Intelligent feedback high-precision incubator carbon dioxide constant temperature shaker full-automatic system

Abstract

This invention discloses a fully automated intelligent feedback-type high-precision incubator-based carbon dioxide constant-temperature shaker system, comprising: a fuzzy neural coupling compensation control module that integrates temperature field prediction data and metabolic parameters, and sends control commands to other modules; a probabilistic temperature field uniformity prediction module that generates temperature field assessment data; an incubator gradient fusion oscillation control module that adjusts the oscillation frequency and gradient distribution; a respiratory metabolic parameter quantification module that converts gas composition data into metabolic parameters; and other modules that respectively regulate carbon dioxide concentration and shaker status to achieve fully automated control of the entire process. This system uses the fuzzy neural coupling compensation control module as its core, linking data and commands from multiple modules to solve the problems of insufficient control coordination and parameter adjustment lag; and through the probabilistic temperature field uniformity prediction module combined with spatial gradient and metabolic dynamics, and the respiratory metabolic parameter quantification module processing data throughout the entire process, it solves the problem of limited accuracy in temperature field prediction and metabolic parameter quantification, meeting the requirements of high-precision culture.

Description

Technical Field

[0001] This invention relates to the field of biological sample culture technology, and in particular to a fully automatic intelligent feedback high-precision incubator carbon dioxide constant temperature shaker system. Background Technology

[0002] In biological sample culture, the demands for temperature stability, accurate carbon dioxide concentration, uniform shaking of the culture environment, and monitoring of sample respiratory and metabolic states are becoming increasingly stringent. Traditional incubators and shakers are mostly controlled independently, making it difficult to achieve coordinated regulation of temperature, carbon dioxide concentration, shaking parameters, and sample metabolic states. Furthermore, the numerous manual interventions involved can easily lead to fluctuations in the culture environment, affecting sample culture results. With the increasing precision requirements of biological experiments, there is an urgent need for a fully automated system that integrates multi-parameter monitoring and intelligent feedback control. Through the collaborative work of multiple modules, this system can capture real-time data on temperature distribution, gas concentration changes, and sample metabolism within the incubator, dynamically adjusting the shaking frequency, temperature, and carbon dioxide supply to adapt to the culture needs of different samples, reduce human error, and improve the stability and controllability of the culture process.

[0003] Existing technologies for CO2 constant-temperature shaker systems in incubators have two significant drawbacks: First, there is insufficient coordination between control modules. Most systems can only achieve independent adjustment of a single parameter, failing to deeply couple temperature field prediction results, sample metabolic data, oscillation control, and concentration regulation. This results in lag in parameter adjustments, making it difficult to respond quickly to changes in the culture environment and thus affecting the overall stability of the culture environment. Second, the accuracy of temperature field uniformity prediction and metabolic parameter quantification is limited. Existing prediction methods mostly rely on fixed models and do not fully consider the impact of spatial gradient differences within the incubator and dynamic changes in sample metabolism on the temperature field distribution. At the same time, metabolic parameter acquisition and conversion are easily affected by interference, failing to accurately reflect the true metabolic state of the sample. This results in a lack of reliable data support for the generation of control commands, making it difficult to meet the requirements of high-precision culture. Summary of the Invention

[0004] In order to overcome the shortcomings and deficiencies of existing technologies, this invention provides a fully automatic intelligent feedback high-precision incubator carbon dioxide constant temperature shaker system.

[0005] The technical solution adopted in this invention is an intelligent feedback high-precision incubator carbon dioxide constant temperature shaker fully automatic system, including: a fuzzy neural coupling compensation control module, a probabilistic temperature field uniformity prediction module, an incubator gradient fusion oscillation control module, a respiratory metabolic parameter quantification module, a carbon dioxide concentration regulation module, and a constant temperature shaker drive module. The fuzzy neural coupling compensation control module receives temperature distribution data from the probabilistic temperature uniformity prediction module and metabolic parameter signals from the respiratory metabolic parameter quantification module. After internal coupling compensation calculation, it sends control commands to the incubator gradient fusion oscillation control module, the carbon dioxide concentration regulation module, and the constant temperature shaker drive module. The probabilistic temperature uniformity prediction module collects temperature detection values ​​from different areas inside the incubator, generates temperature uniformity assessment data through a probabilistic prediction algorithm, and transmits it to the fuzzy neural coupling compensation control module. The incubator gradient fusion oscillation control module receives commands from the fuzzy neural coupling compensation control module and adjusts the oscillation frequency and gradient distribution inside the incubator based on its own gradient oscillation parameters. The respiratory metabolic parameter quantification module collects data on the gas composition and metabolic product concentrations produced by the respiratory metabolism of samples inside the incubator, converts them into quantified parameters, and transmits them to the fuzzy neural coupling compensation control module. The carbon dioxide concentration regulation module receives commands from the fuzzy neural coupling compensation control module and adjusts the carbon dioxide supply and flow rate to maintain a stable carbon dioxide concentration inside the incubator. The constant temperature shaker drive module receives commands from the fuzzy neural coupling compensation control module and controls the shaking amplitude, frequency, and working status of the constant temperature heating element, thus coordinating the control of the shaker temperature and motion state.

[0006] Furthermore, the coupling compensation control model expression adopted by the fuzzy neural coupling compensation control module is as follows: ; In the formula, The output is controlled by fuzzy neural coupling compensation. These are the weight coefficients of the neural fuzzy network. It is the Sigmoid activation function. This is the predicted temperature value output by the probabilistic temperature field uniformity prediction module. For metabolic quantification parameters, For network bias terms, This is the integral coefficient for temperature deviation. For time variables, Set the temperature value for the incubator. This represents the coefficient of change in metabolic parameters. The time rate of change of metabolic quantification parameters; the temperature field prediction model expression used in the probabilistic temperature field uniformity prediction module is: ; In the formula, This represents the probability value of temperature field uniformity. The variance of the temperature readings inside the incubator. For the first The measured values ​​at each temperature detection point This is the average value of all temperature measurement points. This refers to the allowable temperature fluctuation range. This refers to the number of detection points whose temperature readings fall within the set temperature fluctuation range. This represents the total number of temperature detection points.

