An incubator temperature and humidity self-adaptive adjusting method and system

Abstract

The application discloses a kind of incubator temperature and humidity self-adaptive regulation method and system, it is related to incubator temperature and humidity self-adaptive control technical field, comprising: real-time monitoring carbon dioxide concentration in incubator, calculate the real-time metabolic intensity index of embryo;According to real-time metabolic intensity index and current incubation age, query the preset mapping relationship, generate target absolute humidity value in incubator;Current absolute humidity value in incubator is obtained, and in combination with the target absolute humidity value, the optimal control instruction sequence of humidification device and dehumidification device in future period is solved by model predictive control algorithm rolling;Execute immediate control instruction in optimal control instruction sequence.The present application is with carbon dioxide concentration variation to represent embryo metabolic activity and generate dynamic target absolute humidity value, and realize prospective optimization control in combination with humidity dynamic prediction model and model predictive control algorithm, improve control precision, anti-disturbance ability and energy efficiency level under the premise of meeting physical operation constraint.

Description

Technical Field

[0001] This invention relates to the field of adaptive temperature and humidity control technology for incubators, specifically to an adaptive temperature and humidity adjustment method and system for incubators. Background Technology

[0002] In recent years, poultry egg incubation equipment has evolved from simple temperature control to a system of coordinated temperature and humidity control and refined environmental management. The industry generally adopts an architecture combining sensor measurement, actuator adjustment, and closed-loop controller scheduling to maintain a stable microenvironment within the incubator through humidification, ventilation, and dehumidification. With the increasing demands for large-scale incubation and lean farming, the controlled objects have gradually shifted from relative humidity to more physically consistent state variables such as absolute humidity and enthalpy. Control algorithms have expanded from threshold switching control and PID control to model-based predictive control and data-driven parameter tuning methods to improve stability, disturbance rejection, and energy efficiency.

[0003] Current incubator temperature and humidity control still relies primarily on "age curves + experience-based settings." Humidity settings often depend on fixed schedules or segmented thresholds, ignoring the metabolic and moisture exchange differences among embryos in different batches, under different loads, and with different ventilation conditions. This leads to a systematic deviation between humidity supply and actual physiological needs. First, traditional control methods often use relative humidity, which is significantly affected by temperature fluctuations and air pressure changes. The water vapor content corresponding to the same relative humidity is not constant, making it difficult to establish a transferable and comparable control benchmark. Second, common PID control is a passive correction mechanism. When faced with inertia, lag, coupling of humidification and ventilation, and disturbances such as opening doors for egg changing and turning, it is prone to overshoot, oscillation, and frequent conflicts with the actuator, leading to increased energy consumption and equipment wear. Third, current technologies typically do not incorporate embryonic metabolic status into the control loop, failing to translate changes in embryonic respiratory metabolic intensity into changes in humidity demand. Therefore, it is difficult to promptly suppress excessive water loss during periods of increased metabolism and to avoid environmental moisture redundancy during periods of decreased metabolism, ultimately affecting hatching uniformity and chick survival rate. Therefore, existing technologies struggle to achieve "target humidity generation driven by physiological feedback" and "coordinated regulation through forward prediction optimization," and it is even more difficult to obtain stable, energy-efficient, and repeatable control effects under constrained conditions. Summary of the Invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by this invention is that existing incubator temperature and humidity control methods suffer from the following problems: humidity setting depends on age experience curves, leading to demand mismatch; control indicators based on relative humidity are affected by factors such as temperature, resulting in unstable control benchmarks; traditional closed-loop control is prone to overshoot oscillation and increased energy consumption under conditions of high inertia and strong coupling; and the problem of how to convert embryo metabolic feedback into target absolute humidity and achieve optimal coordinated regulation based on model predictive control.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] In a first aspect, embodiments of the present invention provide an adaptive temperature and humidity control method for an incubator, characterized by comprising the following steps:

[0008] S1: Real-time monitoring of carbon dioxide concentration in the incubator, and calculation of real-time metabolic intensity indicators of embryos based on changes in carbon dioxide concentration.

[0009] S2: Based on the real-time metabolic intensity index and the current incubation age, query the preset mapping relationship and dynamically generate the corresponding target absolute humidity value in the incubator;

[0010] S3: Obtain the current absolute humidity value in the incubator, and combine it with the target absolute humidity value. Using the pre-established incubator humidity dynamic prediction model, the optimal control command sequence of the humidification device and dehumidification device in the future period is solved by the model prediction control algorithm.

[0011] S4: Execute the immediate control command in the optimal control command sequence to drive the humidification device and the dehumidification device to adjust.

[0012] As a preferred embodiment of the incubator temperature and humidity adaptive adjustment method of the present invention, in step S1, the calculation of the real-time metabolic intensity index of the embryos includes calculating the carbon dioxide mass release rate generated by the embryo population based on the monitored carbon dioxide concentration, the effective volume of the incubator cavity and the real-time ventilation, and obtaining the real-time metabolic intensity index based on the carbon dioxide mass release rate.

