Oxygen production, disinfection and purification integrated machine intelligent control method and system combining swarm intelligence

Abstract

This invention discloses an intelligent control method and system for an integrated oxygen generator, disinfection, and purification machine that incorporates swarm intelligence, belonging to the field of swarm intelligence technology. It includes: acquiring current indoor environmental parameters and equipment operating parameters of the integrated oxygen generator, disinfection, and purification machine; calculating the oxygen demand for the current control cycle based on the environmental parameters, and calculating the baseline disinfection dwell time for the current control cycle based on the oxygen flow rate of each oxygen generator unit and the actual disinfection dwell time of the previous control cycle; constructing an initial population based on the oxygen demand and the total number of oxygen generator units; performing multi-objective iterative updates on the initial population based on the oxygen demand, the baseline disinfection dwell time, and the total number of oxygen generator units to obtain a Pareto optimal solution set; selecting an optimal solution from the Pareto optimal solution set as the control command for the current control cycle, and outputting the control command. This invention achieves precise matching of oxygen demand and stable, compliant disinfection effect optimization for the integrated oxygen generator, disinfection, and purification machine.

Description

Technical Field

[0001] This invention relates to the field of swarm intelligence technology, specifically to an intelligent control method and system for an integrated oxygen generator, disinfection, and purification machine that incorporates swarm intelligence. Background Technology

[0002] In existing technologies, the control of integrated oxygen generators, disinfection, and purification systems mostly adopts threshold-triggered adjustment or single-target closed-loop control schemes. The equipment collects indoor environmental parameters through built-in sensors, and uses preset target concentration indicators as the basis for adjustment to perform start-stop control and output flow regulation of the oxygen generator unit to maintain the indoor oxygen supply level; the disinfection function usually relies on the equipment's nominal parameters to set fixed operating rules, and works in conjunction with the airflow in the oxygen generator circuit to complete the disinfection and purification process inside the chamber.

[0003] The output flow rate of the oxygen generating unit simultaneously determines the total oxygen supply of the entire unit and the airflow residence time in the disinfection chamber. These two performance indicators are strongly coupled and constrained. A single-objective adjustment logic cannot simultaneously achieve precise matching of oxygen supply demand and dynamic attainment of disinfection residence time. When the status of indoor personnel and environmental parameters fluctuate, fixed parameter or single-objective adjustment modes are prone to lag in oxygen supply response and unstable disinfection effects. The flow distribution and start-stop combinations of multi-unit oxygen generating equipment lack global-level collaborative optimization, making it difficult to obtain the optimal operating scheme under the dual constraints of oxygen supply and disinfection. Summary of the Invention

[0004] This invention provides an intelligent control method and system for an integrated oxygen generator, disinfection and purification machine that combines swarm intelligence. The purpose is to solve the technical problem that the single-objective control logic in the prior art cannot perform global collaborative optimization of the start-stop status and flow distribution of multiple oxygen generator units.

[0005] In view of the above problems, the present invention provides an intelligent control method and system for an integrated oxygen generator, disinfection and purification machine that combines swarm intelligence.

[0006] In a first aspect, the present invention provides an intelligent control method for an integrated oxygen generator, disinfection, and purification machine that combines swarm intelligence, including:

[0007] The system acquires current indoor environmental parameters and equipment operating parameters of the integrated oxygen generator, disinfection and purification unit. The environmental parameters include indoor oxygen concentration, indoor carbon dioxide concentration and number of people in the room. The equipment operating parameters include the total number of oxygen generators, the oxygen flow rate of each oxygen generator, and the actual disinfection dwell time in the previous control cycle.

[0008] The oxygen demand for the current control cycle is calculated based on the environmental parameters, and the baseline disinfection dwell time for the current control cycle is calculated based on the oxygen production flow rate of each oxygen production unit and the actual disinfection dwell time of the previous control cycle.

[0009] Based on the oxygen demand and the total number of oxygen generating units, an initial population is constructed, wherein each particle in the initial population carries the start / stop status and flow rate setting value of each oxygen generating unit.

[0010] Based on the oxygen demand, the baseline disinfection dwell time, and the total number of oxygen generating units, the initial population is iteratively updated using a multi-objective method to obtain the Pareto optimal solution set.

[0011] Based on the Pareto optimal solution set, an optimal solution is selected as the control command for the current control cycle, and the control command is output.

[0012] Secondly, the present invention provides an intelligent control system for an integrated oxygen generator, disinfection, and purification machine that incorporates swarm intelligence, including:

[0013] The parameter acquisition module is used to acquire the current indoor environmental parameters and the equipment operating parameters of the oxygen generator, disinfection and purification integrated machine. The environmental parameters include indoor oxygen concentration, indoor carbon dioxide concentration and number of people in the room. The equipment operating parameters include the total number of oxygen generators, the oxygen flow rate of each oxygen generator and the actual disinfection dwell time of the previous control cycle.

[0014] The demand calculation module is used to calculate the oxygen supply demand for the current control cycle based on the environmental parameters, and to calculate the baseline disinfection dwell time for the current control cycle based on the oxygen production flow rate of each oxygen production unit and the actual disinfection dwell time of the previous control cycle.

[0015] The population construction module is used to construct an initial population based on the oxygen demand and the total number of oxygen generating units, wherein each particle in the initial population carries the start / stop status and flow rate setting value of each oxygen generating unit.

[0016] The iterative optimization module is used to perform multi-objective iterative updates on the initial population based on the oxygen supply demand, the baseline disinfection dwell time, and the total number of oxygen generating units to obtain the Pareto optimal solution set.

[0017] The instruction output module is used to select an optimal solution as the control instruction for the current control cycle based on the Pareto optimal solution set, and output the control instruction.

[0018] One or more technical solutions provided in this invention have at least the following technical effects or advantages:

[0019] This invention provides an intelligent control method and system for an integrated oxygen generator, disinfection, and purification machine that combines swarm intelligence. By simultaneously acquiring indoor environmental parameters and equipment operating parameters, it calculates the oxygen demand and baseline disinfection dwell time for the current control cycle. Based on a swarm intelligence algorithm, a multi-objective optimization framework is constructed to iteratively optimize the start / stop status and flow setpoint of each oxygen generator unit and output the optimal control command. This invention achieves coordinated optimization of the two coupled indicators of oxygen supply and disinfection, adaptively matching the dynamic changes of indoor personnel and environment. While ensuring oxygen supply accuracy and disinfection effectiveness, it completes the globally optimal configuration of multiple oxygen generator unit operation combinations, effectively improving the control accuracy and operating efficiency of the integrated oxygen generator, disinfection, and purification machine. Attached Figure Description

[0020] Figure 1 A flowchart illustrating the intelligent control method for an integrated oxygen generator, disinfection, and purification machine combining swarm intelligence, provided in an embodiment of the present invention.

[0021] Figure 2 A schematic diagram of the overall logic of the intelligent control method for an integrated oxygen generator, disinfection and purification machine combining swarm intelligence provided in an embodiment of the present invention;

[0022] Figure 3 A schematic diagram of the intelligent control system for an integrated oxygen generator, disinfection, and purification machine that combines swarm intelligence, provided in an embodiment of the present invention.

[0023] The components represented by each number in the attached diagram are explained below:

[0024] Parameter acquisition module 11, requirement calculation module 12, population construction module 13, iterative optimization module 14, instruction output module 15. Detailed Implementation

[0025] This invention provides an intelligent control method and system for an integrated oxygen generator, disinfection and purification machine that combines swarm intelligence. It is used to address the technical problem that the single-objective control logic in the prior art cannot perform global collaborative optimization of the start-stop status and flow distribution of multiple oxygen generator units.

[0026] Example 1, as Figure 1 , Figure 2 As shown, this invention provides an intelligent control method for an integrated oxygen generator, disinfection, and purification machine that combines swarm intelligence. The method includes:

[0027] S100: Obtain the current indoor environmental parameters and the equipment operating parameters of the oxygen generator, disinfection and purification integrated machine. The environmental parameters include indoor oxygen concentration, indoor carbon dioxide concentration and number of people in the room. The equipment operating parameters include the total number of oxygen generator units, the oxygen flow rate of each oxygen generator unit and the actual disinfection dwell time of the previous control cycle.

[0028] In this embodiment of the invention, subsequent calculations of oxygen demand, baseline disinfection dwell time, and swarm intelligence multi-objective optimization all rely on accurate input parameters. Single-point sampling data is susceptible to instantaneous environmental disturbances and sensor noise, and cannot stably reflect the true state of the indoor environment; the lack of historical operating data during the initial power-on of the equipment will prevent the subsequent disinfection dwell time correction logic from starting normally. This step uses a sliding window sampling mechanism and an initial value calibration mechanism to obtain complete and smooth indoor environmental parameters and equipment operating parameters of the integrated oxygen generator, providing a reliable input basis for the full-process control calculation and ensuring that the control logic can be continuously executed in all operating scenarios.

[0029] Step S100 in the method provided in this embodiment of the invention includes:

[0030] Within a preset sliding time window, the indoor oxygen concentration, indoor carbon dioxide concentration, number of people in the room, total number of oxygen generating units, oxygen flow rate of each oxygen generating unit, and actual disinfection dwell time of the previous control cycle are collected according to a preset sampling cycle.

[0031] When the current control cycle is the first control cycle of the equipment's first power-on operation, the actual disinfection dwell time of the previous control cycle is taken as the preset initial calibration value.

[0032] First, within the preset sliding time window, indoor oxygen concentration, indoor carbon dioxide concentration, number of people in the room, total number of oxygen generating units, oxygen flow rate of each oxygen generating unit, and actual disinfection dwell time of the previous control cycle are collected according to the preset sampling cycle.

[0033] The preset sliding time window refers to a continuous time interval that extends backward from the current moment to a fixed duration. This window slides forward continuously over time, always containing the latest historical data. The preset sampling period refers to the time interval between two adjacent data collections.

[0034] Indoor oxygen concentration refers to the volume percentage concentration of oxygen continuously collected by an electrochemical or laser oxygen sensor at a height of 1.2 to 1.5 meters above the ground in the central area of ​​the room. Indoor carbon dioxide concentration refers to the volume concentration of carbon dioxide continuously collected by a non-dispersive infrared absorption carbon dioxide sensor at the same measuring point in the room. The number of people in the room refers to the total number of people detected and counted in real time in the indoor space using an infrared pyroelectric sensor array or a millimeter-wave radar people counting sensor.

[0035] The total number of oxygen generating units refers to the total number of pressure swing adsorption (PSA) oxygen generating modules installed inside the integrated oxygen generating, disinfection, and purification machine. The oxygen flow rate of each oxygen generating unit refers to the volume of oxygen output per unit time by a single unit, read in real-time by a mass flow meter installed in the outlet pipeline of each unit. The actual disinfection residence time of the previous control cycle refers to the measured residence time of the airflow through the disinfection chamber in the previous control cycle; this value is stored in the non-volatile memory inside the device.

[0036] Specifically, within a preset sliding time window, the above six types of parameters are collected synchronously according to a preset sampling period. Each collection of each parameter yields a sample value, forming a data sequence arranged in chronological order within a sliding window. The output values ​​of various sensors or counters are directly recorded as the collection results at each sampling moment. The actual disinfection dwell time value of the previous control cycle is derived from the actual measurement result after the completion of the previous control cycle and is directly read from the storage unit during the parameter acquisition phase of the current control cycle.

[0037] For example, the preset sliding time window length is set to 300 seconds, and the preset sampling period is 10 seconds. A certain oxygen generator, disinfection, and purification integrated machine is a diffusion oxygen generator, installed in a conference room with a floor area of ​​120 square meters and a ceiling height of 2.8 meters. There are currently 12 people in the conference room. The indoor oxygen sensor reading at the current moment is 20.5%, the carbon dioxide sensor reading is 1200 ppm, and the infrared pyroelectric personnel counter displays 12 people. The diffusion oxygen generator contains a total of 6 pressure swing adsorption oxygen generation modules, each with a rated oxygen production flow rate of 5 L / min. That is, the total number of oxygen generation units is 6, and the rated oxygen production flow rate of each unit is 5 L / min, for a total installed oxygen production capacity of 30 L / min. The outlet mass flow meters of each oxygen generating unit read 5L / min, 5L / min, 5L / min, 0L / min, 0L / min, and 0L / min at the current sampling time, respectively, indicating that the first, second, and third oxygen generating units are in operation, with a total oxygen production flow rate of 15L / min, while the fourth, fifth, and sixth oxygen generating units are in standby mode.