[0007] Furthermore, the temperature field dynamic prediction model expression used by the probabilistic temperature field uniformity prediction module is as follows: ; In the formula, for Predicted temperature value at time of day for Predicted temperature value at time of day To predict the correction factor, for Average temperature measurement at time [time]. The spatial gradient influence coefficient. This refers to the number of temperature detection points. for Time of the first Temperature measurements at each detection point For the first The distance from each detection point to the center of the incubator; the metabolic coupling regulation model expression used by the fuzzy neural coupling compensation control module is: ; In the formula, This is the metabolic coupling regulatory coefficient. For adjustment coefficients, The hyperbolic tangent activation function is used. This is the carbon dioxide concentration value detected by the carbon dioxide concentration control module. This represents the rate of change of carbon dioxide concentration over time.

[0008] Furthermore, the gradient oscillation control model expression used by the gradient fusion oscillation control module of the incubator is as follows: ; In the formula, The output value is the oscillation frequency. Based on the oscillation frequency, The coefficient representing the influence of temperature field uniformity. This refers to the probability value of temperature field uniformity output by the probabilistic temperature field uniformity prediction module. To compensate for the control influence coefficient, The output is controlled by fuzzy neural coupling compensation. To compensate for the maximum limit of the control output, The oscillation angular frequency, For time variables, The oscillation phase angle is used; the temperature-oscillation cooperative model expression used by the fuzzy neural coupling compensation control module is: ; In the formula, The target value for temperature control, To set the temperature value, To compensate for the control deviation coefficient, To compensate for the average value of the control output, This is the oscillation frequency deviation coefficient. This represents the average oscillation frequency.

[0009] Furthermore, the metabolic quantification model expression used by the respiratory metabolic parameter quantification module is as follows:

[0010] In the formula, For metabolic quantification parameters, For the number of gas samples, For the first Oxygen concentration in the second sample. For the first The carbon dioxide concentration of the second sample. For the first Gas flow rate at the time of the next sampling The sampling time interval, For sample quality; the gradient distribution model expression used by the gradient fusion oscillation control module of the incubator is: ; Coordinates inside the incubator gradient distribution value at, Based on the gradient value, This is the temperature gradient coefficient. coordinates The predicted temperature value at that location, This is the average predicted temperature value. To predict the standard deviation of temperature, The spatial attenuation coefficient, These are the spatial coordinates within the incubator.

[0011] Furthermore, the concentration control model expression used by the carbon dioxide concentration regulation module is as follows: ; Supply flow for carbon dioxide. Basic supply flow, For control coefficients, Set the concentration for carbon dioxide. To measure the concentration of carbon dioxide, For time variables, For metabolic quantification parameters; the temperature-oscillation cooperative model expression used in the isothermal shaker drive module is: ; In the formula, The oscillation speed of the shaking table, Based on the swing speed, This is the temperature deviation coefficient. The target value for temperature control, This is the temperature measurement value of the shaker. The oscillation frequency coefficient is... This is the output value for the oscillation frequency.

[0012] Furthermore, the incubator gradient fusion oscillation control module includes an oscillation frequency adjustment unit, a gradient distribution control unit, a multi-region collaborative unit, and a feedback calibration unit. The oscillation frequency adjustment unit receives control commands from the fuzzy neural coupling compensation control module, acquires the current oscillation frequency signal of the shaker, compares the deviation between the command frequency and the current frequency using an internal frequency adjustment algorithm, generates a frequency adjustment signal, and transmits it to the shaker drive component to adjust the motor speed and transmission ratio of the drive component to change the oscillation frequency. The gradient distribution control unit acquires temperature and gas concentration distribution data in different regions of the incubator, combines it with the frequency signal output by the oscillation frequency adjustment unit, determines the gradient adjustment target for each region using a gradient calculation algorithm, generates a gradient control signal, and transmits it to the airflow guide in the incubator. The device and heating element are modified by changing the airflow direction, velocity, and power distribution of the heating element to adjust the gradient distribution. The multi-region collaborative unit receives the output signals from the oscillation frequency adjustment unit and the gradient distribution control unit, collects the state feedback data of each region in the incubator, coordinates the oscillation and gradient parameters of different regions through a collaborative algorithm to avoid parameter conflicts between regions, generates collaborative control signals, and feeds them back to the oscillation frequency adjustment unit and the gradient distribution control unit respectively. The feedback calibration unit collects the output parameters of the gradient fusion oscillation control module of the incubator and the actual detected oscillation and gradient state data, calculates the parameter deviation value, generates a calibration signal through a calibration algorithm, and transmits it to the oscillation frequency adjustment unit, the gradient distribution control unit, and the multi-region collaborative unit to correct the output parameters of each unit to reduce the deviation.

[0013] Furthermore, the respiratory metabolic parameter quantification module includes a gas sampling unit, a component analysis unit, a parameter conversion unit, and a data filtering unit. The gas sampling unit receives a system start signal or a timed sampling command, controls the sampling pump and valves to extract a quantitative gas sample from different sample areas in the incubator, and delivers the gas sample to the component analysis unit through a sampling pipeline, while simultaneously recording the sampling time, sampling location, and sampling volume data. The component analysis unit receives the gas sample delivered by the gas sampling unit, activates its internal gas sensor and detection circuit, detects the concentrations of oxygen, carbon dioxide, and metabolic product gases in the sample, and converts the detected analog signals into digital signals. The signal is generated to produce concentration detection data and transmitted to the parameter conversion unit. The parameter conversion unit receives the concentration detection data output by the component analysis unit, combines it with the sampling parameters recorded by the gas sampling unit, and converts the concentration data into quantitative parameters reflecting the respiratory metabolic intensity of the sample through a metabolic quantification algorithm. At the same time, it calculates the time change rate of the parameters, generates metabolic quantification parameters and change rate data, and transmits them to the data filtering unit. The data filtering unit receives the metabolic quantification parameters and change rate data output by the parameter conversion unit, uses a filtering algorithm to remove noise interference signals in the data, retains the effective parameter change trend, generates a filtered metabolic parameter signal, and transmits it to the fuzzy neural coupling compensation control module.