[0013] As a preferred embodiment of the incubator temperature and humidity adaptive adjustment method of the present invention, wherein: in step S1, calculating the carbon dioxide mass release rate produced by the embryo population includes:

[0014] According to the law of conservation of mass, the mass release rate of carbon dioxide produced by the embryo population in the incubator... The value consists of two parts: the change in the gas volume inside the chamber and the ventilation exhaust flow rate, and is calculated using the following formula:

[0015] ;

[0016] in:

[0017] The effective volume of the incubator's internal cavity;

[0018] The density of the air inside the incubator;

[0019] : Conversion factor, used to convert the volume concentration of carbon dioxide in ppm to a mass fraction;

[0020] : The current filtered carbon dioxide volume concentration;

[0021] : The volume concentration of carbon dioxide at the previous sampling time;

[0022] The time interval between two consecutive samples;

[0023] Real-time ventilation rate of the incubator at time t;

[0024] Background volume concentration of carbon dioxide in fresh air.

[0025] As a preferred embodiment of the incubator temperature and humidity adaptive adjustment method of the present invention, wherein: in step S2, the preset mapping relationship is a three-dimensional lookup table;

[0026] The three-dimensional lookup table uses the incubation age as the first dimension index and the discretized interval of the real-time metabolic intensity index as the second dimension index, and stores the target absolute humidity value.

[0027] The system uses bilinear interpolation to query the three-dimensional lookup table and dynamically generates the target absolute humidity value.

[0028] As a preferred embodiment of the adaptive temperature and humidity control method for incubators described in this invention, in step S3, the pre-established dynamic humidity prediction model for incubators is a discrete state-space model established through a system identification method.

[0029] Using control commands from humidifiers and dehumidifiers as input and predicted values ​​of absolute humidity inside the incubator as output, this method is used to predict the future trend of absolute humidity inside the incubator under different control commands.

[0030] As a preferred embodiment of the incubator temperature and humidity adaptive adjustment method of the present invention, wherein: in step S3, the rolling solution through the model predictive control algorithm includes:

[0031] In each control cycle, an optimization problem is constructed and solved;

[0032] The objective function of the optimization problem is configured as follows: minimize the tracking error between the predicted absolute humidity value and the target absolute humidity value in the future time period, while minimizing the variation of the control command sequence and energy consumption;

[0033] Solving the optimization problem also requires satisfying the physical operational constraints of the humidification and dehumidification devices.

[0034] As a preferred embodiment of the incubator temperature and humidity adaptive adjustment method described in this invention, step S4 includes:

[0035] The real-time control commands are parsed and converted into low-level control signals that drive the humidification and dehumidification devices.

[0036] Based on the underlying control signals, the humidification and dehumidification devices are driven to perform coordinated adjustment actions;

[0037] During execution, the operating status of the humidification and dehumidification devices is monitored, and safety protection logic is triggered when an abnormality is detected.

[0038] The change in the actual absolute humidity inside the incubator after execution is fed back to step S3 as the new current absolute humidity value to initiate the rolling solution for the next control cycle.

[0039] Secondly, embodiments of the present invention provide an adaptive temperature and humidity control system for an incubator, comprising:

[0040] Real-time monitoring module: Real-time monitoring of carbon dioxide concentration in the incubator, and calculation of real-time metabolic intensity indicators of embryos based on changes in carbon dioxide concentration;

[0041] Mapping query module: Based on the real-time metabolic intensity index and the current incubation age, query the preset mapping relationship and dynamically generate the corresponding target absolute humidity value in the incubator;

[0042] Humidity prediction module: Obtain the current absolute humidity value in the incubator, and combine it with the target absolute humidity value. Using a pre-established dynamic humidity prediction model for the incubator, the optimal control command sequence for the humidification and dehumidification devices in the future period is solved by the model prediction control algorithm.

[0043] The execution control module executes the immediate control commands in the optimal control command sequence to drive the humidification and dehumidification devices to adjust so that the actual absolute humidity inside the incubator approaches the target absolute humidity value.

[0044] The beneficial effects of this invention are as follows: This invention characterizes embryonic metabolic activity by changing carbon dioxide concentration within the incubator, calculates real-time metabolic intensity indicators, and generates a target absolute humidity value by querying a preset mapping relationship based on the incubation age. This transforms humidity setting from a static empirical curve to a dynamic setting driven by physiological feedback. At the control level, a dynamic humidity prediction model and a model predictive control algorithm are introduced to continuously solve for the optimal control command sequence of the humidification and dehumidification devices. Under the premise of satisfying physical operation constraints, this invention achieves forward-looking and stable tracking of the target absolute humidity, thereby improving humidity control accuracy, anti-disturbance capability, and energy efficiency, and improving the consistency of the incubation process. Attached Figure Description

[0045] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort, wherein:

[0046] Figure 1 A flowchart illustrating an adaptive temperature and humidity control method for an incubator, as provided in the first embodiment of the present invention.

[0047] Figure 2 The diagram shows the module connection of an adaptive temperature and humidity control system for an incubator, as provided in the third embodiment of the present invention. Detailed Implementation

[0048] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0049] Example 1, referring to Figure 1 As an embodiment of the present invention, an adaptive temperature and humidity adjustment method for an incubator is provided.

[0050] S1: Real-time monitoring of carbon dioxide concentration in the incubator, and calculation of real-time metabolic intensity indicators of embryos based on changes in carbon dioxide concentration.

[0051] The core purpose of this step is to transform the internal physiological activity of embryonic respiration and metabolism into a digital signal that can be acquired and quantified in real time by the control system, providing a direct basis for subsequent humidity regulation based on physiological needs.