[0038] Specifically, when the current control cycle is the first control cycle of the equipment's initial power-on operation, the actual disinfection dwell time of the previous control cycle is taken as a preset initial calibration value. The determination of whether the current control cycle is the first control cycle of the equipment's initial power-on operation is performed by reading the operating status flag bit stored internally in the equipment. The preset initial calibration value refers to the baseline value of the disinfection dwell time measured under standard test conditions before the equipment leaves the factory.

[0039] Specifically, during the performance calibration phase before the equipment leaves the factory, the equipment is placed in a standard environment with a temperature of 25°C and a relative humidity of 50%, and operated at rated power. The rated flow rate of the airflow passes through the disinfection chamber, and the average time of the airflow passing through the disinfection chamber is actually measured using the tracer gas method. This measured value is stored in the non-volatile memory of the equipment as the initial calibration value.

[0040] For example, the control cycle counter register inside the device is read. If the current value of the register is 0, it indicates that the current control cycle is the first control cycle after the device is first powered on. The initial calibration value written at the factory is read from the device's non-volatile memory. This value indicates that the disinfection dwell time measured under standard test conditions is 22.0 seconds. Therefore, the actual disinfection dwell time of the previous control cycle is set to 22.0 seconds.

[0041] In this embodiment of the invention, by establishing a data acquisition framework with a preset sliding time window and a preset sampling period, a data sequence with a clear time reference and a unified temporal correspondence is provided for the subsequent calculation of the carbon dioxide concentration change rate. By introducing the number of indoor personnel into the environmental parameters and the total number of oxygen generating units and the flow rate of each oxygen generating unit into the equipment operating parameters, personnel load information and upper limit constraints on oxygen generating resources are provided for the calculation of oxygen supply demand. By introducing the actual disinfection residence time of the previous control cycle into the equipment operating parameters and providing a preset initial calibration value for the first power-on operation scenario, a data foundation for a closed-loop correction mechanism is established, enabling the subsequent calculation of the benchmark disinfection residence time to correct the theoretical value using historical measured information.

[0042] S200: Calculate the oxygen demand for the current control cycle based on the environmental parameters, and calculate the baseline disinfection dwell time for the current control cycle based on the oxygen production flow rate of each oxygen production unit and the actual disinfection dwell time of the previous control cycle.

[0043] In this embodiment of the invention, the oxygen demand is the target baseline value for calculating oxygen fitness in subsequent particle swarm optimization, and the baseline disinfection residence time is the target baseline value for calculating disinfection fitness. The oxygen demand consists of three parts: the baseline oxygen demand based on the current number of personnel, the deviation compensation based on oxygen concentration deviation, and the trend compensation based on personnel change trends. The calculation of the baseline disinfection residence time requires closed-loop correction of the theoretical calculated value using the actual disinfection residence time of the previous control cycle to eliminate the deviation between the actual operating conditions and the theoretical nominal value caused by factors such as dust accumulation in the filter and uneven airflow distribution in the disinfection chamber.

[0044] Step S200 in the method provided in this embodiment of the invention includes:

[0045] The calculation of the oxygen demand for the current control cycle based on the environmental parameters includes:

[0046] Calculate the oxygen concentration deviation based on the indoor oxygen concentration and the preset target oxygen concentration;

[0047] The rate of change of indoor carbon dioxide concentration is calculated based on the data sequence of indoor carbon dioxide concentration within the preset sliding time window, and the change in the number of people indoors is calculated in combination with the preset ventilation rate.

[0048] The basic oxygen demand is calculated based on the number of people in the room and the preset oxygen demand coefficient per person.

[0049] The deviation compensation amount is calculated based on the oxygen concentration deviation and the preset concentration deviation compensation coefficient;

[0050] The trend compensation amount is calculated based on the change in the number of people in the room and the preset personnel change compensation coefficient.

[0051] The oxygen demand for the current control cycle is obtained by adding the basic oxygen demand, the deviation compensation amount, and the trend compensation amount.

[0052] First, calculate the oxygen concentration deviation based on the indoor oxygen concentration and the preset target oxygen concentration. The preset target oxygen concentration refers to the desired value for maintaining the indoor oxygen concentration. The preset target oxygen concentration is typically set as the volume percentage concentration of oxygen in the atmosphere at sea level, a fixed value between 20.9% and 21.0%. The oxygen concentration deviation is the difference between the preset target oxygen concentration and the current indoor oxygen concentration. For example, if the preset target oxygen concentration is 21.0% and the current indoor oxygen concentration is 20.5%, the oxygen concentration deviation = 21.0% - 20.5% = 0.5%.

[0053] Secondly, the rate of change of carbon dioxide concentration is calculated based on the data sequence of indoor carbon dioxide concentration within a preset sliding time window, and the change in the number of people indoors is calculated in combination with the preset ventilation rate.

[0054] The carbon dioxide concentration change rate refers to the rate at which the indoor carbon dioxide concentration changes over time within a preset sliding time window. Within this window, all sampled values ​​of the indoor carbon dioxide concentration are acquired. Using the sampling time as the independent variable and the carbon dioxide concentration as the dependent variable, a univariate linear fit is performed using the least squares method. The slope of the fitted line is then taken as the carbon dioxide concentration change rate.

[0055] The change in the number of people indoors refers to the quantified trend of the number of people indoors over a recent period, relative to the current sampling time. This change is calculated by inversely applying the quantitative relationship between the rate of change in carbon dioxide concentration and the preset ventilation rate. Specifically, the rate of change in the number of people indoors at the current moment is calculated using the quantitative relationship between the rate of change in indoor carbon dioxide concentration, the preset ventilation rate, and the indoor space volume. The mass conservation law is: Change in indoor carbon dioxide concentration = Amount of carbon dioxide exhaled by people - Amount of carbon dioxide removed by ventilation. Substituting the rate of change in carbon dioxide concentration and the preset ventilation rate into this equation, the change in the number of people indoors can be calculated.

[0056] For example, within a preset sliding time window of 300 seconds, the data sequence of indoor carbon dioxide concentration at all sampling times is acquired. Using the sampling time as the independent variable t (0 ≤ t ≤ 300) and the carbon dioxide concentration as the dependent variable C(t), a univariate linear fit is performed using the least squares method. The slope of the fitted line is 0.8 ppm / s, meaning the rate of change of carbon dioxide concentration is 0.8 ppm / s. Indoor space volume = 120 × 2.8 = 336 m³ 3 The preset air exchange rate is 0.2 times / hour, which translates to a volumetric flow rate of 336 × 0.2 / 3600 = 0.0187 m³. 3 / s. Given a carbon dioxide concentration change rate of 0.8 ppm / s and a room volume of 336 m². 3 and a preset air exchange rate of 0.0187m 3 In the case of / s, according to the law of conservation of mass, the change in the number of people indoors is calculated to be an increase of 0.5 people / minute.

[0057] Next, the basal oxygen demand (BOD) is calculated based on the number of people in the room and the preset individual oxygen demand coefficient. The preset individual oxygen demand coefficient refers to the rate of oxygen supply required for a single adult to maintain normal physiological activity under standard activity conditions. This coefficient is obtained by statistically averaging oxygen consumption data from a large number of people in a sedentary state. The basal oxygen demand (BOD) refers to the rate of oxygen supply required for normal breathing for all people in the room. The BOD equals the number of people in the room multiplied by the preset individual oxygen demand coefficient. For example, if there are 12 people in the room and the preset individual oxygen demand coefficient is 0.5 L / (min·person), the BOD = 12 × 0.5 = 6.0 L / min.

[0058] Then, the deviation compensation amount is calculated based on the oxygen concentration deviation and the preset concentration deviation compensation coefficient. The preset concentration deviation compensation coefficient is the gain coefficient that converts a unit oxygen concentration deviation into oxygen supply flow rate. The preset concentration deviation compensation coefficient is determined by dividing the total volume of the indoor space by the preset response time constant. The deviation compensation amount refers to the additional increase or decrease in oxygen supply flow rate required to bring the current indoor oxygen concentration back to the preset target oxygen concentration. The deviation compensation amount is obtained by multiplying the oxygen concentration deviation by the preset concentration deviation compensation coefficient. For example, if the preset concentration deviation compensation coefficient is 20L / (min·%), the deviation compensation amount = 0.5% × 20L / (min·%) = 10.0L / min.

[0059] Subsequently, the trend compensation amount is calculated based on the change in the number of people indoors and the preset personnel change compensation coefficient. The preset personnel change compensation coefficient is a gain coefficient that converts the change in the number of people per unit into oxygen supply flow rate. The preset personnel change compensation coefficient is numerically equal to the preset single-person oxygen supply demand coefficient, but their meanings are different: the preset single-person oxygen supply demand coefficient is used to convert the current number of people into the static oxygen demand at the current moment, while the preset personnel change compensation coefficient is used to convert the change in the number of people over a future period into a dynamic adjustment of oxygen demand. The trend compensation amount refers to the increased oxygen supply flow rate required to proactively address the trend of changes in the number of people indoors. The trend compensation amount is obtained by multiplying the change in the number of people indoors by the preset personnel change compensation coefficient. For example, trend compensation amount = 0.5 people / minute × 0.5L / (min·person) = 0.25L / min.

[0060] Finally, the basic oxygen demand, deviation compensation, and trend compensation are added together to obtain the oxygen demand for the current control cycle. For example, the oxygen demand for the current control cycle = 6.0 L / min + 10.0 L / min + 0.25 L / min = 16.25 L / min.

[0061] The baseline disinfection dwell time for the current control cycle is calculated based on the oxygen flow rate of each oxygen generation unit and the actual disinfection dwell time of the previous control cycle, including:

[0062] Obtain the nominal volume of the disinfection chamber of the integrated oxygen generator, disinfection and purification machine;

[0063] Calculate the theoretical value of the total oxygen production flow rate based on the oxygen production flow rate of each oxygen production unit in operation;

[0064] Divide the nominal volume of the disinfection chamber by the theoretical value of the total oxygen flow rate to obtain the theoretical value of the current disinfection residence time. If the theoretical value of the total oxygen flow rate is zero, then set the theoretical value of the current disinfection residence time to the preset maximum allowable residence time.

[0065] When the current control cycle is not the first control cycle, the residence time deviation coefficient is calculated based on the actual disinfection residence time of the previous control cycle and the theoretical value of the disinfection residence time of the previous control cycle.

[0066] The theoretical value of the current disinfection dwell time is corrected by the dwell time deviation coefficient to obtain the benchmark disinfection dwell time for the current control cycle.

[0067] First, obtain the nominal volume of the disinfection chamber of the integrated oxygen generator, disinfection, and purification unit. The nominal volume of the disinfection chamber refers to the geometric volume of the internal disinfection chamber of the integrated oxygen generator, disinfection, and purification unit. Read the nominal volume of the disinfection chamber from the non-volatile memory of the integrated oxygen generator, disinfection, and purification unit. For example, the nominal volume of the disinfection chamber is 0.05 m³. 3 .

[0068] Secondly, the theoretical total oxygen production flow rate is calculated based on the oxygen production flow rate of each oxygen generating unit currently in operation. The theoretical total oxygen production flow rate refers to the sum of the output flow rates of all oxygen generating units currently in operation. To obtain the theoretical total oxygen production flow rate, the oxygen production flow rates of each operating oxygen generating unit are summed. For example, if the currently operating oxygen generating units are the first, second, and third oxygen generating units, and each unit has an oxygen production flow rate of 5 L / min, the theoretical total oxygen production flow rate is 15 L / min.

[0069] Next, the nominal volume of the disinfection chamber is divided by the theoretical value of the total oxygen production flow rate to obtain the theoretical value of the current disinfection residence time. If the theoretical value of the total oxygen production flow rate is zero, the theoretical value of the current disinfection residence time is set as the preset maximum allowable residence time. The theoretical value of the disinfection residence time refers to the time required for the airflow to pass through the disinfection chamber assuming that the airflow is completely uniformly distributed within the disinfection chamber without any leakage or short circuit. It is equal to the nominal volume of the disinfection chamber divided by the theoretical value of the total oxygen production flow rate. The preset maximum allowable residence time is the upper limit substitute value used in subsequent constraint judgments when the total oxygen production flow rate is zero. This value is set to a very large value to ensure that the particle is not marked as an infeasible solution.