[0014] Furthermore, the carbon dioxide concentration control module includes a concentration detection unit, a supply control unit, a flow regulation unit, and a leakage monitoring unit. The concentration detection unit collects carbon dioxide concentration signals from different locations within the incubator, converts the analog concentration signals into digital signals via a detection circuit, performs mean calculation and stability assessment on the digital signals, generates carbon dioxide concentration detection data, and transmits it to the supply control unit and the leakage monitoring unit. The supply control unit receives control commands from the fuzzy neural coupling compensation control module and the concentration detection data from the concentration detection unit, compares the difference between the commanded concentration and the detected concentration, calculates the required carbon dioxide supply amount using a supply algorithm, generates a supply control signal, and transmits it to the carbon dioxide control unit. The supply device controls the valve opening and gas supply pressure to adjust the supply volume; the flow regulation unit receives the supply control signal from the supply control unit, collects airflow speed and pressure data in the incubator, determines the airflow circulation rate and flow path through the flow algorithm, generates a flow control signal and transmits it to the fan and airflow guide component in the incubator, and changes the fan speed and guide component angle to adjust the airflow state; the leakage monitoring unit receives the concentration detection data from the concentration detection unit and the supply volume data from the supply control unit, calculates the deviation between the theoretical concentration change value and the actual concentration change value, determines whether there is a carbon dioxide leak through the leakage judgment algorithm, generates a leakage monitoring signal and transmits it to the supply control unit.

[0015] The intelligent feedback high-precision incubator CO2 constant temperature shaker fully automatic system operates as follows: Step S1, the system modules are started, enabling the fuzzy neural coupling compensation control module, probabilistic temperature field uniformity prediction module, incubator gradient fusion oscillation control module, respiratory metabolic parameter quantification module, CO2 concentration regulation module, and constant temperature shaker drive module to enter the initial working state. Each module completes internal parameter initialization and self-check, and sends an initial state signal to the fuzzy neural coupling compensation control module; Step S2, the probabilistic temperature field uniformity prediction module starts collecting temperature detection values ​​from different areas inside the incubator, organizes and converts the collected temperature data, inputs it into the internal probabilistic prediction algorithm for calculation, generates temperature field uniformity evaluation data, and transmits the evaluation data to the fuzzy neural coupling compensation control module; Step S3, the respiratory metabolic parameter quantification module starts the gas sampling unit, extracts gas samples from the incubator, and sends them to the component analysis unit for gas component detection. The detected concentration data is transmitted to the parameter conversion unit to convert it into metabolic quantification parameters, filtered by the data filtering unit, and then transmitted to the fuzzy neural coupling compensation control module. The coupling compensation control module; Step S4, the fuzzy neural coupling compensation control module receives temperature field assessment data from the probabilistic temperature field uniformity prediction module and metabolic parameter signals from the respiratory metabolic parameter quantification module, and combines them with the target parameters set by the system, and performs calculations through the internal coupling compensation algorithm to generate control commands for the incubator gradient fusion oscillation control module, carbon dioxide concentration regulation module, and constant temperature shaker drive module; Step S5, the incubator gradient fusion oscillation control module, carbon dioxide concentration regulation module, and constant temperature shaker drive module respectively receive the control commands from the fuzzy neural coupling compensation control module, and, combined with the current working status and detection data of their respective modules, adjust the oscillation frequency, gradient distribution, carbon dioxide supply, shaker motion state, and temperature parameters through the internal control algorithm; Step S6, each module feeds back the adjusted working parameters and the actual detected state data to the fuzzy neural coupling compensation control module, the fuzzy neural coupling compensation control module compares the deviation between the feedback data and the set target parameters, performs coupling compensation calculations again, generates new control commands and sends them to each execution module to continuously adjust and stabilize the system parameters.

[0016] Beneficial Effects: This invention proposes a fully automatic intelligent feedback-type high-precision incubator CO2 constant-temperature shaker system. Through the collaborative operation of six modules, including a fuzzy neural coupling compensation control module and a probabilistic temperature field uniformity prediction module, it can capture real-time temperature field distribution, gas concentration, and sample metabolic data within the incubator. It dynamically adjusts the oscillation frequency, temperature, and CO2 supply, effectively improving the stability and controllability of the culture environment, reducing human error, and adapting to different sample culture needs. Its beneficial effects lie in the fact that the fuzzy neural coupling compensation control module deeply couples temperature field prediction results, metabolic data, oscillation control, and concentration regulation, solving the problems of insufficient coordination of control modules and lag in parameter adjustment in existing technologies. This enables rapid response of various parameters to changes in the culture environment, ensuring overall stability. The probabilistic temperature field uniformity prediction module fully considers the impact of spatial gradient differences and dynamic changes in sample metabolism on the temperature field within the incubator. Combined with the respiratory metabolic parameter quantification module, it accurately collects and converts metabolic parameters, reducing interference. This solves the problem of limited accuracy in temperature field prediction and metabolic parameter quantification in existing technologies, providing reliable data support for control command generation and meeting the needs of high-precision culture. Attached Figure Description

[0017] Figure 1 This is a diagram showing the system module composition of the present invention; Figure 2 This is a flowchart of the system operation steps of the present invention. Detailed Implementation

[0018] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0019] like Figure 1 As shown, the intelligent feedback high-precision incubator carbon dioxide constant temperature shaker fully automatic system includes: a fuzzy neural coupling compensation control module, a probabilistic temperature field uniformity prediction module, an incubator gradient fusion oscillation control module, a respiratory metabolic parameter quantification module, a carbon dioxide concentration regulation module, and a constant temperature shaker drive module. The fuzzy neural coupling compensation control module receives the temperature field distribution data output by the probabilistic temperature field uniformity prediction module and the metabolic parameter signal output by the respiratory metabolic parameter quantification module. After internal coupling compensation calculation, it sends control commands to the incubator gradient fusion oscillation control module, the carbon dioxide concentration regulation module, and the constant temperature shaker drive module. Specifically, the fuzzy neural coupling compensation control module receives temperature distribution data output from the probabilistic temperature field uniformity prediction module and metabolic parameter signals generated by the respiratory metabolic parameter quantification module. It then performs coupled calculations using a computational mechanism that integrates fuzzy logic nonlinear processing capabilities with the self-learning characteristics of neural networks. This generates control commands with multi-parameter coordination, which are precisely sent to the incubator gradient fusion oscillation control module, the carbon dioxide concentration regulation module, and the constant-temperature shaker drive module. This overcomes the limitations of independent control by each execution module, achieving coordinated adjustment of parameters such as temperature, oscillation, and gas concentration, avoiding parameter conflicts, and ensuring overall stability of the culture environment. In practice, the module is set to a data sampling interval of 10 seconds, receiving 30 sets of data at a time, including those from the incubator and other modules. The temperature field distribution data of the same area and 15 sets of metabolic parameter signals reflecting the metabolic state of the samples are strictly controlled within a 5-second internal calculation cycle. During the calculation, the control rule weights in the fuzzy logic rule base and the connection weights of the neural network are dynamically adjusted according to the culture stage. The generated control commands include key parameters such as oscillation frequency adjustment amplitude (range ±5 rpm), carbon dioxide supply (range 0-2 liters / minute), and shaker temperature target value (range 5-50℃). The commands are transmitted through industrial Ethernet with a delay of less than 2 seconds. At the same time, the module adopts a low-power design with an operating power of no more than 20 watts to avoid affecting the power supply stability of other modules in the system due to excessive energy consumption, and to ensure the continuous output of accurate control commands during long-term continuous operation.