[0052] To effectively monitor embryonic metabolic products, carbon dioxide gas sensors need to be installed inside the incubator. Furthermore, these sensors are preferably installed in the return air duct or at a representative location within the incubator where the gas is uniformly mixed, ensuring that the collected concentration values ​​reflect the overall metabolic output of the embryo population. Even further, the sensors should preferably be high-precision sensors based on the non-dispersive infrared (NDIR) principle, with a sampling frequency of at least once per minute, to continuously monitor and output the volumetric concentration of carbon dioxide in the incubator air. Its unit is ppm (parts per million by volume).

[0053] It should be noted that the raw concentration signal was directly collected. The data may contain high-frequency interference introduced by airflow disturbances, equipment electronic noise, etc., requiring preprocessing to ensure data quality. Furthermore, digital filtering algorithms (such as the moving average method or a first-order low-pass filter) are used to further filter the data. Processing is performed to suppress high-frequency noise, thereby obtaining a smooth and stable concentration time-series signal. .

[0054] Furthermore, the metabolic intensity of embryos is more fundamentally reflected in the mass of carbon dioxide released per unit time, rather than simply the environmental concentration. Therefore, concentration monitoring values ​​need to be converted into a mass release rate. Furthermore, according to the law of conservation of mass, the mass release rate of carbon dioxide produced by the embryo population within the incubator... (Unit: mg / h), consists of two parts: the change in gas volume inside the chamber and the ventilation exhaust flow rate, and is calculated using the following formula:

[0055] ;

[0056] in:

[0057] : Effective volume of the incubator cavity, in cubic meters (m³).

[0058] The density of the air inside the incubator, measured in kilograms per cubic meter (kg / m³), can be calculated based on the ideal gas law or by consulting a table of humid air density, using real-time monitored temperature and relative humidity. Alternatively, an empirical constant value (e.g., 1.2 kg / m³) can be used if the control accuracy requirements allow.

[0059] : Conversion factor, used to convert the volume concentration of carbon dioxide (ppm) to a mass fraction, with units of milligrams per kilogram per million parts per million (mg / (kg·ppm)). This factor can be derived from the ideal gas law. In practical applications, It can be approximated as a constant (e.g., approximately 1.96 mg / (kg·ppm) under standard conditions);

[0060] : The current filtered carbon dioxide volume concentration, in ppm;

[0061] : Volume concentration of carbon dioxide at the previous sampling time, in ppm;

[0062] The time interval between two consecutive samples, in hours (h).

[0063] Real-time ventilation volume of the incubator at time t, in cubic meters per hour (m³ / h); this value can be directly measured by an air volume sensor installed in the ventilation duct.

[0064] Background volume concentration of carbon dioxide in fresh air, expressed in ppm, typically 400 ppm.

[0065] It should be noted that the first term of this calculation formula This represents the rate of change in carbon dioxide levels within the incubator due to embryonic metabolism; the second term... This represents the net flow rate of carbon dioxide discharged outside the chamber through the ventilation system. Furthermore, this formula is used to calculate... It is a physical quantity that eliminates the interference of ventilation fluctuations and directly characterizes the mass of carbon dioxide produced by the embryo population per unit time. Compared with simply using concentration values, it can reflect the metabolic level more essentially and stably.

[0066] Furthermore, the total weight of eggs introduced into the incubation batches varies. To ensure the comparability of metabolic intensity indicators across batches and to directly correlate them with the needs of individual biological tissues, the release rate needs to be normalized. This requires further normalization of the carbon dioxide mass release rate. Divide by the total mass of the eggs in this batch. (Unit: kilograms) to obtain real-time metabolic intensity indicators :

[0067] ;

[0068] The The unit is milligrams per kilogram per hour (mg / (kg·h)), and its physical meaning is the mass of carbon dioxide produced per unit weight of egg per hour. It should be noted that this indicator... The overall metabolic activity of the embryo population was quantitatively and in real time characterized, and the output was used as a core physiological feedback variable to provide direct input for the humidity demand decision in step S2.

[0069] S2: Based on the real-time metabolic intensity index and the current incubation age, query the preset mapping relationship and dynamically generate the corresponding target absolute humidity value in the incubator.

[0070] This step combines the real-time metabolic intensity signal output from step S1 with the current incubation age, and dynamically calculates the target absolute humidity value of the incubator that best meets the current physiological needs of the embryo by querying the mapping relationship that has been pre-established and stored in the control system.

[0071] The preset mapping relationship defines a quantitative correspondence among the three factors—absolute humidity, environmental humidity, and the level of metabolic intensity required to maintain optimal water balance in an embryo at a specific incubation age and a specific metabolic intensity level—as well as the relative values ​​of these factors.

[0072] This mapping relationship is established based on the coupled physiological mechanism between embryonic respiration and metabolism and water evaporation, and is calibrated through experimental studies. Specifically, in a controlled incubation experimental device, at different incubation stages, a series of different embryonic metabolic level data were obtained by adjusting environmental conditions or monitoring natural embryonic activity; simultaneously, the absolute humidity of the environment was finely adjusted, and the absolute humidity value corresponding to the stable water loss rate of this batch of embryos within the preset optimal physiological range was observed and recorded. By collecting and analyzing experimental data covering the entire incubation cycle and the expected metabolic intensity range, a complete mapping relationship was finally constructed.