[0070] Specifically, if the theoretical total oxygen flow rate is zero, the current theoretical disinfection residence time is set to the preset maximum allowable residence time. If the theoretical total oxygen flow rate is greater than zero, the nominal volume of the disinfection chamber is divided by the theoretical total oxygen flow rate to obtain the theoretical current disinfection residence time. Units must be consistent during calculation: the nominal volume of the disinfection chamber should be converted from m³ to m³. 3 Converting to L, the theoretical unit of the total oxygen production flow rate is L / min. Dividing the two and multiplying by 60 yields the theoretical disinfection residence time in seconds. For example, if the theoretical total oxygen production flow rate is greater than zero, and the nominal volume of the disinfection chamber is 0.05 m³,... 3Converting to 50L, divided by the theoretical total oxygen production flow rate = 50L / 15L / min × 60 = 200 seconds, that is, the theoretical disinfection dwell time is 200 seconds.

[0071] Furthermore, when the current control cycle is not the first control cycle, a residence time deviation coefficient is calculated based on the actual disinfection residence time and the theoretical value of the disinfection residence time in the previous control cycle. The residence time deviation coefficient refers to the ratio between the actual disinfection residence time and the theoretical disinfection residence time in the previous control cycle, reflecting the degree of deviation between the actual operating conditions and the theoretical nominal operating conditions of the disinfection chamber.

[0072] The control cycle counter register is read. If the current value of the register is greater than 0, the dwell time deviation coefficient is calculated. The actual disinfection dwell time of the previous control cycle is divided by the theoretical value of the disinfection dwell time of the previous control cycle to obtain the dwell time deviation coefficient. If the current is the first control cycle, the dwell time deviation coefficient is not calculated, and the baseline disinfection dwell time is directly the theoretical value of the disinfection dwell time.

[0073] For example, the control cycle count register is read; the current value is 100, indicating that this is not the first control cycle. The actual disinfection dwell time in the previous control cycle was 16.5 seconds. The theoretical value of the disinfection dwell time in the previous control cycle is calculated: In the previous control cycle, the first and second oxygen generating units were operating, with a flow rate of 5L / min for each unit, resulting in a theoretical total oxygen generating flow rate of 10L / min. The theoretical value of the disinfection dwell time in the previous control cycle = 50L / 10L / min × 60 = 300.0 seconds. The dwell time deviation coefficient = 16.5 / 300 = 0.055.

[0074] Finally, the theoretical value of the current disinfection residence time is corrected using the residence time deviation coefficient to obtain the baseline disinfection residence time for the current control cycle. The baseline disinfection residence time refers to the target residence time value obtained after correction of the actual measurement value from the previous control cycle, serving as the basis for calculating the disinfection fitness of this control cycle. The baseline disinfection residence time for the current control cycle is obtained by multiplying the theoretical value by the residence time deviation coefficient. For example, the baseline disinfection residence time for the current control cycle = 200.0 seconds × 0.055 = 11.0 seconds.

[0075] In this embodiment of the invention, by decomposing the oxygen demand into three components—basic oxygen demand, deviation compensation, and trend compensation—the calculation of the oxygen demand simultaneously considers three time scales: the current static load, the current concentration deviation, and the future trend, thereby improving the response speed and steady-state accuracy of oxygen supply control. By using the actual disinfection residence time from the previous control cycle to correct the theoretical value of the current disinfection residence time, the baseline disinfection residence time can reflect the actual operating conditions of the disinfection chamber, eliminating theoretical deviations caused by factors such as filter dust accumulation and uneven airflow distribution. This provides a more accurate baseline value for calculating disinfection fitness in subsequent multi-objective optimization.

[0076] S300: Based on the oxygen demand and the total number of oxygen generating units, an initial population is constructed, wherein each particle in the initial population carries the start / stop status and flow rate setting value of each oxygen generating unit.

[0077] After obtaining the oxygen demand and the total number of oxygen-generating units, an initial population needs to be constructed as the starting point for swarm intelligence iterative optimization. Each particle in the population represents a possible control scheme, namely, a combination of the start / stop state of each oxygen-generating unit and the flow rate setpoint. The requirements for population construction include three levels: First, the particle encoding structure must completely cover all adjustable oxygen-generating units, ensuring that the start / stop state and flow rate setpoint of each unit can be expressed in the encoding; Second, the initial range of the flow rate setpoint needs to be reasonably determined based on the oxygen demand to ensure that the deviation between the oxygen supply capacity and the demand of most particles in the initial population is not too large, thereby improving the convergence efficiency of subsequent iterations; Third, the population size needs to be pre-set. Too small a size will lead to insufficient search space coverage, while too large a size will increase the computational burden.

[0078] Step S300 in the method provided in this embodiment of the invention includes:

[0079] The number of coding groups for population particles is determined based on the total number of oxygen generating units. Each coding group corresponds to one oxygen generating unit, and each coding group includes the start / stop status and flow rate setting value of the oxygen generating unit.

[0080] Based on the ratio of the oxygen demand to the total number of oxygen generating units, the initial range of the flow rate setting value for each oxygen generating unit is determined. The lower limit of the initial range of the flow rate setting value is 0, and the upper limit is the ratio of the oxygen demand to the total number of oxygen generating units.

[0081] Within the initial range of the flow rate setting value, a flow rate setting value is randomly generated for each group of codes of each particle to generate an initial population of a preset size.

[0082] First, the number of coding groups for the population particles is determined based on the total number of oxygen-generating units. Each coding group corresponds to one oxygen-generating unit, and each coding group includes the start / stop status and flow rate setpoint of that oxygen-generating unit. The number of coding groups refers to the total number of coding groups contained in each particle, and its value is equal to the total number of oxygen-generating units. Each coding group refers to a set of data in the particle coding corresponding to a single oxygen-generating unit, containing two fields: the first field is the start / stop status of the oxygen-generating unit, and the second field is the flow rate setpoint of the oxygen-generating unit. The start / stop status refers to the operating status flag of the oxygen-generating unit, with a value of "on" or "off". The flow rate setpoint refers to the target value of oxygen output per unit time set for the oxygen-generating unit. When the start / stop status is off, the flow rate setpoint is fixed at 0 L / min; when the start / stop status is on, the flow rate setpoint is greater than 0 and does not exceed a certain value of the unit's rated maximum oxygen production flow rate.

[0083] Specifically, the total number of oxygen generating units is read, and this value is directly used to determine the number of code groups for each particle. Each oxygen generating unit is assigned a unique code, with the code order corresponding one-to-one with the physical number of the oxygen generating unit. In each code group, the start / stop status is represented by a binary flag, and the flow rate setting is represented by a floating-point number or a fixed-point number.

[0084] For example, the diffused oxygen generator contains a total of 6 pressure swing adsorption oxygen generation modules, that is, a total of 6 oxygen generation units. The number of code groups for each particle is determined to be 6 groups. The first group of codes corresponds to the first oxygen generation unit, including the start / stop status flag bit and flow rate setting value of the unit; the second group of codes corresponds to the second oxygen generation unit, and so on until the sixth group of codes corresponds to the sixth oxygen generation unit.

[0085] Secondly, based on the ratio of oxygen demand to the total number of oxygen generating units, the initial range of flow rate settings for each oxygen generating unit is determined. The lower limit of the initial flow rate setting range is 0, and the upper limit is the ratio of oxygen demand to the total number of oxygen generating units. The initial range of flow rate settings refers to the random value interval set for the flow rate setting value in each group of codes for each particle when generating the initial population. The lower and upper limits of the initial range of flow rate settings constitute the value boundaries of the flow rate settings.

[0086] Specifically, the oxygen demand is divided by the total number of oxygen-generating units to obtain the average oxygen contribution per unit. The lower limit of the initial flow rate setpoint is set to 0 L / min, and the upper limit is set to this average oxygen contribution. During the initial population generation process, the flow rate setpoint in each group of codes for each particle is randomly selected between this lower and upper limit.

[0087] For example, the oxygen demand in the current control cycle is 16.25 L / min, and there are a total of 6 oxygen generating units, so 16.25 L / min / 6 = 2.71 L / min. Therefore, the initial range of the flow rate setpoint for each oxygen generating unit is determined to be [0 L / min, 2.71 L / min], that is, the flow rate setpoint for each oxygen generating unit in the initial population is randomly generated between 0 L / min and 2.71 L / min.

[0088] Next, within the initial range of the flow rate setpoint, a flow rate setpoint is randomly generated for each set of codes for each particle, generating an initial population of a preset size. The preset size refers to the total number of particles contained in the initial population, which is preset in the device control parameters. The value of the preset size needs to balance the coverage of the search space and the consumption of computing resources. The initial population refers to the first generation of particle sets generated before the start of iterative optimization. Each particle in the population carries a complete set of codes, representing a complete control scheme.

[0089] Specifically, read the predetermined numerical values. For each particle in the population, for each set of codes containing the flow rate setting value, use a uniformly distributed random number generator to generate a random number within the initial range of the flow rate setting value, which serves as the flow rate setting value for that set of codes. Determine the corresponding start / stop state based on the flow rate setting value for each set of codes: if the flow rate setting value is greater than 0, the start / stop state of the oxygen generation unit corresponding to that set of codes is set to on; if the flow rate setting value is equal to 0, the start / stop state of the oxygen generation unit corresponding to that set of codes is set to off. Repeat the above steps until the number of generated particles reaches the predetermined scale.

[0090] For example, the preset scale is set to 50 particles. Taking particle A as an example, flow rate settings are generated for each of the six sets of codes for this particle. The flow rate setting for each set of codes is independently and randomly selected within the range of [0L / min, 2.71L / min]. The six generated random numbers are 2.1L / min, 0L / min, 1.5L / min, 0.8L / min, 0L / min, and 0L / min, respectively. The start / stop status is determined according to the flow rate setting of each set: the first set of flow rate is 2.1L / min, set to on; the second set of flow rate is 0L / min, set to off; the third set of flow rate is 1.5L / min, set to on; the fourth set of flow rate is 0.8L / min, set to on; the fifth set of flow rate is 0L / min, set to off; and the sixth set of flow rate is 0L / min, set to off. The complete code for this particle is [On 2.1, Off 0, On 1.5, On 0.8, Off 0, Off 0], representing a control scheme: the first oxygen generator operates at 2.1 L / min, the second oxygen generator is off, the third oxygen generator operates at 1.5 L / min, the fourth oxygen generator operates at 0.8 L / min, and the fifth and sixth oxygen generators are off, with a total oxygen flow rate of 4.4 L / min. This random generation process is repeated until 50 particles are generated, forming the initial population.

[0091] In this embodiment of the invention, by encoding each particle into a code group equal to the total number of oxygen-generating units, each control scheme fully covers the start / stop states and flow rate setpoints of all oxygen-generating units, providing a structured decision variable space for subsequent multi-objective iterative optimization. By setting the upper limit of the initial range of flow rate setpoints to the ratio of oxygen demand to the total number of oxygen-generating units, the flow rate setpoint of a single oxygen-generating unit in the initial population will not exceed the average oxygen supply contribution, avoiding an excessively high proportion of extreme solutions in the initial population, thus improving the quality of the initial solution and the convergence speed of subsequent iterations. By randomly generating the flow rate setpoints of each particle within the initial range, the diversity of the initial population in the solution space is ensured.

[0092] S400: Based on the oxygen demand, the baseline disinfection dwell time, and the total number of oxygen generating units, perform multi-objective iterative updates on the initial population to obtain the Pareto optimal solution set.

[0093] After constructing the initial population, a Pareto optimal solution set needs to be searched through multi-objective iterative optimization. The task of iterative optimization is to balance the two objectives of oxygen supply fitness and disinfection fitness, while satisfying constraints such as purification pressure drop and disinfection residence time, gradually guiding the population towards the Pareto front. In each iteration, the total oxygen production flow rate and corresponding oxygen supply fitness and disinfection fitness of each particle need to be calculated. All particles are divided into different levels through non-dominated sorting, and the globally optimal particle is selected from the highest level to guide the population update. This process is repeated until the convergence condition is met. When updating the position and velocity of particles, the overall system noise value needs to be introduced as an adjustment factor to avoid the optimization results from concentrating on high-noise solutions.

[0094] Step S400 in the method provided in this embodiment of the invention includes:

[0095] Obtain the start / stop status and flow rate setpoint of each oxygen-generating unit carried by each particle in the current population;

[0096] The total oxygen production flow rate of each oxygen generating unit in operation is obtained by adding the flow rate setpoints of each unit.