[0020] The probabilistic temperature field uniformity prediction module collects temperature detection values ​​from different areas inside the incubator, generates temperature field uniformity assessment data through a probabilistic prediction algorithm, and transmits it to the fuzzy neural coupling compensation control module. Specifically, the probabilistic temperature uniformity prediction module collects temperature data from different areas within the incubator and generates temperature uniformity assessment results. This provides a reliable environmental condition basis for the core control module, capturing temperature distribution differences in real time and preventing uneven sample culture results caused by local temperature deviations (exceeding ±0.3℃). In practice, 24 temperature detection points are arranged inside the incubator in three layers (top, middle, and bottom, each layer spaced 20 cm apart) and six directions (left, right, front, and back, each direction spaced 30 cm apart). Each detection point is equipped with a platinum resistance temperature sensor with an accuracy of ±0.1℃. The sampling frequency is set to once per second. The collected data is first processed by software to remove outliers (data that deviates from the average value by more than ±0.5℃). After retaining the valid data, the probability value of temperature field uniformity is calculated by a probability statistical model. When the probability value is lower than 90%, the module immediately sends an early warning signal to the core control module. The power consumption of the module is controlled within 15 watts during operation, and the sensor zero-point calibration program is started every 2 hours. A 0℃ standard constant temperature bath is used as the calibration benchmark to ensure that the detection error of the sensor does not exceed ±0.05℃ after long-term use, providing accurate data support for subsequent control adjustments.

[0021] The gradient fusion oscillation control module of the incubator receives instructions from the fuzzy neural coupling compensation control module and adjusts the oscillation frequency and gradient distribution inside the incubator in combination with its own gradient oscillation parameters. Specifically, the gradient fusion oscillation control module of the incubator performs dual functions of oscillation regulation and temperature gradient control within the incubator. After receiving control commands from the fuzzy neural coupling compensation control module, it coordinates the oscillation mechanism and gradient control elements based on pre-stored basic parameters such as the oscillation frequency range (10-150 rpm) and gradient adjustment threshold (0-5℃ / m). This promotes material exchange within the incubator by adjusting the oscillation state and ensures uniform temperature across all areas by controlling the temperature gradient, meeting the dynamic culture requirements of the samples. In practice, the oscillation mechanism is driven by a stepper motor with a step angle of 1.8°. Microstepping technology achieves a speed adjustment accuracy of 1 rpm. Gradient control is achieved through independent control of eight zones within the incubator. The heating element is made of ceramic and has a power adjustment accuracy of 5 watts (range 20-100 watts). During operation, the noise level is controlled below 55 decibels through a soundproof enclosure to avoid interfering with sample growth. Every 30 seconds, the current oscillation frequency (error ±0.5 rpm) and gradient distribution data (error ±0.2℃ / m) are fed back to the core control module via RS485 bus. The transmission components of the oscillation mechanism are made of 304 stainless steel with a hardened surface, a wear resistance coefficient ≤0.15, and a designed service life ≥10,000 hours. The thermal response time of the gradient control element is ≤3 seconds. The module is also equipped with an overload protection device that automatically cuts off the power supply when the detected operating current exceeds 10A to ensure safe operation.

[0022] The respiratory metabolism parameter quantification module collects data on the gas composition and metabolite concentration of samples in the incubator, converts them into quantitative parameters, and then transmits them to the fuzzy neural coupling compensation control module. Specifically, the respiratory metabolic parameter quantification module collects data on oxygen consumption rate, carbon dioxide generation rate, and metabolic product concentration from the samples in the incubator. This data is then converted into quantifiable metabolic parameters using specific data conversion logic. This provides sample status feedback to the core control module, ensuring that control commands accurately match the sample's metabolic needs and preventing culture effects from being affected by mismatches between environmental parameters and metabolic status. In practice, the module uses a needle-type polytetrafluoroethylene (PTFE) sampling probe, inserted 2-3 cm above the sample in the incubator (ensuring the collected gas reflects the sample's real-time metabolic status). The sampling frequency is set to once every 2 minutes, with a single sample volume of 5 ml (error ±0.1 ml). The sampled gas passes through a PTFE tube with an inner diameter of 2 mm and a length ≤1 m. The data is transmitted through PTFE tubing to the detection unit. Oxygen and carbon dioxide concentrations are analyzed by a gas chromatograph (detection accuracy ±0.01%), and metabolite concentrations are detected by an ultraviolet spectrophotometer (detection wavelength 254nm, accuracy ±0.001 g / L). The concentration data is then fed into a preset conversion logic to generate two quantitative parameters: metabolic intensity (range 0-100 units) and metabolic change rate (range ±5 units / hour). The parameters are transmitted via Modbus protocol at intervals consistent with the sampling intervals. The module is calibrated every 24 hours using a 5% carbon dioxide standard gas (error ±0.02%) and a 21% oxygen standard gas (error ±0.02%) as calibration benchmarks to ensure detection and conversion accuracy.

