical molecular sieve control strategies, historical oxygen generation output, historical oxygen generation purity, historical oxygen generation energy consumption, and historical molecular sieve life attenuation influence coefficients. Based on the multiple oxygen generation record groups, a molecular sieve control prediction model that meets the prediction accuracy constraint is trained. The molecular sieve control prediction model includes an input layer, a multi-feature prediction parallel layer, and an output layer. The multi-feature prediction parallel layer includes an oxygen generation output prediction layer, an oxygen generation purity prediction layer, an oxygen generation energy consumption prediction layer, and a molecular sieve life attenuation influence prediction layer. Based on the air detection data and the first molecular sieve control strategy, the first molecular sieve control prediction result is output according to the molecular sieve control prediction model.

[0045] Further, the function establishment module 15 is used to execute the following method: The multi-objective coupled oxygen generation optimization function is: ; where COP represents the coupled oxygen generation optimization coefficient, exp represents the exponential function with the natural constant e as the base, OPX represents the predicted oxygen generation output, OPXO represents the target oxygen generation output, OCI represents the predicted oxygen generation purity, OCIO represents the target oxygen generation purity, OEK represents the predicted oxygen generation energy consumption, OEKO represents the target oxygen generation energy consumption, DIC represents the predicted molecular sieve life attenuation influence coefficient, DICO represents the molecular sieve life attenuation influence threshold, represents the coupled oxygen generation optimization predetermined weight condition.

[0046] Further, the second optimization module 16 is used to execute the following method: According to the second space of the molecular sieve control strategy, the nth molecular sieve control optimization strategy is extracted, where n is a positive integer. According to the multi-objective coupled oxygen generation optimization function, the nth coupled oxygen generation optimization coefficient corresponding to the nth molecular sieve control optimization strategy is calculated. According to the second space of the molecular sieve control strategy, the (n + 1)th molecular sieve control optimization strategy is extracted, and according to the multi-objective coupled oxygen generation optimization function, the (n + 1)th coupled oxygen generation optimization coefficient is calculated. According to the nth coupled oxygen generation optimization coefficient and the (n + 1)th coupled oxygen generation optimization coefficient, a winning analysis is performed on the nth molecular sieve control optimization strategy and the (n + 1)th molecular sieve control optimization strategy to determine the current winning coupled oxygen generation optimization coefficient and the current winning molecular sieve control optimization strategy corresponding to the current winning coupled oxygen generation optimization coefficient. Based on the current winning coupled oxygen generation optimization coefficient, iterative optimization is performed on the current winning molecular sieve control optimization strategy according to the multi-objective coupled oxygen generation optimization function and the second space of the molecular sieve control strategy to determine the molecular sieve control optimization result that meets the predetermined number of iterative optimization times.

[0047] It should be noted that the above order of the embodiments of the present application is only for description and does not represent the superiority or inferiority of the embodiments. And the above specific embodiments of this specification have been described. The processes depicted in the drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0048] The above are only the preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included within the protection scope of the present application.

[0049] This specification and the drawings are only exemplary descriptions of the present application and are considered to have covered any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalent technologies, the present application is intended to include these changes and modifications.

Claims

1. An energy-saving control method for oxygen-producing molecular sieves, characterized in that: The method comprises: According to the oxygen production system, obtaining the air to be processed of the oxygen production molecular sieve assembly, wherein the oxygen production molecular sieve assembly includes a plurality of oxygen production molecular sieves; Acquire air detection data of the air to be treated, and perform control analysis on the oxygen-generating molecular sieve component in combination with a molecular sieve control network to establish a first space of a molecular sieve control strategy that meets a first strategy capacity constraint; Based on the air to be processed, a multi-feature oxygen production target is set, wherein the multi-feature oxygen production target includes a target oxygen production output, a target oxygen production purity, a target oxygen production energy consumption, and a molecular sieve life attenuation impact threshold; Based on the multi-feature oxygen production target, a first space of the molecular sieve control strategy is subjected to multilateral constraint optimization to establish a second space of the molecular sieve control strategy that satisfies the capacity constraint of the second strategy; Based on the multi-feature oxygen production target, a multi-objective coupled oxygen production optimization function is established; Perform multi-objective coupling optimization on the second space of the molecular sieve control strategy according to the multi-objective coupling oxygen production optimization function to generate a molecular sieve control optimization result; Based on the molecular sieve control optimization result and the oxygen-producing molecular sieve assembly, oxygen is produced according to the air to be treated.