[0097] The oxygen supply adaptability of each particle is calculated based on the absolute value of the difference between the total oxygen production flow rate of each particle and the oxygen supply demand, wherein the oxygen supply adaptability is negatively correlated with the absolute value of the oxygen supply difference.

[0098] Calculate the individual disinfection residence time for each particle based on the total oxygen production flow rate of each particle and the nominal volume of the disinfection chamber.

[0099] The disinfection fitness of each particle is calculated based on the absolute value of the disinfection difference between the individual disinfection dwell time and the preset disinfection dwell time constraint threshold, wherein the disinfection fitness is negatively correlated with the absolute value of the disinfection difference.

[0100] Based on the oxygen supply adaptability and disinfection adaptability of each particle, and combined with preset constraints, all particles are sorted in a non-dominated order, and all particles are divided into a first level, a second level and a third level. The first level consists of feasible solution particles that are not dominated by any other particle, the second level consists of feasible solution particles that are dominated by at least one other particle, and the third level consists of infeasible solution particles.

[0101] Select a particle from the first level as the global optimal particle;

[0102] If the first level is empty, then the particle with the smallest purification pressure drop is selected from the second level as the global optimal particle.

[0103] If both the first and second levels are empty, then the particle with the smallest purification pressure drop is selected from the third level as the global optimal particle.

[0104] Based on the start / stop status and flow rate setting of each oxygen-generating unit carried by the global optimal particle, update the position and velocity of all particles in the current population to generate the next generation population.

[0105] Repeat the current step until the preset convergence condition is met, and use the particles in the current population located at the first level as the Pareto optimal solution set.

[0106] First, obtain the start / stop status and flow rate settings of each oxygen-generating unit carried by each particle in the current population. The current population refers to the set of particles in the current iteration round. When S400 is executed for the first time, the current population is the initial population constructed in S300; in subsequent iterations, the current population is the next generation population generated in the previous iteration. For example, the current population is the initial population containing 50 particles. Taking particle A as an example, the code carried by this particle is [On 2.1, Off 0, On 1.5, On 0.8, Off 0, Off 0]. The oxygen-generating units in the running state are the first unit, the third unit, and the fourth unit, with flow rate settings of 2.1L / min, 1.5L / min, and 0.8L / min, respectively.

[0107] Secondly, the flow rate setpoints of each oxygen-generating unit in operation are summed to obtain the total oxygen-generating flow rate for each particle. The total oxygen-generating flow rate refers to the sum of the flow rate setpoints of all oxygen-generating units in operation within the control scheme carried by a single particle. For example, the total oxygen-generating flow rate of particle A = 2.1 + 1.5 + 0.8 = 4.4 L / min.

[0108] Next, based on the absolute value of the difference between the total oxygen production flow rate and the oxygen demand for each particle, the oxygen supply fitness of each particle is calculated. The oxygen supply fitness is negatively correlated with the absolute value of the oxygen supply difference. Oxygen supply fitness is a numerical value used to evaluate the particle's performance in meeting its oxygen demand; a higher oxygen supply fitness indicates that the particle's total oxygen production flow rate is closer to its oxygen demand. The oxygen supply fitness is negatively correlated with the absolute value of the oxygen supply difference between the total oxygen production flow rate and the oxygen demand; that is, the larger the absolute value of the oxygen supply difference, the lower the oxygen supply fitness.

[0109] Specifically, the oxygen supply adaptability is calculated as 1 / (|total oxygen production flow rate − oxygen demand| + 1). Here, |total oxygen production flow rate − oxygen demand| is the absolute value of the difference between the total oxygen production flow rate and the oxygen demand; adding 1 to the denominator prevents overflow due to a zero denominator. The oxygen supply adaptability reaches its maximum value of 1 when the total oxygen production flow rate equals the oxygen demand; when the total oxygen production flow rate deviates from the oxygen demand, the larger the absolute value of the difference, the closer the oxygen supply adaptability is to 0. For example, if the oxygen demand is 16.25 L / min, the absolute value of the oxygen supply difference = |4.4 − 16.25| = 11.85 L / min, and the oxygen supply adaptability = 1 / (11.85 + 1) ≈ 0.078.

[0110] Subsequently, based on the total oxygen production flow rate of each particle and the nominal volume of the disinfection chamber, the individual disinfection residence time for each particle is calculated. The individual disinfection residence time refers to the disinfection residence time corresponding to the control scheme carried by a single particle, and is equal to the nominal volume of the disinfection chamber divided by the total oxygen production flow rate of that particle. For example, the nominal volume of the disinfection chamber is 0.05 m³. 3 Converted to 50L. The total oxygen production flow rate is 4.4L / min, which is greater than zero. Individual disinfection residence time = 50L / 4.4L / min × 60 = 681.8 seconds.

[0111] Then, based on the absolute value of the disinfection difference between the individual disinfection residence time of each particle and the preset disinfection residence time constraint threshold, the disinfection fitness of each particle is calculated, wherein the disinfection fitness is negatively correlated with the absolute value of the disinfection difference. Disinfection fitness is a numerical value used to evaluate the performance of a particle in meeting the disinfection residence time constraint; a higher disinfection fitness indicates that the individual disinfection residence time of the particle is closer to the preset disinfection residence time constraint threshold. Disinfection fitness is negatively correlated with the absolute value of the disinfection difference between the individual disinfection residence time and the preset disinfection residence time constraint threshold.

[0112] The preset disinfection dwell time constraint threshold is a fixed value pre-set based on the physical structural parameters of the disinfection chamber and the minimum effective irradiation time or minimum effective contact time required by the disinfection method. It is used to determine whether any candidate solution meets the hard constraint of disinfection effectiveness. Specifically, when using ultraviolet disinfection, the preset disinfection dwell time constraint threshold is determined by the power, wavelength, and effective irradiation distance of the ultraviolet lamp; when using photocatalytic disinfection, the preset disinfection dwell time constraint threshold is determined by the active area of ​​the photocatalytic material and the intensity of the light source. Before the equipment leaves the factory, the above theoretical values ​​are verified and corrected through microbial killing experiments, and finally calibrated into a fixed threshold that does not change with changes in indoor environmental parameters or equipment operating conditions.

[0113] Specifically, the disinfection fitness is calculated as 1 / (|Individual disinfection dwell time − Preset disinfection dwell time constraint threshold| + 1). The disinfection fitness reaches its maximum value of 1 when the individual disinfection dwell time equals the preset disinfection dwell time constraint threshold. When the individual disinfection dwell time deviates from the preset disinfection dwell time constraint threshold, the larger the absolute value of the difference, the closer the disinfection fitness approaches 0. For example, if the preset disinfection dwell time constraint threshold is 300 seconds, the absolute value of the disinfection difference = |681.8 − 300| = 381.8 seconds, and the disinfection fitness = 1 / (381.8 + 1) ≈ 0.0026.

[0114] Furthermore, based on the oxygen supply adaptability and disinfection adaptability of each particle, and combined with preset constraints, all particles are sorted in a non-dominated order, and all particles are divided into a first level, a second level, and a third level. The first level consists of feasible solution particles that are not dominated by any other particle, the second level consists of feasible solution particles that are dominated by at least one other particle, and the third level consists of infeasible solution particles.

[0115] Non-dominated sorting refers to the process of dividing the population into hierarchical groups from high to low based on the dominance relationships between particle fitness vectors. A feasible solution is a particle that satisfies all preset constraints, including an individual disinfection residence time greater than or equal to the dynamic disinfection residence time constraint threshold, and a purification pressure drop less than or equal to the preset upper limit of the purification pressure drop threshold. An infeasible solution is a particle that does not satisfy at least one preset constraint, including an individual disinfection residence time less than the dynamic disinfection residence time constraint threshold, or a purification pressure drop greater than the preset upper limit of the purification pressure drop threshold.

[0116] Based on the oxygen supply and disinfection adaptability of each particle, and combined with preset constraints, all particles are non-dominated and sorted, dividing them into three levels: Level 1, Level 2, and Level 3, including:

[0117] Based on the baseline disinfection dwell time and the preset baseline disinfection threshold, a dynamic disinfection dwell time constraint threshold is generated;

[0118] Based on the start / stop status and flow rate settings of each oxygen-generating unit carried by each particle in the current population, and based on the series and parallel relationship of the flow resistance of each oxygen-generating unit's gas channel, the purification pressure drop corresponding to the particle is calculated through a preset pressure drop model.

[0119] Particles whose individual disinfection dwell time is greater than or equal to the dynamic disinfection dwell time constraint threshold, or particles whose purification pressure drop is greater than the preset purification pressure drop upper limit threshold, are marked as infeasible solutions and classified into the third level.

[0120] Particles not marked as infeasible solutions are identified as feasible solutions. For feasible solution particles whose purification pressure drop is greater than the preset purification pressure drop inflection point threshold and less than or equal to the preset purification pressure drop upper limit threshold, their oxygen supply adaptability is multiplied by a preset penalty coefficient.

[0121] Compare the dominance relationships of all feasible solution particles: For any two feasible solution particles, if the oxygen supply fitness of the first particle is greater than or equal to the oxygen supply fitness of the second particle, and the disinfection fitness of the first particle is greater than or equal to the disinfection fitness of the second particle, and at least one fitness is greater than the corresponding fitness of the second particle, then it is determined that the first particle dominates the second particle.

[0122] Feasible solution particles that are not dominated by any other feasible solution particles are classified into the first level, and feasible solution particles that are dominated by at least one other feasible solution particle are classified into the second level.

[0123] First, a dynamic disinfection time constraint threshold is generated based on the baseline disinfection time and the preset baseline disinfection threshold. The dynamic disinfection time constraint threshold is a dynamic threshold determined by both the baseline disinfection time and the preset baseline disinfection threshold to determine whether an individual's disinfection time meets the standard. The preset baseline disinfection threshold is obtained, and the smaller of the preset baseline disinfection threshold and the baseline disinfection time is used as the dynamic disinfection time constraint threshold, i.e., Dynamic disinfection time constraint threshold = min(Preset baseline disinfection threshold, Baseline disinfection time). For example, if the baseline disinfection time is 11.0 seconds and the preset baseline disinfection threshold is 300 seconds, then the dynamic disinfection time constraint threshold = min(300, 11.0) = 11.0 seconds.

[0124] The preset baseline disinfection threshold refers to the lower limit set for the shortest residence time of airflow through the disinfection chamber to ensure disinfection effect. The preset baseline disinfection threshold is determined by the physical structural parameters of the disinfection chamber and the minimum effective irradiation time or minimum effective contact time required by the disinfection method used, and is calibrated after verification by microbial killing experiments.

[0125] Secondly, based on the start / stop status and flow rate settings of each oxygen-generating unit carried by each particle in the current population, and according to the series and parallel relationship of the flow resistance of each oxygen-generating unit's gas channels, the purification pressure drop corresponding to each particle is calculated through a preset pressure drop model. Purification pressure drop refers to the static pressure difference generated between the inlet and outlet sides when the airflow passes through the internal purification components of the integrated oxygen generator, disinfection, and purification machine.

[0126] Specifically, based on the start / stop status and flow rate setpoint of each oxygen-generating unit carried by each particle in the current population, the number of oxygen-generating units in operation and the flow rate setpoint of each operating unit are determined. Based on the series and parallel flow resistance relationship of the gas channels in each oxygen-generating unit, the purification pressure drop corresponding to each particle is calculated using a preset pressure drop model. The preset pressure drop model takes the start / stop status and flow rate setpoint of each oxygen-generating unit as input and the purification pressure drop as output.

[0127] The preset pressure drop model is a mathematical model based on the principle of series and parallel flow resistance in fluid mechanics. It is used to calculate the internal purification pressure drop of the equipment based on the start-up and shutdown status and flow setpoint of each oxygen production unit. When constructing the preset pressure drop model, the gas channels inside the equipment are first divided into independent flow resistance elements according to components such as pipes, valves, filters, and adsorption towers. The series and parallel flow resistance network topology is determined based on the actual connection relationship of each element in the equipment. Then, a quantitative relationship equation between pressure drop and flow rate is established for each flow resistance element. The flow resistance coefficient in the equation is obtained through theoretical calculation of the element's geometric parameters or through actual measurement and calibration before delivery. Finally, the flow setpoint of each oxygen production unit carried by particles is substituted into the flow resistance network equation set for solution to obtain the corresponding purification pressure drop value. Taking the structure of two oxygen generating units merging and passing through a shared filter as an example, the outlet pipes of each unit are connected in parallel, and after merging, they enter the shared filter in series. The total pressure drop is equal to the sum of the pressure drop of the parallel pipes and the pressure drop of the filter. The pressure drop of the parallel pipes is determined by the flow rate of each branch and its respective flow resistance coefficient, while the pressure drop of the filter is determined by the total flow rate and the flow resistance coefficient of the filter. The total purification pressure drop can be calculated by substituting the set values ​​of the flow rates of each unit carried by the particles.