[0073] As a specific implementation of this pre-defined mapping relationship, it can be constructed as a three-dimensional lookup table. The lookup table uses the incubation age (e.g., an integer from 1 to 21) as the first-dimensional index and the discretized interval of the real-time metabolic intensity index (e.g., in intervals of 10 mg / (kg·h)) as the second-dimensional index. The value stored in each cell of the table is the corresponding target absolute humidity value (unit: g / m³).

[0074] This preset mapping relationship can also be represented as a mathematical function model fitted based on the experimental data mentioned above. This function takes the incubation age and real-time metabolic intensity index as input and directly outputs the target absolute humidity value.

[0075] The preset optimal physiological range refers to the range within which the cumulative water loss of the embryo through the eggshell during the entire incubation period reaches 12% to 15% of the initial egg weight. This is the generally accepted optimal water exchange range for avian embryonic development.

[0076] During the incubation process, the system receives real-time metabolic intensity indicators from step S1, as well as the current incubation age.

[0077] Furthermore, the system immediately uses both of the above as joint input conditions to query the preset mapping relationship. This query is a dynamic calculation process: if it is the aforementioned three-dimensional lookup table, the system uses a bilinear interpolation algorithm to query the table to obtain a continuous and smooth target value; if the pre-stored model is a fitted function model, it is obtained by substitution calculation.

[0078] It should be noted that the result generated by the above real-time query process is the dynamic target absolute humidity value. The dynamism of this value is reflected in two aspects: First, it exhibits a slow, phased change as the incubation period progresses, which follows the macroscopic physiological laws of the humidity requirements of the embryo at different developmental stages; second, and more importantly, it undergoes real-time fine-tuning with the fluctuations of the real-time metabolic intensity index at the minute level or even shorter cycles, reflecting a precise response to the instantaneous physiological changes of the embryo population.

[0079] It should also be noted that the target absolute humidity value generated in real time is used as the set value (or reference trajectory) for humidity control and is immediately output to the subsequent step S3.

[0080] This step transforms the humidity control target from a static preset value into a dynamic variable that responds to physiological states by querying the "metabolism-age-humidity" mapping relationship established based on embryonic fluid balance physiology. When metabolism is enhanced, a higher target humidity is automatically generated to suppress excessive water loss; when metabolism is weakened, a lower target humidity is generated. This achieves adaptive synchronization between humidity control and embryonic life activities from the decision-making source, effectively solving the problem of humidity supply and demand mismatch caused by individual differences and metabolic fluctuations.

[0081] S3: Obtain the current absolute humidity value in the incubator, and combine it with the target absolute humidity value. Using the pre-established incubator humidity dynamic prediction model, the optimal control command sequence for the humidification and dehumidification devices in the future period is solved by the model prediction control algorithm.

[0082] The purpose of this step is to use a mathematical model that can predict the dynamic changes in incubator humidity, combined with the current environmental state and the dynamic target generated in step S2, to calculate and output the optimal action sequence for driving the humidification and dehumidification devices in the future through an advanced algorithm called model predictive control, thereby achieving accurate, stable and efficient tracking of the target humidity.

[0083] S3.1 Acquisition of Current System Status: To enable effective prediction and control, the current status of the incubator environment must first be acquired. This includes real-time measurement of absolute humidity values ​​at multiple representative spatial locations within the incubator. These values ​​can be obtained through a calibrated absolute humidity sensor network and filtered before being used as the current system output status. Simultaneously, the current operating status (e.g., power, opening degree) of humidification and dehumidification devices (such as ventilation motors and spray valves) also needs to be acquired.

[0084] S3.2, Invocation of the incubator humidity dynamic prediction model: The control system has a pre-stored incubator humidity dynamic prediction model. This model is a simplified mathematical description of the physical system consisting of the incubator chamber, internal airflow, egg cluster, and humidification / dehumidification actuators. It can predict the trend of humidity changes within the incubator over a future period under a given control action.

[0085] Furthermore, the prediction model is preferably obtained through a system identification method. Specifically, during the incubator development or debugging phase, a series of known test signals (such as step signals or pseudo-random sequences) are applied to its humidification and dehumidification mechanisms, and the humidity response data inside the chamber is collected simultaneously. Based on this input-output data, an identification algorithm (such as the least squares method) is used to establish a discrete state-space model or transfer function model with control commands as input and humidity prediction values ​​as output. This model can characterize the key dynamic characteristics of the system, including the time inertia of the humidity response, the delay effect of control actions, and the coupling relationship between humidification and ventilation actions.

[0086] S3.3, Rolling optimization solution of the model predictive control algorithm: In each control cycle (e.g., every 30 seconds), the system performs a rolling optimization calculation as described below:

[0087] State initialization: The current absolute humidity value and equipment status obtained from S3.1 are used as the initial state for the prediction model calculation.

[0088] Input setting value: Use the target absolute humidity value generated in real time in step S2 as a reference trajectory to be tracked in the next predicted time domain (e.g., the next 30 minutes).

[0089] Problem Formulation: Within a finite prediction time domain, a constrained optimization problem is constructed. The objective function aims to minimize the sum of two terms: the first is the tracking error between the future humidity sequence output by the prediction model and the target reference trajectory; the second is the magnitude of change in the control command sequence itself or the energy consumption. For example, the objective function can be expressed as minimizing the weighted sum of the squares of the future humidity deviation and the sum of the squares of the changes in the control quantity.