[0023] The carbon dioxide concentration regulation module receives instructions from the fuzzy neural coupling compensation control module to adjust the carbon dioxide supply and flow rate in order to maintain a stable carbon dioxide concentration in the incubator. Specifically, after receiving instructions from the fuzzy neural coupling compensation control module, the carbon dioxide concentration control module precisely controls the amount and speed of carbon dioxide entering the incubator by adjusting the valve opening and gas flow rate of the carbon dioxide supply device, stabilizing the carbon dioxide concentration in the incubator within a set range (typically 3-10%) to meet the specific gas environment requirements of different samples. In practice, the module connects to a 10-liter carbon dioxide cylinder (working pressure 0.5-1.2 MPa), uses an electronic proportional valve to control the gas supply, with a valve opening adjustment range of 0-100% and an adjustment accuracy of 1%. The matching gas flow meter has a measurement range of 0-5 liters / minute and an accuracy of ±0.01 liters / minute. The system continuously collects the carbon dioxide concentration inside the chamber (detected by an infrared sensor with an accuracy of ±0.1%). When the concentration deviates from the set value by more than 0.1%, the adjustment program is immediately activated. The gas supply is changed by adjusting the valve opening to bring the concentration back to the set value. The adjustment response time is ≤10 seconds. The module adopts a double-sealed structure design, and the sealing material of the valve and pipeline interface is nitrile rubber (carbon dioxide corrosion resistance grade ≥V0). The gas leakage rate is less than 0.001 liters / minute. It is also equipped with a diffused silicon pressure sensor to monitor the cylinder pressure in real time. When the pressure is lower than 0.3MPa, the system alarm module sends a low-pressure warning signal to remind the staff to replace the cylinder, ensuring a continuous and stable gas supply.

[0024] The constant temperature shaker drive module receives instructions from the fuzzy neural coupling compensation control module to control the shaking amplitude, frequency, and working status of the constant temperature heating element, thereby coordinating the control of the shaker temperature and motion state.

[0025] Specifically, the constant-temperature shaker drive module is responsible for the shaker's temperature control and motion state adjustment. After receiving instructions from the core control module, it maintains a stable shaker temperature by adjusting the power of the constant-temperature heating element, and adjusts the shaker's oscillation amplitude and frequency by controlling the drive motor's speed and direction. This provides a constant temperature environment and suitable dynamic oscillation conditions for the samples, balancing temperature stability and motion requirements. In practice, the constant-temperature heating element uses a 304 stainless steel heating tube with a heating power range of 50-500 watts. The power is adjusted using a PID control algorithm, with a temperature control range of 5-50℃, a control accuracy of ±0.1℃, and a temperature fluctuation range of ≤±0.05℃. The shaker drive uses a brushless DC motor with a rated speed of 300 rpm, controlled by pulse width modulation (PWM) technology. The speed is adjustable from 0-300 rpm, corresponding to a shaking amplitude of 0-50 mm (amplitude adjustment accuracy 1 mm). The module collects the shaking table surface temperature (detected by a patch-type temperature sensor, error ±0.05℃) and motor speed (detected by an encoder, error ±0.1 rpm) every 15 seconds and feeds the data back to the core control module. The heating element surface is polished, with a temperature uniformity error ≤ ±0.2℃ and a motor operating noise ≤ 50 decibels. The shaking table is made of 6061 aluminum alloy, with a surface flatness error ≤ 0.1 mm. The module is also equipped with overheat protection (cuts off the heating power when the temperature exceeds the set value by 5℃) and motor overload protection (stops the motor when the torque exceeds 1.5 times the rated torque), fully ensuring the safe operation of the module.

[0026] Preferably, the coupling compensation control model expression used by the fuzzy neural coupling compensation control module is as follows: ; In the formula, The output is controlled by fuzzy neural coupling compensation. These are the weight coefficients of the neural fuzzy network. It is the Sigmoid activation function. This is the predicted temperature value output by the probabilistic temperature field uniformity prediction module. For metabolic quantification parameters, For network bias terms, This is the integral coefficient for temperature deviation. For time variables, Set the temperature value for the incubator. This represents the coefficient of change in metabolic parameters. The time rate of change of metabolic quantification parameters; the temperature field prediction model expression used in the probabilistic temperature field uniformity prediction module is: ; In the formula, This represents the probability value of temperature field uniformity. The variance of the temperature readings inside the incubator. For the first The measured values ​​at each temperature detection point This is the average value of all temperature measurement points. This refers to the allowable temperature fluctuation range. This refers to the number of detection points whose temperature readings fall within the set temperature fluctuation range. This represents the total number of temperature detection points.

[0027] Specifically, during the computation process, the fuzzy neural coupling compensation control module sets the initial value range of the neural fuzzy network weight coefficients to 0.1-0.8, the network bias term value range to 0.05-0.3, the temperature deviation integral coefficient to 0.02-0.1, and the metabolic parameter change rate coefficient to 0.01-0.05. The activation function is the standard Sigmoid function. During computation, it receives the predicted temperature value output by the probabilistic temperature field uniformity prediction module and the metabolic quantification parameters output by the respiratory metabolic parameter quantification module in real time. Combining these with the set temperature value and the metabolic parameter change rate over time, it generates coupling compensation. The output is controlled. When the probabilistic temperature field uniformity prediction module is running, the variance of the temperature detection values ​​is calculated using the most recent 100 sets of valid detection data. The allowable temperature fluctuation range is set to ±0.3℃. The ratio of the number of detection points within the set temperature fluctuation range to the total number of detection points is calculated. Combined with the Gaussian distribution function, the probability value of temperature field uniformity is calculated. When the probability value is lower than 92%, the warning information is immediately transmitted to the fuzzy neural coupling compensation control module. The module's operation cycle is consistent with the data sampling cycle, both being 10 seconds, to ensure that the calculation results can reflect environmental changes in real time and provide accurate basis for subsequent control adjustments.

[0028] Preferably, the temperature field dynamic prediction model expression used by the probabilistic temperature field uniformity prediction module is: ; In the formula, for Predicted temperature value at time of day for Predicted temperature value at time of day To predict the correction factor, for Average temperature measurement at time [time]. The spatial gradient influence coefficient. This refers to the number of temperature detection points. for Time of the first Temperature measurements at each detection point For the first The distance from each detection point to the center of the incubator; the metabolic coupling regulation model expression used by the fuzzy neural coupling compensation control module is: ; In the formula, This is the metabolic coupling regulatory coefficient. For adjustment coefficients, The hyperbolic tangent activation function is used. This is the carbon dioxide concentration value detected by the carbon dioxide concentration control module. This represents the rate of change of carbon dioxide concentration over time.