2. The method according to claim 1, characterized in that The air detection data of the air to be treated is obtained, and the oxygen-generating molecular sieve component is controlled and analyzed in combination with the molecular sieve control network to establish a first space of the molecular sieve control strategy that meets the first strategy capacity constraint, including: The molecular sieve control network includes molecular sieve control multivariate indicators, and the molecular sieve control multivariate indicators include adsorption pressure, desorption pressure, adsorption time, desorption time, molecular sieve temperature, intake air flow rate and molecular sieve regeneration cycle; Based on the air detection data, control adaptation mining is performed on the multiple oxygen-producing molecular sieves according to the molecular sieve control network to obtain multiple molecular sieve control adaptation domains; Based on the multiple molecular sieve control adaptation domains, control decisions are made on the multiple oxygen-producing molecular sieves to obtain multiple molecular sieve control decision domains, wherein each molecular sieve control decision domain includes multiple molecular sieve control decisions corresponding to each oxygen-producing molecular sieve; Based on the multiple oxygen-producing molecular sieves, control decisions are combined on the multiple molecular sieve control decision domains to obtain the first space of the molecular sieve control strategy, wherein the first space of the molecular sieve control strategy includes multiple molecular sieve control strategies, and the number of the multiple molecular sieve control strategies meets the first strategy capacity constraint.

3. The method according to claim 2, characterized in that Based on the air detection data, control adaptation mining is performed on the multiple oxygen-producing molecular sieves according to the molecular sieve control network to obtain multiple molecular sieve control adaptation domains, including: Collect control constraint parameters of the plurality of oxygen-producing molecular sieves according to the molecular sieve control network to obtain a plurality of molecular sieve control constraint domains; Based on the air detection data, control records of the plurality of oxygen-producing molecular sieves are retrieved to obtain a plurality of molecular sieve control retrieval domains; Based on the molecular sieve control network, control trigger feature identification is performed on the multiple molecular sieve control search domains respectively to obtain multiple molecular sieve control trigger domains, wherein each molecular sieve control trigger domain includes multiple control trigger feature intervals corresponding to each oxygen-producing molecular sieve; Based on the multiple oxygen-producing molecular sieves, an intersection analysis is performed on the multiple molecular sieve control trigger domains and the multiple molecular sieve control constraint domains to generate the multiple molecular sieve control adaptation domains.

4. The method according to claim 1, characterized in that Based on the multi-feature oxygen production target, a first space of the molecular sieve control strategy is subjected to multilateral constraint optimization to establish a second space of the molecular sieve control strategy that satisfies the capacity constraint of the second strategy, including: Extracting a first molecular sieve control strategy according to the first molecular sieve control strategy space; Based on the air detection data, performing multi-feature prediction on the first molecular sieve control strategy to obtain a first molecular sieve control prediction result; Based on the multi-feature oxygen production target, a multilateral constraint optimization rule is established, wherein the multilateral constraint optimization rule includes predicting that the oxygen production output is greater than / equal to the target oxygen production output, predicting that the oxygen production purity is greater than / equal to the target oxygen production purity, predicting that the oxygen production energy consumption is less than / equal to the target oxygen production energy consumption, and predicting that the molecular sieve life attenuation influence coefficient is less than / equal to the molecular sieve life attenuation influence threshold; Determining whether the first molecular sieve control prediction result satisfies the multilateral constraint optimization rule; If the first molecular sieve control prediction result satisfies the multilateral constraint optimization rule, the first molecular sieve control strategy is recorded as the first molecular sieve control optimization strategy, and the first molecular sieve control optimization strategy is added to the second space of the molecular sieve control strategy; Based on the multilateral constraint optimization rule and the second strategy capacity constraint, the first space of the molecular sieve control strategy is continuously subjected to multilateral constraint optimization to obtain the second space of the molecular sieve control strategy.