[0090] Meanwhile, the optimization problem must include constraints, such as the humidifier power not exceeding its maximum value, the dehumidifier fan speed having upper and lower limits, and the two being logically interlocked (to avoid simultaneous full-power humidification and ventilation), as well as physical and safety constraints.

[0091] Sequence solution: In each control cycle, solve the optimization problem constructed above to obtain a set of optimal control command sequences that start from the current moment and cover the future control time domain (e.g., a series of humidifier power setpoints and fan speed setpoints).

[0092] Real-time command output: Extract the first control command from the optimal control command sequence and use it as the real-time control command that should be sent to the humidifier and dehumidifier immediately at the current moment.

[0093] It should be noted that the above process is performed "rolling". That is, in the next control cycle, the system will repeat the entire optimization calculation of S3.3 based on the latest measured actual humidity value (as the new initial state) and the updated target humidity value, thereby always using the latest information to continuously generate and execute new optimal control command sequences. This mechanism gives the control system strong disturbance rejection and real-time correction capabilities.

[0094] It should also be noted that step S3, by applying the Model Predictive Control (MPC) algorithm, successfully solved many problems in traditional incubator humidity control methods, such as regulation lag, overshoot oscillation, and excessive energy consumption. Traditional methods typically employ simple on / off control or PID feedback control, but in the environment of incubators with high inertia and strong coupling, problems such as inaccurate humidity control and excessive equipment operation easily occur. Through the MPC algorithm, the system can predict humidity change trends in advance and perform smooth control, avoiding the oscillation and lag phenomena of traditional control methods. MPC can also incorporate the physical and logical constraints of actuators into the optimization solution, intelligently coordinating the operation of humidification and dehumidification equipment, reducing ineffective resistance, thereby significantly reducing energy consumption and increasing equipment lifespan. Simultaneously, MPC's rolling optimization mechanism can quickly respond to external disturbances (such as door opening or seasonal changes), maintaining system stability and precise humidity regulation. In summary, step S3 achieves a transformation from traditional passive reactive control to feedforward-feedback coordinated active optimization control, providing incubators with a high-precision, high-efficiency humidity regulation solution.

[0095] S4: Execute the immediate control command in the optimal control command sequence to drive the humidification device and the dehumidification device to adjust.

[0096] The purpose of this step is to accurately and reliably convert the digital optimal control command obtained in step S3 into the actual physical actions of the humidification and dehumidification actuators, and to monitor and protect them during the execution process, thereby forming a complete adaptive control closed loop from "perception-decision-execution-feedback", so that the actual absolute humidity in the incubator dynamically and stably approaches the target value determined by physiological needs.

[0097] Furthermore, this execution and adjustment process is specifically implemented through the following sub-steps:

[0098] S4.1, Control Command Parsing and Distribution: The immediate control command from step S3 is a digital signal containing specific control parameters. The main controller of the control system parses this command and converts it into low-level control signals that can be directly driven by different actuators.

[0099] Furthermore, depending on the type of actuator, the underlying control signal can be:

[0100] For proportionally controlled humidifiers (such as ultrasonic or electrode humidifiers), the output is an analog signal (such as 0-10VDC) or a pulse width modulation (PWM) signal to linearly control its humidification power.

[0101] For ventilation and dehumidification devices (such as variable frequency fans or dampers), the output is an analog signal or a digital frequency command to adjust the fan speed or damper opening, thereby controlling the ventilation volume.

[0102] For switching actuators, the output is a relay on / off signal.

[0103] The analyzed control signals are synchronously and precisely sent to the designated humidification and dehumidification devices through the corresponding drive circuits or communication buses.

[0104] S4.2, Drive and Response of Actuators: After receiving their respective control signals, the humidifier and dehumidifier immediately change their working state according to the instructions.

[0105] The humidifier adjusts the amount of water atomization or steam generation based on the received power or flow command, precisely injecting water into the incubator cavity.

[0106] The dehumidification device (mainly achieved through ventilation) adjusts the rate at which dry outside air is introduced and humid air is expelled from the chamber according to the received air volume or opening command, thereby achieving controllable humidity removal.

[0107] It should be emphasized that, since the instruction sequence provided in step S3 is optimized, the actions of the humidifier and dehumidifier are pre-coordinated and mutually supportive, rather than independent switching actions. For example, the system may instruct the humidifier to operate continuously at medium power while simultaneously instructing the fan to run at low speed for fine-tuning, thereby achieving efficient and energy-saving precise humidity control and avoiding the frequent start-stop and conflicting operations of equipment common in traditional methods.

[0108] S4.3, Execution Status Monitoring and Safety Protection: During execution, the system continuously monitors key parameters to ensure safety and reliability. This includes monitoring the water level and operating current of the humidifier, as well as the operating feedback signals of the dehumidifier fan.

[0109] Furthermore, the system has pre-installed security protection logic:

[0110] Interlock protection: When the humidifier is detected to be short of water or malfunctioning, the humidification output will be automatically shut off and an alarm may be triggered, regardless of the control command. At the same time, the control strategy can be automatically switched to a safe mode that only regulates through ventilation.

[0111] Over-limit protection: If the actual humidity sensor reading deviates abnormally from the expected range after the command is executed, the system will determine that the execution is abnormal or the sensor is faulty, and trigger the predefined safety plan (such as maintaining the existing action, switching to backup control mode, etc.).