[0029] Specifically, in the dynamic prediction calculation of the probabilistic temperature field uniformity prediction module, the prediction correction coefficient is set to 0.15-0.4, the spatial gradient influence coefficient is set to 0.08-0.2, and the number of temperature detection points is determined according to the volume of the incubator, with 8-12 points per cubic meter. The difference between the measured temperature value and the current predicted temperature value at each detection point is calculated, and a weighted sum is performed based on the distance from the detection point to the center of the incubator to achieve temperature prediction at the next moment. In the metabolic coupling adjustment calculation of the fuzzy neural coupling compensation control module, the adjustment coefficients are set to 0.2-0.6, 0.05-0.3, 0.03-0.2, and 0.02-0.15, respectively. The hyperbolic tangent activation function is used to process the coupled data of metabolic quantification parameters and carbon dioxide concentration. At the same time, the carbon dioxide concentration time change rate is introduced for dynamic correction to generate metabolic coupling adjustment coefficients. The adjustment coefficients are updated every 5 seconds during module operation to ensure that the coefficients can adapt to the dynamic changes of sample metabolism. The calculation results of the predicted temperature value and the adjustment coefficient are verified by the internal verification program. If the error exceeds ±0.05, the calculation is recalculated to ensure data accuracy.

[0030] Preferably, the gradient oscillation control model expression used by the gradient fusion oscillation control module of the incubator is: ; In the formula, The output value is the oscillation frequency. Based on the oscillation frequency, The coefficient representing the influence of temperature field uniformity. This refers to the probability value of temperature field uniformity output by the probabilistic temperature field uniformity prediction module. To compensate for the control influence coefficient, The output is controlled by fuzzy neural coupling compensation. To compensate for the maximum limit of the control output, The oscillation angular frequency, For time variables, The oscillation phase angle is used; the temperature-oscillation cooperative model expression used by the fuzzy neural coupling compensation control module is: ; In the formula, The target value for temperature control, To set the temperature value, To compensate for the control deviation coefficient, To compensate for the average value of the control output, This is the oscillation frequency deviation coefficient. This represents the average oscillation frequency.

[0031] Specifically, in the gradient oscillation control operation of the gradient fusion oscillation control module of the incubator, the basic oscillation frequency is set to 20-80 rpm, the temperature field uniformity influence coefficient is set to 0.1-0.3, the compensation control influence coefficient is set to 0.05-0.2, the maximum limit of the compensation control output is determined according to the rated load of the oscillation mechanism, usually 0.8-1.2 times the rated output, the oscillation angular frequency is calculated based on the basic oscillation frequency, the initial value of the oscillation phase angle is set to 0, and the range of 0-π / 2 is dynamically adjusted with the operation cycle to generate the oscillation frequency output value; fuzzy neural coupling compensation control. In the temperature-oscillation collaborative calculation of the module, the temperature control target value is based on the set temperature value, the compensation control deviation coefficient is set to 0.08-0.25, the oscillation frequency deviation coefficient is set to 0.03-0.15, the average value of the compensation control output and the average value of the oscillation frequency are the arithmetic mean of the most recent 20 sets of calculation data, the oscillation frequency output value and the temperature detection value are collected every 8 seconds during the calculation, and the collaborative calculation results are corrected to ensure that the temperature control target value can match the oscillation state. The module calculation results are transmitted to the execution unit in real time through the communication interface, and the delay is controlled within 1.5 seconds.

[0032] Preferably, the metabolic quantification model expression used by the respiratory metabolic parameter quantification module is as follows: In the formula, For metabolic quantification parameters, For the number of gas samples, For the first Oxygen concentration in the second sample. For the first The carbon dioxide concentration of the second sample. For the first Gas flow rate at the time of the next sampling The sampling time interval, For sample quality; the gradient distribution model expression used by the gradient fusion oscillation control module of the incubator is: ; Coordinates inside the incubator gradient distribution value at, Based on the gradient value, This is the temperature gradient coefficient. coordinates The predicted temperature value at that location, This is the average predicted temperature value. To predict the standard deviation of temperature, The spatial attenuation coefficient, These are the spatial coordinates within the incubator.

[0033] Specifically, in the metabolic quantification calculation of the respiratory metabolic parameter quantification module, the number of gas samplings is set to 5-10 times, the sampling time interval is set to 1-3 minutes, and the sample mass is pre-measured using a high-precision electronic scale with an accuracy of ±0.01 grams. Oxygen concentration, carbon dioxide concentration, and gas flow rate data are collected for each sampling. The difference between oxygen and carbon dioxide concentrations is calculated, and combined with the sampling volume, time interval, and sample mass, metabolic quantification parameters are generated. The concentration detection data needs to be filtered to remove fluctuation interference within ±0.005%. The gradient fusion oscillation control module of the incubator... In the distribution calculation, the basic gradient value is set to 0.5-2℃ / m, the temperature gradient coefficient is set to 0.12-0.35, the standard deviation of the predicted temperature is calculated using the most recent 30 sets of predicted temperature data, the spatial attenuation coefficient is set to 0.05-0.18, and 10-15 spatial regions are divided according to the three-dimensional coordinates of the incubator. The difference between the predicted temperature value and the average predicted temperature value in each region is calculated, and attenuation correction is performed in combination with the spatial coordinate distance to generate the gradient distribution value. The calculation result of the gradient distribution value must meet the requirement that the gradient difference between each region does not exceed ±0.3℃ / m; otherwise, the coefficients are readjusted and the calculation is performed.

[0034] Preferably, the concentration control model expression used by the carbon dioxide concentration regulation module is: ; Supply flow for carbon dioxide. Basic supply flow, For control coefficients, Set the concentration for carbon dioxide. To measure the concentration of carbon dioxide, For time variables, For metabolic quantification parameters; the temperature-oscillation cooperative model expression used in the isothermal shaker drive module is: ; In the formula, The oscillation speed of the shaking table, Based on the swing speed, This is the temperature deviation coefficient. The target value for temperature control, This is the temperature measurement value of the shaker. The oscillation frequency coefficient is... This is the output value for the oscillation frequency.