[0128] For example, taking particle A as an example, this particle is coded as [On 2.1, Off 0, On 1.5, On 0.8, Off 0, Off 0], and its operating units are the first unit (2.1L / min), the third unit (1.5L / min), and the fourth unit (0.8L / min), with a total flow rate of 4.4L / min. Based on the series and parallel relationship of the flow resistance of the gas channels in each oxygen generation unit, the purification pressure drop is calculated to be 45Pa using a preset pressure drop model. The remaining particles are processed in the same way, with the corresponding purification pressure drop values ​​calculated using a preset pressure drop model based on the start / stop status and flow rate setpoint of each oxygen generation unit in their respective codes.

[0129] Next, particles with an individual disinfection residence time greater than or equal to the dynamic disinfection residence time constraint threshold, or particles with a purification pressure drop greater than the preset upper limit threshold for purification pressure drop, are marked as infeasible solutions and classified into the third level. The preset upper limit threshold for purification pressure drop refers to the maximum allowable purification pressure drop. This threshold is jointly determined by the mechanical strength of the internal purification components of the integrated oxygen generator, disinfection, and purification unit, the rated head of the fan, and the system's sealing performance. When the purification pressure drop exceeds this value, it indicates that the purification components are severely blocked or the airflow resistance has exceeded the safe operating range of the equipment. For example, a particle has an individual disinfection residence time of 184.6 seconds, which is greater than the dynamic disinfection residence time constraint threshold of 11.0 seconds, satisfying the disinfection constraint; however, its purification pressure drop is 350 Pa, which is greater than the preset upper limit threshold for purification pressure drop of 300 Pa, failing to meet the pressure drop constraint. This particle is marked as an infeasible solution and classified into the third level.

[0130] Then, particles not marked as infeasible solutions are identified as feasible solutions. For feasible solution particles whose purification pressure drop is greater than the preset purification pressure drop inflection point threshold and less than or equal to the preset purification pressure drop upper limit threshold, their oxygen supply fitness is multiplied by a preset penalty coefficient.

[0131] The preset purification pressure drop inflection point threshold refers to the starting point at which a penalty needs to be imposed on the particle oxygen supply adaptability after the purification pressure drop exceeds this value. The preset purification pressure drop inflection point threshold is less than the preset purification pressure drop upper limit threshold. The preset purification pressure drop inflection point threshold is calibrated by the inflection point of the performance degradation curve of the purification component during its normal life cycle, and is used to apply selective pressure in advance when the pressure drop increases significantly but has not yet reached the upper limit.

[0132] The preset penalty coefficient refers to the discount factor applied to oxygen supply adaptability when the purification pressure drop exceeds the preset inflection point threshold but does not exceed the preset upper limit threshold. The preset penalty coefficient is a constant greater than 0 and less than 1. The preset penalty coefficient is calibrated using multiple sets of purification efficiency test data of the equipment under different pressure drop ranges, and is used to quantify the negative impact weight of increased pressure drop on the overall system performance.

[0133] For example, if the purification pressure drop of a particle is 150 Pa, which is greater than the preset purification pressure drop inflection point threshold of 100 Pa and less than the preset purification pressure drop upper limit threshold of 300 Pa, the oxygen supply adaptability of the particle is multiplied by a preset penalty coefficient of 0.8. The original oxygen supply adaptability of the particle is 0.364, and the oxygen supply adaptability after penalty = 0.364 × 0.8 ≈ 0.291.

[0134] Then, the dominance relationship of all feasible solution particles is compared: for any two feasible solution particles, if the oxygen supply fitness of the first particle is greater than or equal to the oxygen supply fitness of the second particle, and the disinfection fitness of the first particle is greater than or equal to the disinfection fitness of the second particle, and at least one fitness is greater than the corresponding fitness of the second particle, then it is determined that the first particle dominates the second particle.

[0135] For example, the currently feasible solution particles are particles A, B, and C, with fitness vectors of (oxygen supply fitness 0.078, disinfection fitness 0.0026), (oxygen supply fitness 0.291, disinfection fitness 0.0086), and (oxygen supply fitness 0.952, disinfection fitness 0.00855), respectively. Pairwise comparisons show that particle A's oxygen supply fitness is lower than both particles B and C, and its disinfection fitness is also lower than both. Therefore, particle A is dominated by particles B and C. Compared to particle C, particle B's oxygen supply fitness (0.291) is lower than particle C's (0.952), but its disinfection fitness (0.0086) is higher than particle C's (0.00855). Both particles have their advantages and disadvantages and do not dominate each other.

[0136] Finally, feasible solution particles not dominated by any other feasible solution particles are classified into the first level, and feasible solution particles dominated by at least one other feasible solution particle are classified into the second level. For example, particles B and C are not dominated by any other feasible solution particles and are classified into the first level; particle A is dominated and is classified into the second level. The remaining particles are classified according to the same rules: those not dominated by any other feasible solution particles are classified into the first level, those dominated by at least one other feasible solution particle are classified into the second level, and all infeasible solutions are classified into the third level.

[0137] Based on this, a particle is selected from the first level as the globally optimal particle; if the first level is empty, the particle with the smallest purification pressure drop is selected from the second level as the globally optimal particle; if both the first and second levels are empty, the particle with the smallest purification pressure drop is selected from the third level as the globally optimal particle. The globally optimal particle is the reference particle selected in the current iteration to guide the updating direction and velocity of all particles, and its control scheme represents the best or second-best candidate solution in the current population.

[0138] For example, after non-dominated sorting, the first level contains 8 particles, the second level contains 25 particles, and the third level contains 17 particles. The first level is not empty, and one of the 8 particles in the first level is randomly selected as the global optimal particle. The code carried by this global optimal particle is [On 2.8, On 2.7, On 2.6, On 2.5, On 2.5, On 2.2]. The total oxygen production flow rate is 2.8 + 2.7 + 2.6 + 2.5 + 2.5 + 2.2 = 16.3 L / min. The absolute value of the oxygen supply difference is |16.3 − 16.25| = 0.05 L / min, and the oxygen supply fitness is 1 / (0.05 + 1) = 1 / 1.05 ≈ 0.952. The individual disinfection residence time is 50 L / 16.3 L / min × 60 = 184 seconds. The absolute value of the disinfection difference = |184.0−300| = 116 seconds, and the disinfection fitness = 1 / (116+1)≈0.00855. The purification pressure drop calculated using the preset pressure drop model is 85 Pa, which is less than the inflection point threshold of 100 Pa, and therefore no penalty is imposed. This particle is selected as the globally optimal particle.

[0139] Then, based on the start / stop status and flow rate settings of each oxygen-generating unit carried by the globally optimal particle, the positions and velocities of all particles in the current population are updated to generate the next generation population. Position refers to the coordinate vector of a particle in the multi-dimensional search space, with each dimension corresponding to the flow rate setting of an oxygen-generating unit. Velocity refers to the direction and step size vector of a particle's movement in the multi-dimensional search space.

[0140] Specifically, based on the start / stop status and flow rate settings of each oxygen-generating unit carried by the globally optimal particle, the positions and velocities of all particles in the current population are updated, including:

[0141] Based on the start / stop status and flow rate setting of each oxygen generation unit carried by the global optimal particle, the preset noise mapping table is queried to obtain the overall noise value corresponding to the global optimal particle.

[0142] The noise gradient is calculated based on the overall noise level and the number of oxygen-generating units operating at the global optimal particle level.

[0143] The position update step size and velocity update step size of the particles in the next generation population are multiplied by an adjustment factor, which is inversely proportional to the noise gradient.

[0144] The adjusted position update step size and velocity update step size are used to update the position and velocity of all particles in the current population, generating the next generation population.

[0145] First, based on the start / stop status and flow rate setpoint of each oxygen-generating unit carried by the globally optimal particle, a preset noise mapping table is consulted to obtain the overall noise value corresponding to the globally optimal particle. The preset noise mapping table is a table establishing the correspondence between oxygen-generating unit combinations and overall noise values ​​through actual measurement and calibration before the equipment leaves the factory. The input to the preset noise mapping table is the start / stop status and flow rate setpoint of each oxygen-generating unit, and the output is the corresponding overall noise value. The overall noise value refers to the comprehensive noise generated by the integrated oxygen-generating, disinfecting, and purifying machine under specific operating conditions.

[0146] Specifically, the overall noise value corresponding to each particle is obtained. Based on the start / stop status and flow rate setting of each oxygen generating unit carried by the globally optimal particle, a preset noise mapping table is consulted to obtain the overall noise value corresponding to the globally optimal particle. For example, the globally optimal particle carries the code [On 2.8, On 2.7, On 2.6, On 2.5, On 2.5, On 2.2], and all 6 oxygen generating units are in operation, with flow rate setting values ​​of 2.8L / min, 2.7L / min, 2.6L / min, 2.5L / min, 2.5L / min, and 2.2L / min respectively, for a total oxygen generation flow rate of 16.3L / min. Consulting the preset noise mapping table, the overall noise value of all 6 units of the globally optimal particle when all are running is 52dB.

[0147] Secondly, based on the overall noise level and the number of operating oxygen-generating units for the globally optimal particle, the noise gradient is calculated. The noise gradient refers to the average change in the overall noise level as the number of operating oxygen-generating units increases by one. All particles in the current population are grouped according to the number of operating oxygen-generating units, ranging from 1 to the total number of oxygen-generating units, and the average overall noise level of particles within each group is calculated. The average noise increase for each segment as the number of operating oxygen-generating units increases from 1 to 2, from 2 to 3, and up to N-1 to N is summed, and divided by the total number of segments to obtain the noise gradient. Where k is the number of operating oxygen generating units, ranging from 1 to N−1, and N is the total number of oxygen generating units.

[0148] For example, the overall noise values ​​for running 1 to 6 units were queried. The results are: 42dB for 1 unit, 46dB for 2 units, 50dB for 3 units, 54dB for 4 units, 56dB for 5 units, and 58dB for 6 units. The noise difference between adjacent unit numbers is: 46-42=4dB, 50-46=4dB, 54-50=4dB, 56-54=2dB, 58-56=2dB. Summing up these differences (4+4+4+2+2=16dB) and dividing by the number of unit number changes (6-1=5), the noise gradient is 3.2dB / unit.

[0149] Next, the position update step size and velocity update step size of the particles in the next generation population are multiplied by an adjustment factor, which is inversely proportional to the noise gradient. The position update step size refers to the distance a particle moves in each dimension during one iteration. The velocity update step size refers to the magnitude of the change in the velocity vector of a particle during one iteration. The adjustment factor is a coefficient used to scale the position update step size and velocity update step size, and it is inversely proportional to the noise gradient.

[0150] Specifically, the adjustment factor = 1 / (1 + noise gradient). The adjusted position update step size is obtained by multiplying the preset position update step size by the adjustment factor; the adjusted velocity update step size is obtained by multiplying the preset velocity update step size by the adjustment factor. For example, the adjustment factor = 1 / (1+3.2) ≈ 0.238. The preset position update step size is 0.5 L / min, and the preset velocity update step size is 0.3 L / min. The adjusted position update step size = 0.5 × 0.238 ≈ 0.119 L / min; the adjusted velocity update step size = 0.3 × 0.238 ≈ 0.071 L / min.

[0151] Then, the positions and velocities of all particles in the current population are updated using the adjusted position update step size and velocity update step size to generate the next generation population. The next generation population refers to the new population formed after all particles in the current population have completed their position and velocity updates, and this population will serve as the current population for the next iteration.

[0152] Specifically, for each particle in the current swarm, we first obtain its current position vector, current velocity vector, individual historical best position vector, and global best particle position vector. We then update the particle's velocity vector using the standard particle velocity update formula in particle swarm optimization. The particle's new velocity consists of three parts: the particle's original velocity at the previous moment multiplied by an inertia weight coefficient; the difference between the particle's individual historical best position and current position multiplied by an individual cognitive coefficient and a random number; and the difference between the global best particle position and current position multiplied by a social cognitive coefficient and a random number. We add these three parts together to obtain the particle's new velocity vector. Finally, we add the particle's current position vector to the new velocity vector to obtain the particle's new position vector.