[0112] S4.4, Closed-loop feedback and continuous operation: The effect of this step, namely the change in the actual absolute humidity inside the incubator, will be immediately sensed by the humidity sensor network involved in steps S1 and S3, and fed back as the new "current absolute humidity value" to the next round of rolling optimization calculation in step S3.

[0113] Meanwhile, step S2 is also continuously generating new dynamic target values. Therefore, in the next control cycle (e.g., after tens of seconds), step S3 will recalculate a new set of optimal control commands based on the latest actual humidity feedback and the latest target humidity setting, and execute them again through step S4.

[0114] This process repeats itself, forming a closed-loop adaptive control system that enables real-time perception, dynamic decision-making, precise execution, and continuous correction, driving the actual ambient humidity to continuously approach and eventually stabilize within a very small fluctuation range around the target value for physiological needs.

[0115] It should be noted that the execution stage of traditional control methods is often disconnected from the decision-making stage, or the actuators are crude (such as simple switch control), resulting in poor control performance, high energy consumption, large equipment wear, and difficulty in forming a fast and accurate closed-loop response.

[0116] Step S4 of this invention translates the theoretically optimal instructions calculated by the MPC algorithm into coordinated physical actions using standardized industrial signals, ensuring the effective implementation of advanced control strategies. Driving the humidification and dehumidification devices to cooperate rather than act in opposition significantly improves regulation efficiency, reduces energy consumption, and extends equipment lifespan. Built-in status monitoring and safety interlocks ensure the long-term stability and safety of the system, preventing drastic environmental changes due to equipment failure.

[0117] Example 2 is the second embodiment of the present invention, which differs from the previous embodiment in that:

[0118] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art or the current technical solution, can be embodied in the form of a software product. This current computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0119] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0120] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0121] Example 3, referring to Figure 2 As an embodiment of the present invention, an adaptive temperature and humidity control system for an incubator is provided, which includes a real-time monitoring module, a mapping query module, a humidity prediction module, and an execution control module;

[0122] Real-time monitoring module: Real-time monitoring of carbon dioxide concentration in the incubator, and calculation of real-time metabolic intensity indicators of embryos based on changes in carbon dioxide concentration;

[0123] Mapping query module: Based on the real-time metabolic intensity index and the current incubation age, query the preset mapping relationship and dynamically generate the corresponding target absolute humidity value in the incubator;

[0124] Humidity prediction module: Obtain the current absolute humidity value in the incubator, and combine it with the target absolute humidity value. Using a pre-established dynamic humidity prediction model for the incubator, the optimal control command sequence for the humidification and dehumidification devices in the future period is solved by the model prediction control algorithm.

[0125] The execution control module executes the immediate control commands in the optimal control command sequence to drive the humidification and dehumidification devices to adjust so that the actual absolute humidity inside the incubator approaches the target absolute humidity value.

[0126] Example 4 is an embodiment of the present invention, which provides an adaptive temperature and humidity control method for an incubator. To verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculations and simulation / comparative experiments.

[0127] In this embodiment, fertilized eggs of the same strain were selected as the experimental subjects. Egg weight was sampled and statistically analyzed before incubation to control batch consistency. The sampling results showed that the initial egg weight was concentrated in the range of 65±2 g. Six incubators with identical structures were arranged in the same incubation workshop, designated as Invention Box A (Metabolic Feedback MPC-1 to 3) and Control Box B (Fixed Curve PID-1 to 3). The chamber volume, humidification device, dehumidification device (ventilation and dehumidification), air duct structure, and control interface of the six incubators were kept consistent. The temperature control adopted the same and consistent setting strategy to ensure that the humidity control strategy was the primary variable. The experimental period covered incubation up to day 18, including the period of significant increase in embryonic metabolism, to examine the control differences under dynamic physiological demand scenarios.

[0128] Before the experiment began, a "pre-defined mapping relationship" was established for the embryos of this strain, as required in step S2. The mapping relationship was constructed using a pre-calibration process: in a controlled incubation environment, multiple representative incubation age points were selected, and the metabolic levels of the embryo population were categorized and observed. Simultaneously, the absolute humidity was adjusted until the embryo's water loss stabilized near the pre-defined optimal physiological range. The corresponding age, metabolic intensity level, and absolute humidity combination were recorded to form a dataset covering the experimental age and metabolic range. After discretization, the dataset was organized into a three-dimensional lookup table. The lookup table used incubation age as the first-dimensional index and the discretized range of the real-time metabolic intensity index as the second-dimensional index. The stored value was the target absolute humidity value, pre-stored in the control system of the invention box A. During control operation, bilinear interpolation was used to achieve continuous target output.