[0035] Specifically, in the concentration control calculation of the carbon dioxide concentration regulation module, the basic supply flow rate is set to 0.2-1 L / min, and the control coefficients are set to 0.15-0.45, 0.05-0.2, and 0.03-0.18 respectively. The carbon dioxide set concentration is set to 3-10% according to the sample culture requirements. The concentration measurement value is the average of 3 detection points, and the concentration data is collected every 3 seconds. The difference and integral value between the set concentration and the measured concentration are calculated. Combined with the metabolic quantification parameters, the supply flow rate is adjusted. The supply flow rate adjustment range does not exceed 0.1 L / min each time to avoid sudden changes in concentration. In the temperature-oscillation coordinated calculation of the constant temperature shaker drive module... The basic swing speed is set to 5-20 mm / s, the temperature deviation coefficient is set to 0.08-0.22, and the oscillation frequency coefficient is set to 0.02-0.12. The temperature control target value comes from the fuzzy neural coupling compensation control module. The temperature measurement value of the shaker is the average of three detection points on the table. Temperature and oscillation frequency data are collected every 4 seconds. The correction values ​​of temperature deviation and oscillation frequency on swing speed are calculated. The swing speed is adjusted in a step-like manner, with each adjustment not exceeding 1 mm / s to ensure smooth operation of the shaker. The calculation results of concentration control and swing coordination are verified through closed-loop feedback. If the error exceeds ±0.05, it is recalculated.

[0036] Preferably, the incubator gradient fusion oscillation control module includes an oscillation frequency adjustment unit, a gradient distribution control unit, a multi-region collaborative unit, and a feedback calibration unit. The oscillation frequency adjustment unit receives control commands sent by the fuzzy neural coupling compensation control module, acquires the current oscillation frequency signal of the shaker, compares the deviation between the command frequency and the current frequency using an internal frequency adjustment algorithm, generates a frequency adjustment signal, and transmits it to the shaker drive component to adjust the motor speed and transmission ratio of the drive component to change the oscillation frequency. The gradient distribution control unit acquires temperature and gas concentration distribution data in different regions of the incubator, combines it with the frequency signal output by the oscillation frequency adjustment unit, determines the gradient adjustment target for each region using a gradient calculation algorithm, generates a gradient control signal, and transmits it to the airflow guide in the incubator. The device and heating element adjust the gradient distribution by changing the airflow direction, velocity, and power distribution of the heating element. The multi-region collaborative unit receives the output signals from the oscillation frequency adjustment unit and the gradient distribution control unit, collects the state feedback data of each region in the incubator, coordinates the oscillation and gradient parameters of different regions through a collaborative algorithm to avoid parameter conflicts between regions, generates collaborative control signals, and feeds them back to the oscillation frequency adjustment unit and the gradient distribution control unit respectively. The feedback calibration unit collects the output parameters of the gradient fusion oscillation control module of the incubator and the actual detected oscillation and gradient state data, calculates the parameter deviation value, generates a calibration signal through a calibration algorithm, and transmits it to the oscillation frequency adjustment unit, the gradient distribution control unit, and the multi-region collaborative unit to correct the output parameters of each unit to reduce the deviation.

[0037] Specifically, the gradient fusion oscillation control module of the incubator achieves precise control of gradient and oscillation through the coordinated work of the oscillation frequency adjustment unit, gradient distribution control unit, multi-region collaborative unit, and feedback calibration unit. This solves the problems of uneven local temperature field and insufficient adaptability of oscillation parameters within the incubator, ensuring the dynamic stability of the sample culture environment. In practice, the oscillation frequency adjustment unit receives control commands from the fuzzy neural coupling compensation control module, with a command frequency range of 10-150 rpm. It acquires the current oscillation frequency of the shaker (detected by an encoder, accuracy ±0.1 rpm), compares the deviation between the command frequency and the current frequency, and generates a frequency adjustment signal when the deviation exceeds ±0.5 rpm. This signal is transmitted to the shaker drive motor, and the oscillation frequency is changed by adjusting the motor speed (adjustment accuracy ±0.2 rpm) and transmission ratio (fixed transmission ratio 1:5). The gradient distribution control unit acquires the temperature (accuracy ±0.1℃) and gas concentration (accuracy ±0.5℃) of eight zones within the incubator. The system uses 0.05% data, combined with the frequency signal output from the oscillation frequency adjustment unit (updated every 5 seconds), to determine the gradient adjustment target (gradient range 0-5℃ / m) for each zone through a gradient algorithm. It then generates a gradient control signal that is transmitted to the airflow guide device (adjustment angle range 0-90°, accuracy ±1°) and the heating element (power adjustment range 20-100 watts, accuracy ±2 watts). The multi-zone coordination unit receives output signals from each unit every 10 seconds, collects status feedback data from 12 zones, and coordinates parameters through a coordination algorithm to avoid parameter conflicts between zones (e.g., the coupling deviation between oscillation frequency and gradient adjustment is controlled within ±0.3℃ / m). It then generates a coordination control signal that is fed back to each unit. The feedback calibration unit collects module output parameters and actual detection data (oscillation frequency error ±0.3 rpm, gradient error ±0.2℃ / m) every 15 seconds, calculates the deviation value, and generates a calibration signal to correct the output of each unit, ensuring that the deviation does not exceed ±0.1 rpm and ±0.1℃ / m.

[0038] Preferably, the respiratory metabolic parameter quantification module includes a gas sampling unit, a component analysis unit, a parameter conversion unit, and a data filtering unit. The gas sampling unit receives a system start signal or a timed sampling command, controls the sampling pump and valves to extract a quantitative gas sample from different sample areas in the incubator, and delivers the gas sample to the component analysis unit through a sampling pipeline, while simultaneously recording the sampling time, sampling location, and sampling volume data. The component analysis unit receives the gas sample delivered by the gas sampling unit, activates its internal gas sensor and detection circuit, detects the concentrations of oxygen, carbon dioxide, and metabolic product gases in the sample, and converts the detected analog signals into digital signals. The signal is generated to produce concentration detection data and transmitted to the parameter conversion unit. The parameter conversion unit receives the concentration detection data output by the component analysis unit, combines it with the sampling parameters recorded by the gas sampling unit, and converts the concentration data into quantitative parameters reflecting the respiratory metabolic intensity of the sample through a metabolic quantification algorithm. At the same time, it calculates the time change rate of the parameters, generates metabolic quantification parameters and change rate data, and transmits them to the data filtering unit. The data filtering unit receives the metabolic quantification parameters and change rate data output by the parameter conversion unit, uses a filtering algorithm to remove noise interference signals in the data, retains the effective parameter change trend, generates a filtered metabolic parameter signal, and transmits it to the fuzzy neural coupling compensation control module.