[0153] Secondly, boundary constraints are applied to the new position vector. The flow rate setting for each dimension is checked to ensure it is within the allowable range. If the flow rate setting for a certain dimension is less than 0, it is forced to be 0; if it is greater than the rated maximum oxygen production flow rate of the oxygen generating unit, it is forced to be set to the rated maximum oxygen production flow rate.

[0154] Next, the encoding of each particle is updated based on the new position vector after boundary constraints. The start / stop state of the corresponding oxygen generating unit is determined based on the flow rate setting value in each dimension of the new position: if the flow rate setting value is greater than 0, the start / stop state of the oxygen generating unit is set to on; if the flow rate setting value is equal to 0, the start / stop state of the oxygen generating unit is set to off. The updated start / stop state and the flow rate setting value together constitute the new encoding of the particle.

[0155] Finally, the above steps are performed on all particles in sequence to update the position and velocity of all particles in the current population. All the updated particles together constitute the next generation population.

[0156] For example, consider the first particle in the current swarm. The particle's current position x(t) = [2.1, 0, 1.5, 0.8, 0, 0] L / min, current velocity v(t) = [0.1, 0, 0.2, -0.1, 0, 0] L / min, individual historical best position pbest = [2.5, 0, 2.0, 1.0, 0, 0] L / min, and global best particle position gbest = [2.8, 2.7, 2.6, 2.5, 2.5, 2.2] L / min. The inertia weight coefficient w = 0.8, individual cognitive coefficient c1 = 1.5, social cognitive coefficient c2 = 1.5, and the random number rand() is set to 0.5. The velocity is updated using the particle swarm velocity update formula, and the updated velocity v(t+1) = [0.905, 0, 1.36, 1.345, 0, 0] L / min. Add the current position to the updated velocity to obtain the new position x(t+1) = [3.005, 0, 2.86, 2.145, 0, 0] L / min. Boundary constraint check: all dimensions are within the rated range of 0 to 5 L / min, no correction is required. Determine the start / stop state based on the updated flow rate settings: the first unit's flow rate of 3.005 L / min is greater than 0, set to on; the second unit's flow rate of 0 is set to off; the third unit's flow rate of 2.86 L / min is greater than 0, set to on; the fourth unit's flow rate of 2.145 L / min is greater than 0, set to on; the fifth and sixth units' flow rates are both 0, set to off. The updated code for this particle is [on 3.005, off 0, on 2.86, on 2.145, off 0, off 0]. The remaining 49 particles in the current population complete the position and velocity updates in the same way, and all the updated particles together constitute the next generation population.

[0157] Finally, the current step is repeated until the preset convergence condition is met, and the particles in the current population located at the first level are taken as the Pareto optimal solution set. The preset convergence condition refers to the criterion for terminating the iteration, which is that the number of iterations reaches a preset maximum number of iterations. The preset maximum number of iterations refers to the pre-set maximum number of iterations for swarm intelligence optimization; iteration terminates when the actual number of iterations reaches this preset value. The Pareto optimal solution set refers to the set of all particles that are not dominated by any other feasible solution after multi-objective iterative optimization. Each particle in this set represents a candidate control scheme that achieves Pareto optimality between oxygen supply fitness and disinfection fitness.

[0158] Specifically, at the end of each iteration, the number of iterations completed is recorded. It is then determined whether the current iteration count has reached the preset maximum number of iterations. If the current iteration count is less than the preset maximum number of iterations, the next iteration round continues. If the current iteration count equals the preset maximum number of iterations, the iteration terminates. After terminating the iteration, all particles in the first level of the current population are extracted, and the set of these particles is output as the Pareto optimal solution set.

[0159] For example, the preset maximum number of iterations is 500. The 200th iteration has been completed. Since the current iteration count (200) is less than the preset maximum number of iterations (500), the preset convergence condition has not been met. The next generation population generated in the above steps is used as the current population, and the next round of iterations continues. This iteration process is repeated, incrementing the iteration count by 1 after each iteration, until the 500th iteration is completed. After the 500th iteration, the current iteration count (500) equals the preset maximum number of iterations (500), satisfying the preset convergence condition, and the iteration terminates. At this point, there are 8 particles in the first level of the current population. The set of these 8 particles is the Pareto optimal solution set. Each particle in this Pareto optimal solution set is a candidate control scheme that cannot be dominated by other feasible solutions. For example, two schemes are: Scheme A [On 2.8, On 2.7, On 2.6, On 2.5, On 2.5, On 2.2], with a total oxygen flow rate of 16.3 L / min, oxygen fitness of 0.952, individual disinfection residence time of 184.0 seconds, disinfection fitness of 0.00855, and purification pressure drop of 85 Pa; Scheme B [On 2.5, On 2.5, On 2.5, On 2.5, On 2.5, On 2.5], with a total oxygen flow rate of 15.0 L / min, oxygen fitness of 0.444, individual disinfection residence time of 200.0 seconds, disinfection fitness of 0.0099, and purification pressure drop of 70 Pa. Scheme A has a higher oxygen fitness, and Scheme B has a higher disinfection fitness; they are not mutually dominant and are both Pareto optimal solutions.

[0160] In this embodiment of the invention, by quantifying oxygen supply fitness and disinfection fitness into continuous values ​​between 0 and 1, the matching degree of oxygen supply demand and the matching degree of disinfection residence time are unified under the same evaluation framework for non-dominated ranking and hierarchical division. Infeasible solutions are marked and penalized through dynamic disinfection residence time constraints and purification pressure drop constraints, guiding the population to search within the feasible region. The position and velocity update step size of all particles are dynamically adjusted based on the noise gradient calculated from the globally optimal particle's running state. When the noise gradient is large, the search step size is automatically reduced to avoid generating high-noise solutions; when the noise gradient is small, a larger search step size is maintained to ensure convergence speed. Through the particle swarm velocity and position update mechanism, the population, guided by the globally optimal particle, approaches the Pareto front generation by generation, ultimately obtaining a set of candidate control schemes that achieve Pareto optimality between the two objectives of oxygen supply and disinfection, providing a diverse set of optimal solutions for subsequent user mode decisions.

[0161] S500: Select an optimal solution from the Pareto optimal solution set as the control command for the current control cycle, and output the control command.

[0162] Through multi-objective iterative optimization, the Pareto optimal solution set contains multiple candidate control schemes where oxygen supply fitness and disinfection fitness are mutually independent. These candidate control schemes cannot be directly ranked based on a simple dominance relationship. In practical applications, different scenarios have different priorities regarding oxygen supply and disinfection: high-altitude areas or densely populated places may prioritize oxygen supply efficiency, hospitals or laboratories with high air quality requirements may prioritize disinfection effectiveness, while regular office or residential scenarios require a balanced approach. Therefore, based on the user's preset control mode, it is necessary to assign different weight coefficients to oxygen supply fitness and disinfection fitness, select the optimal solution from the Pareto optimal solution set that best suits the current user preferences, and output its accompanying control commands to the device for execution.

[0163] Step S500 in the method provided in this embodiment of the invention includes:

[0164] Obtain user-preset control modes, including oxygen supply priority mode, disinfection priority mode, and balanced mode;

[0165] The first weighting coefficient corresponding to oxygen supply adaptability and the second weighting coefficient corresponding to disinfection adaptability are determined according to the control mode.

[0166] The oxygen fitness of each particle in the Pareto optimal solution set is multiplied by the first weighting coefficient to obtain the first weighted value;

[0167] The disinfection fitness of each particle is multiplied by the second weighting coefficient to obtain the second weighted value;

[0168] The first weighted value and the second weighted value are added together to obtain the overall fitness of each particle;

[0169] The particle with the highest overall fitness is selected, and the start / stop status and flow rate set value of each oxygen generation unit carried by the particle with the highest overall fitness are used as the control command.

[0170] First, the system acquires the user's preset control modes, including oxygen supply priority mode, disinfection priority mode, and balanced mode. A control mode refers to the decision preference type pre-set by the user based on the current usage scenario and needs, used to guide the weight allocation when selecting the optimal solution from the Pareto optimal solution set. The oxygen supply priority mode is suitable for scenarios with high population density and urgent oxygen demand, prioritizing oxygen supply adaptability during decision-making. The disinfection priority mode is suitable for scenarios with high air quality requirements, prioritizing disinfection adaptability during decision-making. The balanced mode is suitable for regular usage scenarios, giving equal weight to oxygen supply adaptability and disinfection adaptability during decision-making.

[0171] Specifically, the system reads the control mode flag stored internally. This flag is set by the user via the device control panel or a remote terminal. If the flag indicates oxygen supply priority mode, then the system is confirmed to be in oxygen supply priority mode; if the flag indicates disinfection priority mode, then the system is confirmed to be in disinfection priority mode; if the flag indicates equalization mode, then the system is confirmed to be in equalization mode. For example, the system reads the control mode flag stored internally. If the user sets the control mode to equalization mode via a remote terminal, the current value of the flag is equalization mode. Therefore, the current control mode is confirmed to be equalization mode.

[0172] Secondly, based on the control mode, determine the first weighting coefficient corresponding to oxygen supply adaptability and the second weighting coefficient corresponding to disinfection adaptability. The first weighting coefficient refers to the weight ratio of oxygen supply adaptability in the overall adaptability calculation, and is a constant greater than 0 and less than 1. The second weighting coefficient refers to the weight ratio of disinfection adaptability in the overall adaptability calculation, and is a constant greater than 0 and less than 1. The sum of the first weighting coefficient and the second weighting coefficient equals 1.

[0173] Specifically, if the control mode is oxygen supply priority mode, the first weighting coefficient is set to 0.7 and the second weighting coefficient is set to 0.3. If the control mode is disinfection priority mode, the first weighting coefficient is set to 0.3 and the second weighting coefficient is set to 0.7. If the control mode is balanced mode, the first weighting coefficient is set to 0.5 and the second weighting coefficient is set to 0.5. For example, if the current control mode is balanced mode, according to the weighting allocation rules of balanced mode, the first weighting coefficient corresponding to oxygen supply fitness is set to 0.5, and the second weighting coefficient corresponding to disinfection fitness is set to 0.5.

[0174] Next, the oxygen fitness of each particle in the Pareto optimal solution set is multiplied by the first weighting coefficient to obtain the first weighted value; the disinfection fitness of each particle is multiplied by the second weighting coefficient to obtain the second weighted value; the first weighted value and the second weighted value are added together to obtain the comprehensive fitness of each particle. The comprehensive fitness is the weighted sum obtained by multiplying the oxygen fitness and disinfection fitness by their respective weighting coefficients, and is used to rank the particles in the Pareto optimal solution set based on a single index.

[0175] Specifically, the oxygen supply fitness and disinfection fitness of each particle in the Pareto optimal solution set are obtained. For each particle in the Pareto optimal solution set, the oxygen supply fitness is multiplied by a first weighting coefficient to obtain a first weighted value. For each particle in the Pareto optimal solution set, the disinfection fitness is multiplied by a second weighting coefficient to obtain a second weighted value. For each particle in the Pareto optimal solution set, the first weighted value and the second weighted value are added together to obtain the comprehensive fitness.

[0176] For example, the first weighting coefficient is 0.5, and the second weighting coefficient is 0.5. The Pareto optimal solution set contains 8 particles, taking Scheme A and Scheme B as examples. Scheme A has an oxygen supply fitness of 0.952 and a disinfection fitness of 0.00855. Scheme B has an oxygen supply fitness of 0.444 and a disinfection fitness of 0.0099. The first weighted value of Scheme A = 0.952 × 0.5 = 0.476, the second weighted value = 0.00855 × 0.5 = 0.004275, and the overall fitness = 0.476 + 0.004275 = 0.480275. The first weighted value of Scheme B = 0.444 × 0.5 = 0.222, the second weighted value = 0.0099 × 0.5 = 0.00495, and the overall fitness = 0.222 + 0.00495 = 0.22695. The remaining 6 particles in the Pareto optimal solution set are all obtained by multiplying the oxygen supply fitness of each particle by the first weighting coefficient and the disinfection fitness by the second weighting coefficient, and then adding them together to obtain the comprehensive fitness of each particle.

[0177] Finally, the particle with the highest overall adaptability is selected, and the start / stop status and flow rate setpoint of each oxygen-generating unit carried by the particle with the highest overall adaptability are used as control commands. The control commands are the control commands that are finally output to the integrated oxygen generator, disinfection and purification machine for execution, and include the start / stop status and flow rate setpoint of each oxygen-generating unit.