[0129] During the trial operation phase, the invention chamber A strictly followed the sequence of steps S1 to S4 to form a closed loop: Step S1 continuously monitored the time-series data of carbon dioxide concentration in the incubator, combined with the effective volume of the incubator cavity and real-time ventilation, and obtained the carbon dioxide mass release rate of the embryo population according to the mass conservation principle, and obtained the real-time metabolic intensity index accordingly; Step S2 read the current incubation age, used the incubation age and the real-time metabolic intensity index as joint inputs, queried a pre-set three-dimensional lookup table and generated the target absolute humidity value through bilinear interpolation; Step S3 obtained the current absolute humidity value in the incubator and called the pre-established incubator humidity dynamic prediction model. (Through the discrete state-space model established by system identification), an optimization problem is constructed and solved within a fixed control cycle. The optimal control command sequence for the humidifier and dehumidifier is obtained through rolling solutions for future time periods. Step S4 executes the immediate control commands in the optimal control command sequence, parses and converts these commands into underlying control signals to drive the humidifier and dehumidifier, driving both devices to perform coordinated adjustment actions and monitoring their operating status. When an anomaly is detected, safety protection logic is triggered. The actual absolute humidity change after execution is fed back as the new current absolute humidity value to step S3 to enter the next control cycle. Control box B uses a fixed humidity curve and conventional PID control. The humidity setting only switches according to the pre-defined curve based on age, without introducing real-time metabolic feedback. The PID passively corrects deviations, and humidification and ventilation lack global collaborative optimization. During the experiment, a standardized door opening disturbance (uniform duration and frequency) is set to evaluate the disturbance recovery capabilities of the two methods. Simultaneously, humidity error statistics, overshoot, execution load, energy consumption, cumulative water loss rate, hatching rate, and healthy chick quality score are recorded to comprehensively verify the technical effectiveness.

[0130] The key data recorded is shown in Table 1:

[0131] Table 1: Experimental Data Recording Table

[0132]

[0133] From the perspective of the verifiability of the control chain, the index settings in this embodiment correspond to two core contributions of S2 and S3 respectively: First, S2 maps the real-time metabolic intensity index and the hatching age to the target absolute humidity value through a "preset mapping relationship," avoiding rigid deviations of the fixed curve from the source; Second, S3 uses a model predictive control algorithm to solve for the optimal control command sequence in a rolling manner, achieving lower errors, lower overshoot, and lower execution load in a humidity system with inertia and coupling. Data comparison can form a closed-loop evidence chain.

[0134] At the target setting level, the average target absolute humidity value for Invention Box A was 14.63–14.92 g / m³, while that for Control Box B was 13.73–13.79 g / m³. The two groups were similar in average real-time metabolic intensity (53.5–56.8 mg / (kg·h)), but the target absolute humidity value for Invention Box A shifted upwards overall and maintained reasonable dispersion across the three devices, indicating that the target was driven by a combination of incubation age and real-time metabolic intensity, rather than simply switching in segments with age. This difference aligns with the physiological pattern of increased water exchange demand due to enhanced metabolism in the later stages of incubation, demonstrating that the dynamic target generation mechanism of S2 can explicitly transform physiological state changes into a control reference trajectory, thereby avoiding the structural risk of "mismatch between setpoint and demand" in Control Box B under dynamic load. Further observation of the actual average absolute humidity values ​​showed that the three devices in Invention Box A had values ​​of 14.58–14.86 g / m³, closely adhering to the target average, indicating that the target was not only "reasonably generated" but also "trackable."

[0135] In terms of control accuracy and transient performance, the root mean square error (RMSE) of absolute humidity tracking for Inventive Box A is 0.38–0.47 g / m³, while that for Control Box B is 1.06–1.29 g / m³, showing a significant difference in error magnitude. Regarding peak overshoot, Inventive Box A is 0.72–0.86 g / m³, while Control Box B is 1.98–2.41 g / m³. This difference aligns with the typical advantages of model predictive control: rolling solutions explicitly consider the predicted absolute humidity evolution for future periods in each control cycle, enabling forward-looking planning of control actions and reducing overshoot and fluctuations caused by lag correction. Control Box B, employing passive PID control, is prone to forming an oscillating chain of "overcorrection-reverse correction" under disturbance and coupling conditions, leading to a simultaneous increase in RMSE and overshoot. Consistent with the key technical points of S3, Inventive Box A simultaneously considers tracking error, the amplitude of control command sequence changes, and energy consumption in the optimization problem, resulting in smoother control actions.

[0136] Regarding disturbance rejection capability, the time required to recover to the target ±0.5 g / m³ after a standardized door opening disturbance was 4.4–5.3 min for the invention box A and 11.2–13.5 min for the control box B. This result directly reflects the disturbance absorption capability of the "predictive model + rolling solution + execution feedback" closed loop: after a disturbance occurs, the next control cycle updates the state and resolves using the new current absolute humidity value, which can quickly provide a new coordinated control sequence; the PID of the control box B can only react based on instantaneous deviation, and there is a lack of global coordination between humidification and ventilation, resulting in a longer recovery process and a greater likelihood of overshoot.

[0137] In terms of execution load and energy consumption, the humidification / dehumidification operation frequency of the invention box A is 24–29 times / day, significantly lower than the 57–66 times / day of the control box B; the daily energy consumption is 3.55–3.71 kWh / day, lower than the 4.36–4.61 kWh / day of the control box B. The decrease in the number of operations can be reasonably explained as follows: the optimal control command sequence of S3 suppresses the amplitude of control command changes and reduces ineffective countermeasures during the solution process, and the execution stage tends to make continuous small adjustments rather than frequent start-stop; the reduction in energy consumption is consistent with the "reduction in operations + reduction in countermeasures", which is a verifiable engineering result.