5. The method according to claim 4, characterized in that Based on the air detection data, a multi-feature prediction is performed on the first molecular sieve control strategy to obtain a first molecular sieve control prediction result, including Based on the oxygen production molecular sieve component, oxygen production records are traced back to obtain multiple oxygen production record groups, wherein each oxygen production record group includes historical air detection data, historical molecular sieve control strategy, historical oxygen production output, historical oxygen production purity, historical oxygen production energy consumption and historical molecular sieve life attenuation influence coefficient; Based on the multiple oxygen production record groups, a molecular sieve control prediction model that meets the prediction accuracy constraint is trained, wherein the molecular sieve control prediction model includes an input layer, a multi-feature prediction parallel layer and an output layer, and the multi-feature prediction parallel layer includes an oxygen production output prediction layer, an oxygen production purity prediction layer, an oxygen production energy consumption prediction layer and a molecular sieve life attenuation impact prediction layer; Based on the air detection data and the first molecular sieve control strategy, the first molecular sieve control prediction result is output according to the molecular sieve control prediction model.

6. The method according to claim 1, characterized in that The multi-objective coupled oxygen production optimization function is: ； Among them, COP represents the coupled oxygen production optimization coefficient, exp represents the exponential function with the natural constant e as the base, OPX represents the predicted oxygen production output, OPXO represents the target oxygen production output, OCI represents the predicted oxygen production purity, OCIO represents the target oxygen production purity, OEK represents the predicted oxygen production energy consumption, OEKO represents the target oxygen production energy consumption, DIC represents the predicted molecular sieve life attenuation influence coefficient, DICO represents the molecular sieve life attenuation influence threshold, Characterize the predetermined weight conditions for optimizing coupled oxygen production.

7. The method according to claim 1, characterized in that According to the multi-objective coupled oxygen production optimization function, the second space of the molecular sieve control strategy is optimized by multi-objective coupling to generate a molecular sieve control optimization result, including According to the second molecular sieve control strategy space, extracting the nth molecular sieve control optimization strategy, where n is a positive integer; Calculating the nth coupled oxygen production optimization coefficient corresponding to the nth molecular sieve control optimization strategy according to the multi-objective coupled oxygen production optimization function; According to the second space of the molecular sieve control strategy, extract the n+1th molecular sieve control optimization strategy, and calculate the n+1th coupled oxygen production optimization coefficient according to the multi-objective coupled oxygen production optimization function; According to the nth coupled oxygen production optimization coefficient and the n+1th coupled oxygen production optimization coefficient, the nth molecular sieve control optimization strategy and the n+1th molecular sieve control optimization strategy are analyzed to determine the current winning coupled oxygen production optimization coefficient and the current winning molecular sieve control optimization strategy corresponding to the current winning coupled oxygen production optimization coefficient; Based on the current superior coupled oxygen production optimization coefficient, the current superior molecular sieve control optimization strategy is iteratively optimized according to the multi-objective coupled oxygen production optimization function and the second space of the molecular sieve control strategy to determine the molecular sieve control optimization result that meets the predetermined number of iterative optimization times.

8. An energy-saving control device for oxygen-producing molecular sieves, characterized in that: The energy-saving control method for oxygen-producing molecular sieve according to any one of claims 1 to 7 is used, and the device comprises: A module for obtaining air to be processed, wherein the module is used to obtain air to be processed of an oxygen-producing molecular sieve assembly according to an oxygen-producing system, wherein the oxygen-producing molecular sieve assembly includes a plurality of oxygen-producing molecular sieves; An analysis module, the analysis module is used to obtain air detection data of the air to be processed, and to control and analyze the oxygen-producing molecular sieve component in combination with a molecular sieve control network, so as to establish a first space of a molecular sieve control strategy that satisfies a first strategy capacity constraint; A target setting module, the target setting module is used to set a multi-feature oxygen production target based on the air to be processed, wherein the multi-feature oxygen production target includes a target oxygen production output, a target oxygen production purity, a target oxygen production energy consumption and a molecular sieve life attenuation impact threshold; A first optimization module, the first optimization module is used to perform multilateral constraint optimization on the first space of the molecular sieve control strategy based on the multi-feature oxygen production target, and establish a second space of the molecular sieve control strategy that meets the second strategy capacity constraint; A function establishment module, wherein the function establishment module is used to establish a multi-objective coupled oxygen production optimization function based on the multi-feature oxygen production target; A second optimization module, the second optimization module is used to perform multi-objective coupling optimization on the second space of the molecular sieve control strategy according to the multi-objective coupling oxygen production optimization function to generate a molecular sieve control optimization result; A control module is used to produce oxygen according to the air to be processed based on the molecular sieve control optimization result and the oxygen-producing molecular sieve component.

Images