[0178] Specifically, among all particles in the Pareto optimal solution set, the overall fitness of each particle is compared, and the particle with the highest overall fitness is identified. The start / stop status and flow rate setpoint of each oxygen generation unit carried by this particle are extracted and used as the control command for the current control cycle. The control command is sent to the actuator of the integrated oxygen generation, disinfection, and purification machine, which then performs actual operation according to the start / stop status and flow rate setpoint of each oxygen generation unit in the control command.

[0179] For example, the Pareto optimal solution set contains 8 particles, with the following overall fitness values: Scheme A has a fitness of 0.480275, Scheme B has a fitness of 0.22695, and the remaining 6 particles have fitness values ​​of 0.312, 0.287, 0.198, 0.175, 0.321, and 0.256, respectively. Scheme A has the highest overall fitness value of 0.480275, therefore Scheme A is selected as the optimal solution. Scheme A carries the codes [On 2.8, On 2.7, On 2.6, On 2.5, On 2.5, On 2.2], which represent the start / stop status and flow rate settings of each oxygen generating unit as follows: First oxygen generating unit on, flow rate setting 2.8 L / min; Second oxygen generating unit on, flow rate setting 2.7 L / min; Third oxygen generating unit on, flow rate setting 2.6 L / min; Fourth oxygen generating unit on, flow rate setting 2.5 L / min; Fifth oxygen generating unit on, flow rate setting 2.5 L / min; Sixth oxygen generating unit on, flow rate setting 2.2 L / min. This code is output as the control command for the current control cycle to the actuator of the integrated oxygen generator, disinfection, and purification machine. The actuator controls the operation of each oxygen generating unit according to the aforementioned start / stop status and flow rate settings.

[0180] In this embodiment of the invention, by acquiring the user-preset control mode and determining the weight coefficients corresponding to oxygen supply fitness and disinfection fitness according to the mode, the same Pareto optimal solution set can adapt to the decision preferences of different application scenarios, thus improving scenario adaptability. By multiplying the oxygen supply fitness and disinfection fitness of each particle by their corresponding weight coefficients and then summing them, the overall fitness is obtained, transforming multi-objective decision-making into a single-index ranking, simplifying the selection process of the optimal solution. By selecting the particle with the highest overall fitness and outputting its start / stop status and flow setpoint as control commands to the actuator, a complete closed-loop control from optimization decision-making to equipment execution is achieved.

[0181] Through the specific implementation methods described above, the embodiments of the present invention achieve the following technical effects:

[0182] This invention provides an intelligent control method and system for an integrated oxygen generator, disinfection, and purification machine that combines swarm intelligence. By acquiring indoor environmental parameters and equipment operating parameters, it calculates the oxygen demand and the baseline disinfection residence time corrected by the measured values ​​of the previous cycle, providing an accurate data foundation for multi-objective optimization. Based on this, an initial population is constructed using the start / stop status and flow rate setpoints of each oxygen generator unit as decision variables. Non-dominated sorting is performed with oxygen supply fitness and disinfection fitness as dual objectives. During iteration, a dynamic disinfection constraint threshold, a purification pressure drop penalty mechanism, and an adaptive step size adjustment based on noise gradient are introduced to guide the population towards the Pareto front. Finally, the optimal solution is selected from the Pareto optimal solution set according to the user-preset mode as the control command output. This invention achieves a Pareto optimal trade-off between oxygen supply efficiency and disinfection effect, while also considering the constraints of purification pressure drop and overall machine noise, effectively improving the comprehensive control efficiency of the integrated machine in dynamic indoor environments.

[0183] Example 2, as Figure 3 As shown, this invention provides an intelligent control system for an integrated oxygen generator, disinfection, and purification machine that combines swarm intelligence. The system includes:

[0184] The parameter acquisition module 11 is used to acquire the current indoor environmental parameters and the equipment operating parameters of the oxygen generation, disinfection and purification integrated machine. The environmental parameters include indoor oxygen concentration, indoor carbon dioxide concentration and number of people in the room. The equipment operating parameters include the total number of oxygen generation units, the oxygen generation flow rate of each oxygen generation unit and the actual disinfection dwell time of the previous control cycle.

[0185] The demand calculation module 12 is used to calculate the oxygen supply demand for the current control cycle based on the environmental parameters, and to calculate the baseline disinfection dwell time for the current control cycle based on the oxygen production flow rate of each oxygen production unit and the actual disinfection dwell time of the previous control cycle.

[0186] Population building module 13 is used to build an initial population based on the oxygen demand and the total number of oxygen generating units, wherein each particle in the initial population carries the start / stop status and flow rate setting value of each oxygen generating unit.

[0187] Iterative optimization module 14 is used to perform multi-objective iterative updates on the initial population based on the oxygen supply demand, the baseline disinfection dwell time, and the total number of oxygen generating units to obtain a Pareto optimal solution set;

[0188] The instruction output module 15 is used to select an optimal solution as the control instruction for the current control cycle based on the Pareto optimal solution set, and output the control instruction.

[0189] In one embodiment, the parameter acquisition module 11 is further configured to:

[0190] Within a preset sliding time window, the indoor oxygen concentration, indoor carbon dioxide concentration, number of people in the room, total number of oxygen generating units, oxygen flow rate of each oxygen generating unit, and actual disinfection dwell time of the previous control cycle are collected according to a preset sampling cycle.

[0191] When the current control cycle is the first control cycle of the equipment's first power-on operation, the actual disinfection dwell time of the previous control cycle is taken as the preset initial calibration value.

[0192] In one embodiment, the demand calculation module 12 is further configured to:

[0193] The calculation of the oxygen demand for the current control cycle based on the environmental parameters includes:

[0194] Calculate the oxygen concentration deviation based on the indoor oxygen concentration and the preset target oxygen concentration;

[0195] The rate of change of indoor carbon dioxide concentration is calculated based on the data sequence of indoor carbon dioxide concentration within the preset sliding time window, and the change in the number of people indoors is calculated in combination with the preset ventilation rate.

[0196] The basic oxygen demand is calculated based on the number of people in the room and the preset oxygen demand coefficient per person.

[0197] The deviation compensation amount is calculated based on the oxygen concentration deviation and the preset concentration deviation compensation coefficient;

[0198] The trend compensation amount is calculated based on the change in the number of people in the room and the preset personnel change compensation coefficient.

[0199] The oxygen demand for the current control cycle is obtained by adding the basic oxygen demand, the deviation compensation amount, and the trend compensation amount.

[0200] The baseline disinfection dwell time for the current control cycle is calculated based on the oxygen flow rate of each oxygen generation unit and the actual disinfection dwell time of the previous control cycle, including:

[0201] Obtain the nominal volume of the disinfection chamber of the integrated oxygen generator, disinfection and purification machine;

[0202] Calculate the theoretical value of the total oxygen production flow rate based on the oxygen production flow rate of each oxygen production unit in operation;

[0203] Divide the nominal volume of the disinfection chamber by the theoretical value of the total oxygen flow rate to obtain the theoretical value of the current disinfection residence time. If the theoretical value of the total oxygen flow rate is zero, then set the theoretical value of the current disinfection residence time to the preset maximum allowable residence time.

[0204] When the current control cycle is not the first control cycle, the residence time deviation coefficient is calculated based on the actual disinfection residence time of the previous control cycle and the theoretical value of the disinfection residence time of the previous control cycle.

[0205] The theoretical value of the current disinfection dwell time is corrected by the dwell time deviation coefficient to obtain the benchmark disinfection dwell time for the current control cycle.

[0206] In one embodiment, population building module 13 is further configured to:

[0207] The number of coding groups for population particles is determined based on the total number of oxygen generating units. Each coding group corresponds to one oxygen generating unit, and each coding group includes the start / stop status and flow rate setting value of the oxygen generating unit.

[0208] Based on the ratio of the oxygen demand to the total number of oxygen generating units, the initial range of the flow rate setting value for each oxygen generating unit is determined. The lower limit of the initial range of the flow rate setting value is 0, and the upper limit is the ratio of the oxygen demand to the total number of oxygen generating units.

[0209] Within the initial range of the flow rate setting value, a flow rate setting value is randomly generated for each group of codes of each particle to generate an initial population of a preset size.

[0210] In one embodiment, the iterative optimization module 14 is further configured to:

[0211] Obtain the start / stop status and flow rate setpoint of each oxygen-generating unit carried by each particle in the current population;

[0212] The total oxygen production flow rate of each oxygen generating unit in operation is obtained by adding the flow rate setpoints of each unit.

[0213] The oxygen supply adaptability of each particle is calculated based on the absolute value of the difference between the total oxygen production flow rate of each particle and the oxygen supply demand, wherein the oxygen supply adaptability is negatively correlated with the absolute value of the oxygen supply difference.

[0214] Calculate the individual disinfection residence time for each particle based on the total oxygen production flow rate of each particle and the nominal volume of the disinfection chamber.

[0215] The disinfection fitness of each particle is calculated based on the absolute value of the disinfection difference between the individual disinfection dwell time and the preset disinfection dwell time constraint threshold, wherein the disinfection fitness is negatively correlated with the absolute value of the disinfection difference.

[0216] Based on the oxygen supply adaptability and disinfection adaptability of each particle, and combined with preset constraints, all particles are sorted in a non-dominated order, and all particles are divided into a first level, a second level and a third level. The first level consists of feasible solution particles that are not dominated by any other particle, the second level consists of feasible solution particles that are dominated by at least one other particle, and the third level consists of infeasible solution particles.

[0217] Select a particle from the first level as the global optimal particle;

[0218] If the first level is empty, then the particle with the smallest purification pressure drop is selected from the second level as the global optimal particle.

[0219] If both the first and second levels are empty, then the particle with the smallest purification pressure drop is selected from the third level as the global optimal particle.

[0220] Based on the start / stop status and flow rate setting of each oxygen-generating unit carried by the global optimal particle, update the position and velocity of all particles in the current population to generate the next generation population.

[0221] Repeat the current step until the preset convergence condition is met, and use the particles in the current population located at the first level as the Pareto optimal solution set.

[0222] Based on the oxygen supply and disinfection adaptability of each particle, and combined with preset constraints, all particles are non-dominated and sorted, dividing them into three levels: Level 1, Level 2, and Level 3, including:

[0223] Based on the baseline disinfection dwell time and the preset baseline disinfection threshold, a dynamic disinfection dwell time constraint threshold is generated;

[0224] Based on the start / stop status and flow rate settings of each oxygen-generating unit carried by each particle in the current population, and based on the series and parallel relationship of the flow resistance of each oxygen-generating unit's gas channel, the purification pressure drop corresponding to the particle is calculated through a preset pressure drop model.

[0225] Particles whose individual disinfection dwell time is greater than or equal to the dynamic disinfection dwell time constraint threshold, or particles whose purification pressure drop is greater than the preset purification pressure drop upper limit threshold, are marked as infeasible solutions and classified into the third level.

[0226] Particles not marked as infeasible solutions are identified as feasible solutions. For feasible solution particles whose purification pressure drop is greater than the preset purification pressure drop inflection point threshold and less than or equal to the preset purification pressure drop upper limit threshold, their oxygen supply adaptability is multiplied by a preset penalty coefficient.

[0227] Compare the dominance relationships of all feasible solution particles: For any two feasible solution particles, if the oxygen supply fitness of the first particle is greater than or equal to the oxygen supply fitness of the second particle, and the disinfection fitness of the first particle is greater than or equal to the disinfection fitness of the second particle, and at least one fitness is greater than the corresponding fitness of the second particle, then it is determined that the first particle dominates the second particle.

[0228] Feasible solution particles that are not dominated by any other feasible solution particles are classified into the first level, and feasible solution particles that are dominated by at least one other feasible solution particle are classified into the second level.

[0229] Specifically, based on the start / stop status and flow rate setpoint of each oxygen-generating unit carried by the globally optimal particle, the positions and velocities of all particles in the current population are updated, including:

[0230] Based on the start / stop status and flow rate setting of each oxygen generation unit carried by the global optimal particle, the preset noise mapping table is queried to obtain the overall noise value corresponding to the global optimal particle.

[0231] The noise gradient is calculated based on the overall noise level and the number of oxygen-generating units operating at the global optimal particle level.

[0232] The position update step size and velocity update step size of the particles in the next generation population are multiplied by an adjustment factor, which is inversely proportional to the noise gradient.

[0233] The adjusted position update step size and velocity update step size are used to update the position and velocity of all particles in the current population, generating the next generation population.