[0138] Regarding hatching outcome indicators, the cumulative water loss rate of the invention box A on day 18 was 13.3–13.9%, concentrated near the preset optimal physiological range, while the control box B showed a dispersion of 10.9–11.2% which was lower and 15.0% which was higher, reflecting that the fixed curve was unable to stably maintain water exchange under different disturbances and load conditions. The hatching rate of the invention box A was 90.4–91.6%, and the chick health quality score was 91.9–93.4, while those of the control box B were 85.1–87.8% and 83.5–86.7, respectively. This result is not an isolated "result improvement", but is consistent with the logical chain of "dynamic generation of target absolute humidity value → reduction of humidity tracking error → stabilization of water loss rate → improvement of hatching outcome", thus supporting the technical effect of the invention in achieving high-precision, high-stability and high-efficiency humidity regulation under dynamic physiological demand scenarios.

[0139] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for adaptive temperature and humidity control in an incubator, characterized in that, include: S1: Real-time monitoring of carbon dioxide concentration in the incubator, and calculation of real-time metabolic intensity indicators of embryos based on changes in carbon dioxide concentration. S2: Based on the real-time metabolic intensity index and the current incubation age, query the preset mapping relationship and dynamically generate the corresponding target absolute humidity value in the incubator; S3: Obtain the current absolute humidity value in the incubator, and combine it with the target absolute humidity value. Using the pre-established incubator humidity dynamic prediction model, the optimal control command sequence of the humidification device and dehumidification device in the future period is solved by the model prediction control algorithm. S4: Execute the immediate control command in the optimal control command sequence to drive the humidification device and the dehumidification device to adjust; The calculation of the real-time metabolic intensity index of the embryos includes calculating the carbon dioxide mass release rate of the embryo population based on the monitored carbon dioxide concentration, the effective volume of the incubator cavity and the real-time ventilation, and obtaining the real-time metabolic intensity index based on the carbon dioxide mass release rate. The calculation of the carbon dioxide mass release rate produced by the embryo population includes: According to the law of conservation of mass, the mass release rate of carbon dioxide produced by the embryo population in the incubator... The value consists of two parts: the change in the gas volume inside the chamber and the ventilation exhaust flow rate, and is calculated using the following formula: ； in: : The effective volume of the incubator's interior, in cubic meters; The density of air inside an incubator, expressed in kilograms per cubic meter. : Conversion factor, used to convert the volume concentration of carbon dioxide in ppm to a mass fraction, with units of milligrams per kilogram per million parts per million; : The current filtered carbon dioxide volume concentration, in ppm; : Volume concentration of carbon dioxide at the previous sampling time, in ppm; The time interval between two consecutive samplings, in hours; : Real-time ventilation volume of the incubator at time t, in cubic meters per hour; Background volume concentration of carbon dioxide in fresh air, in ppm; Divide the carbon dioxide mass release rate by the total mass of the eggs in this batch to obtain the real-time metabolic intensity index.

2. The incubator temperature and humidity adaptive adjustment method as described in claim 1, characterized in that, In step S2, the preset mapping relationship is a three-dimensional lookup table; The three-dimensional lookup table uses the incubation age as the first dimension index and the discretized interval of the real-time metabolic intensity index as the second dimension index, and stores the target absolute humidity value. The system uses bilinear interpolation to query the three-dimensional lookup table and dynamically generates the target absolute humidity value.

3. The incubator temperature and humidity adaptive adjustment method as described in claim 2, characterized in that, In step S3, the pre-established incubator humidity dynamic prediction model is a discrete state-space model established through the system identification method. Using control commands from humidifiers and dehumidifiers as input and predicted values ​​of absolute humidity inside the incubator as output, this method is used to predict the future trend of absolute humidity inside the incubator under different control commands.

4. The incubator temperature and humidity adaptive adjustment method as described in claim 3, characterized in that, In step S3, the rolling solution using the model predictive control algorithm includes: In each control cycle, an optimization problem is constructed and solved; The objective function of the optimization problem is configured as follows: minimize the tracking error between the predicted absolute humidity value and the target absolute humidity value in the future time period, while minimizing the variation of the control command sequence and energy consumption; Solving the optimization problem also requires satisfying the physical operational constraints of the humidification and dehumidification devices.

5. The incubator temperature and humidity adaptive adjustment method as described in claim 4, characterized in that, Step S4 includes: The real-time control commands are parsed and converted into low-level control signals that drive the humidification and dehumidification devices. Based on the underlying control signals, the humidification and dehumidification devices are driven to perform coordinated adjustment actions; During execution, the operating status of the humidification and dehumidification devices is monitored, and safety protection logic is triggered when an abnormality is detected. The change in the actual absolute humidity inside the incubator after execution is fed back to step S3 as the new current absolute humidity value to initiate the rolling solution for the next control cycle.

6. An adaptive temperature and humidity control system for an incubator, used to implement the adaptive temperature and humidity control method for an incubator as described in any one of claims 1 to 5, characterized in that, include: Real-time monitoring module: Real-time monitoring of carbon dioxide concentration in the incubator, and calculation of real-time metabolic intensity indicators of embryos based on changes in carbon dioxide concentration; Mapping query module: Based on the real-time metabolic intensity index and the current incubation age, query the preset mapping relationship and dynamically generate the corresponding target absolute humidity value in the incubator; Humidity prediction module: Obtain the current absolute humidity value in the incubator, and combine it with the target absolute humidity value. Using a pre-established dynamic humidity prediction model for the incubator, the optimal control command sequence for the humidification and dehumidification devices in the future period is solved by the model prediction control algorithm. The execution control module executes the immediate control commands in the optimal control command sequence to drive the humidification and dehumidification devices to adjust so that the actual absolute humidity inside the incubator approaches the target absolute humidity value.

Images