[0234] In one embodiment, the instruction output module 15 is further configured to:

[0235] Obtain user-preset control modes, including oxygen supply priority mode, disinfection priority mode, and balanced mode;

[0236] The first weighting coefficient corresponding to oxygen supply adaptability and the second weighting coefficient corresponding to disinfection adaptability are determined according to the control mode.

[0237] The oxygen fitness of each particle in the Pareto optimal solution set is multiplied by the first weighting coefficient to obtain the first weighted value;

[0238] The disinfection fitness of each particle is multiplied by the second weighting coefficient to obtain the second weighted value;

[0239] The first weighted value and the second weighted value are added together to obtain the overall fitness of each particle;

[0240] The particle with the highest overall fitness is selected, and the start / stop status and flow rate set value of each oxygen generation unit carried by the particle with the highest overall fitness are used as the control command.

[0241] It should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for intelligent control of an integrated oxygen generator, disinfection, and purification machine combining swarm intelligence, characterized in that: The method includes: The system acquires current indoor environmental parameters and equipment operating parameters of the integrated oxygen generator, disinfection and purification unit. The environmental parameters include indoor oxygen concentration, indoor carbon dioxide concentration and number of people in the room. The equipment operating parameters include the total number of oxygen generators, the oxygen flow rate of each oxygen generator, and the actual disinfection dwell time in the previous control cycle. The oxygen demand for the current control cycle is calculated based on the environmental parameters, and the baseline disinfection dwell time for the current control cycle is calculated based on the oxygen production flow rate of each oxygen production unit and the actual disinfection dwell time of the previous control cycle. Based on the oxygen demand and the total number of oxygen generating units, an initial population is constructed, wherein each particle in the initial population carries the start / stop status and flow rate setting value of each oxygen generating unit. Based on the oxygen demand, the baseline disinfection dwell time, and the total number of oxygen generating units, the initial population is iteratively updated using a multi-objective method to obtain the Pareto optimal solution set. Based on the Pareto optimal solution set, an optimal solution is selected as the control command for the current control cycle, and the control command is output.

2. The intelligent control method for an integrated oxygen generator, disinfection, and purification machine combining swarm intelligence as described in claim 1, is characterized in that, Obtain current indoor environmental parameters and equipment operating parameters of the integrated oxygen generator, disinfection, and purification unit, including: Within a preset sliding time window, the indoor oxygen concentration, indoor carbon dioxide concentration, number of people in the room, total number of oxygen generating units, oxygen flow rate of each oxygen generating unit, and actual disinfection dwell time of the previous control cycle are collected according to a preset sampling cycle. When the current control cycle is the first control cycle of the equipment's first power-on operation, the actual disinfection dwell time of the previous control cycle is taken as the preset initial calibration value.

3. The intelligent control method for an integrated oxygen generator, disinfection, and purification machine combining swarm intelligence as described in claim 2, is characterized in that... The oxygen demand for the current control cycle is calculated based on the environmental parameters, including: Calculate the oxygen concentration deviation based on the indoor oxygen concentration and the preset target oxygen concentration; The rate of change of indoor carbon dioxide concentration is calculated based on the data sequence of indoor carbon dioxide concentration within the preset sliding time window, and the change in the number of people indoors is calculated in combination with the preset ventilation rate. The basic oxygen demand is calculated based on the number of people in the room and the preset oxygen demand coefficient per person. The deviation compensation amount is calculated based on the oxygen concentration deviation and the preset concentration deviation compensation coefficient; The trend compensation amount is calculated based on the change in the number of people in the room and the preset personnel change compensation coefficient. The oxygen demand for the current control cycle is obtained by adding the basic oxygen demand, the deviation compensation amount, and the trend compensation amount.

4. The intelligent control method for an integrated oxygen generator, disinfection, and purification machine combining swarm intelligence as described in claim 1, characterized in that, The baseline disinfection dwell time for the current control cycle is calculated based on the oxygen flow rate of each oxygen generation unit and the actual disinfection dwell time of the previous control cycle, including: Obtain the nominal volume of the disinfection chamber of the integrated oxygen generator, disinfection and purification machine; Calculate the theoretical value of the total oxygen production flow rate based on the oxygen production flow rate of each oxygen production unit in operation; Divide the nominal volume of the disinfection chamber by the theoretical value of the total oxygen flow rate to obtain the theoretical value of the current disinfection residence time. If the theoretical value of the total oxygen flow rate is zero, then set the theoretical value of the current disinfection residence time to the preset maximum allowable residence time. When the current control cycle is not the first control cycle, the residence time deviation coefficient is calculated based on the actual disinfection residence time of the previous control cycle and the theoretical value of the disinfection residence time of the previous control cycle. The theoretical value of the current disinfection dwell time is corrected by the dwell time deviation coefficient to obtain the benchmark disinfection dwell time for the current control cycle.

5. The intelligent control method for an integrated oxygen generator, disinfection, and purification machine combining swarm intelligence as described in claim 1, characterized in that, Based on the oxygen demand and the total number of oxygen production units, an initial population is constructed, including: The number of coding groups for population particles is determined based on the total number of oxygen generating units. Each coding group corresponds to one oxygen generating unit, and each coding group includes the start / stop status and flow rate setting value of the oxygen generating unit. Based on the ratio of the oxygen demand to the total number of oxygen generating units, the initial range of the flow rate setting value for each oxygen generating unit is determined. The lower limit of the initial range of the flow rate setting value is 0, and the upper limit is the ratio of the oxygen demand to the total number of oxygen generating units. Within the initial range of the flow rate setting value, a flow rate setting value is randomly generated for each group of codes of each particle to generate an initial population of a preset size.

6. The intelligent control method for an integrated oxygen generator, disinfection, and purification machine combining swarm intelligence as described in claim 1, characterized in that, Based on the oxygen demand, the baseline disinfection dwell time, and the total number of oxygen generating units, the initial population is iteratively updated using a multi-objective method to obtain a Pareto optimal solution set, including: Obtain the start / stop status and flow rate setpoint of each oxygen-generating unit carried by each particle in the current population; The total oxygen production flow rate of each oxygen generating unit in operation is obtained by adding the flow rate setpoints of each unit. The oxygen supply adaptability of each particle is calculated based on the absolute value of the difference between the total oxygen production flow rate of each particle and the oxygen supply demand, wherein the oxygen supply adaptability is negatively correlated with the absolute value of the oxygen supply difference. Calculate the individual disinfection residence time for each particle based on the total oxygen production flow rate of each particle and the nominal volume of the disinfection chamber. The disinfection fitness of each particle is calculated based on the absolute value of the disinfection difference between the individual disinfection dwell time and the preset disinfection dwell time constraint threshold, wherein the disinfection fitness is negatively correlated with the absolute value of the disinfection difference. Based on the oxygen supply adaptability and disinfection adaptability of each particle, and combined with preset constraints, all particles are sorted in a non-dominated order, and all particles are divided into a first level, a second level and a third level. The first level consists of feasible solution particles that are not dominated by any other particle, the second level consists of feasible solution particles that are dominated by at least one other particle, and the third level consists of infeasible solution particles. Select a particle from the first level as the global optimal particle; If the first level is empty, then the particle with the smallest purification pressure drop is selected from the second level as the global optimal particle. If both the first and second levels are empty, then the particle with the smallest purification pressure drop is selected from the third level as the global optimal particle. Based on the start / stop status and flow rate setting of each oxygen-generating unit carried by the global optimal particle, update the position and velocity of all particles in the current population to generate the next generation population. Repeat the current step until the preset convergence condition is met, and use the particles in the current population located at the first level as the Pareto optimal solution set.

7. The intelligent control method for an integrated oxygen generator, disinfection, and purification machine combining swarm intelligence as described in claim 6, characterized in that, Based on the oxygen supply and disinfection adaptability of each particle, and combined with preset constraints, all particles are non-dominated and sorted, dividing them into three levels: Level 1, Level 2, and Level 3, including: Based on the baseline disinfection dwell time and the preset baseline disinfection threshold, a dynamic disinfection dwell time constraint threshold is generated; Based on the start / stop status and flow rate settings of each oxygen-generating unit carried by each particle in the current population, and based on the series and parallel relationship of the flow resistance of each oxygen-generating unit's gas channel, the purification pressure drop corresponding to the particle is calculated through a preset pressure drop model. Particles whose individual disinfection dwell time is greater than or equal to the dynamic disinfection dwell time constraint threshold, or particles whose purification pressure drop is greater than the preset purification pressure drop upper limit threshold, are marked as infeasible solutions and classified into the third level. Particles not marked as infeasible solutions are identified as feasible solutions. For feasible solution particles whose purification pressure drop is greater than the preset purification pressure drop inflection point threshold and less than or equal to the preset purification pressure drop upper limit threshold, their oxygen supply adaptability is multiplied by a preset penalty coefficient. Compare the dominance relationships of all feasible solution particles: For any two feasible solution particles, if the oxygen supply fitness of the first particle is greater than or equal to the oxygen supply fitness of the second particle, and the disinfection fitness of the first particle is greater than or equal to the disinfection fitness of the second particle, and at least one fitness is greater than the corresponding fitness of the second particle, then it is determined that the first particle dominates the second particle. Feasible solution particles that are not dominated by any other feasible solution particles are classified into the first level, and feasible solution particles that are dominated by at least one other feasible solution particle are classified into the second level.

8. The intelligent control method for an integrated oxygen generator, disinfection, and purification machine combining swarm intelligence as described in claim 6, characterized in that, Based on the start / stop status and flow rate setpoint of each oxygen-generating unit carried by the globally optimal particle, update the position and velocity of all particles in the current population, including: Based on the start / stop status and flow rate setting of each oxygen generation unit carried by the global optimal particle, the preset noise mapping table is queried to obtain the overall noise value corresponding to the global optimal particle. The noise gradient is calculated based on the overall noise level and the number of oxygen-generating units operating at the global optimal particle level. The position update step size and velocity update step size of the particles in the next generation population are multiplied by an adjustment factor, which is inversely proportional to the noise gradient. The adjusted position update step size and velocity update step size are used to update the position and velocity of all particles in the current population, generating the next generation population.

9. The intelligent control method for an integrated oxygen generator, disinfection, and purification machine combining swarm intelligence as described in claim 6, characterized in that, Based on the Pareto optimal solution set, an optimal solution is selected as the control command for the current control cycle, including: Obtain user-preset control modes, including oxygen supply priority mode, disinfection priority mode, and balanced mode; The first weighting coefficient corresponding to oxygen supply adaptability and the second weighting coefficient corresponding to disinfection adaptability are determined according to the control mode. The oxygen fitness of each particle in the Pareto optimal solution set is multiplied by the first weighting coefficient to obtain the first weighted value; The disinfection fitness of each particle is multiplied by the second weighting coefficient to obtain the second weighted value; The first weighted value and the second weighted value are added together to obtain the overall fitness of each particle; The particle with the highest overall fitness is selected, and the start / stop status and flow rate set value of each oxygen generation unit carried by the particle with the highest overall fitness are used as the control command.

10. An intelligent control system for an integrated oxygen generator, disinfection, and purification machine combining swarm intelligence, characterized in that: The system is used to implement the intelligent control method for an integrated oxygen generator, disinfection, and purification machine combining swarm intelligence as described in any one of claims 1 to 9, the system comprising: The parameter acquisition module is used to acquire the current indoor environmental parameters and the equipment operating parameters of the oxygen generator, disinfection and purification integrated machine. The environmental parameters include indoor oxygen concentration, indoor carbon dioxide concentration and number of people in the room. The equipment operating parameters include the total number of oxygen generators, the oxygen flow rate of each oxygen generator and the actual disinfection dwell time of the previous control cycle. The demand calculation module is used to calculate the oxygen supply demand for the current control cycle based on the environmental parameters, and to calculate the baseline disinfection dwell time for the current control cycle based on the oxygen production flow rate of each oxygen production unit and the actual disinfection dwell time of the previous control cycle. The population construction module is used to construct an initial population based on the oxygen demand and the total number of oxygen generating units, wherein each particle in the initial population carries the start / stop status and flow rate setting value of each oxygen generating unit. The iterative optimization module is used to perform multi-objective iterative updates on the initial population based on the oxygen demand, the baseline disinfection dwell time, and the total number of oxygen generating units to obtain the Pareto optimal solution set. The instruction output module is used to select an optimal solution as the control instruction for the current control cycle based on the Pareto optimal solution set, and output the control instruction.

Images