[0039] Specifically, the respiratory metabolic parameter quantification module, through a gas sampling unit, component analysis unit, parameter conversion unit, and data filtering unit, achieves precise quantification of metabolic parameters, accurately capturing the metabolic state of samples in real time, providing reliable sample status feedback to the core control module, and avoiding control deviations caused by distorted metabolic data. In practice, the gas sampling unit receives the system start signal or a timed sampling command every 2 minutes, controlling the sampling pump (flow range 0-10 ml / min, accuracy ±0.1 ml / min) and solenoid valve (response time ≤0.5 seconds) to extract 5 ml of quantitative gas samples (error ±0.1 ml) from 6 sample areas in the incubator. These samples are then transported to the component analysis unit through PTFE tubing (2 mm inner diameter, 1 m length), while simultaneously recording the sampling time (accuracy ±1 second), location, and volume data. The component analysis unit activates the gas sensor (oxygen detection range 0- The system, consisting of a 25% carbon dioxide detection range (0-15%, accuracy ±0.01%) and a detection circuit, converts analog signals into digital signals (sampling rate 100Hz), generating concentration detection data. This data is transmitted to the parameter conversion unit every 3 seconds. The parameter conversion unit, combining sampling parameters (sampling time interval, volume, sample mass, sample mass accuracy ±0.01g), uses a metabolic quantification algorithm to convert the concentration data into metabolic quantification parameters (range 0-100 units, accuracy ±0.5 units). It calculates the parameter time change rate (range ±5 units / hour, accuracy ±0.2 units / hour) and transmits it to the data filtering unit every 5 seconds. The data filtering unit uses a moving average filtering algorithm (window size 10 sets of data) to remove noise interference within ±0.3 units, retaining the parameter change trend, and generating a filtered metabolic parameter signal (updated every 2 minutes) which is then transmitted to the fuzzy neural coupling compensation control module.

[0040] Preferably, the carbon dioxide concentration control module includes a concentration detection unit, a supply control unit, a flow regulation unit, and a leakage monitoring unit. The concentration detection unit collects carbon dioxide concentration signals from different locations within the incubator, converts the analog concentration signals into digital signals via a detection circuit, performs mean calculation and stability assessment on the digital signals, generates carbon dioxide concentration detection data, and transmits it to the supply control unit and the leakage monitoring unit. The supply control unit receives control commands from the fuzzy neural coupling compensation control module and the concentration detection data from the concentration detection unit, compares the difference between the commanded concentration and the detected concentration, calculates the required carbon dioxide supply amount using a supply algorithm, generates a supply control signal, and transmits it to the carbon dioxide control unit. The supply device controls the valve opening and gas supply pressure to adjust the supply volume; the flow regulation unit receives the supply control signal from the supply control unit, collects airflow speed and pressure data in the incubator, determines the airflow circulation rate and flow path through the flow algorithm, generates a flow control signal and transmits it to the fan and airflow guide component in the incubator, and changes the fan speed and guide component angle to adjust the airflow state; the leakage monitoring unit receives the concentration detection data from the concentration detection unit and the supply volume data from the supply control unit, calculates the deviation between the theoretical concentration change value and the actual concentration change value, determines whether there is a carbon dioxide leak through the leakage judgment algorithm, generates a leakage monitoring signal and transmits it to the supply control unit.

[0041] Specifically, the carbon dioxide concentration control module maintains stable carbon dioxide concentration through the coordinated operation of the concentration detection unit, supply control unit, flow regulation unit, and leakage monitoring unit. This addresses environmental instability caused by carbon dioxide concentration fluctuations and leaks within the incubator, meeting the sample's specific gas environment requirements. In practice, the concentration detection unit collects carbon dioxide concentration signals from four locations within the incubator (detected via infrared sensors, accuracy ±0.05%), converts the analog signals to digital signals (sampling rate 50Hz), calculates the mean of 10 data sets (updating the mean every 2 seconds), determines concentration stability (fluctuation range ≤ ±0.03% is considered stable), and generates concentration detection data which is transmitted to the supply control unit and leakage monitoring unit. The supply control unit receives control commands (concentration setting range 3-10%) and concentration detection data from the fuzzy neural coupling compensation control module. When the detected concentration deviates from the set concentration by more than ±0.1%, the required supply amount (range 0-2 liters / minute) is calculated using a supply algorithm, generating a supply control signal. The gas is supplied to the carbon dioxide supply unit, controlling the valve opening (adjustment range 0-100%, accuracy ±1%) and supply pressure (range 0.1-0.5MPa, accuracy ±0.01MPa). The flow regulation unit receives the supply control signal every 3 seconds, collects data on the airflow velocity (range 0-1 m / s, accuracy ±0.05 m / s) and pressure (range 0.09-0.11MPa, accuracy ±0.001MPa) within the chamber, and determines the circulation rate (range 0.2-1 m / s) and flow path using a flow algorithm. The system generates control signals that are transmitted to the fan (speed range 0-2000 rpm, accuracy ±50 rpm) and guide components (angle adjustment range 0-60°, accuracy ±1°). The leak monitoring unit receives concentration detection data and supply data (accuracy ±0.02 liters / minute) every 10 seconds, calculates the deviation between the theoretical and actual concentration changes, and determines that a leak exists when the deviation exceeds ±0.05% / minute. It then generates a monitoring signal and transmits it to the supply control unit, triggering a leak response mechanism (such as reducing the supply and triggering an alarm).

[0042] like Figure 2As shown, the intelligent feedback high-precision incubator CO2 constant temperature shaker fully automatic system operates in the following steps: Step S1, start all modules of the system, so that the fuzzy neural coupling compensation control module, probabilistic temperature field uniformity prediction module, incubator gradient fusion oscillation control module, respiratory metabolic parameter quantification module, CO2 concentration regulation module, and constant temperature shaker drive module enter the initial working state. Each module completes internal parameter initialization and self-check, and sends an initial state signal to the fuzzy neural coupling compensation control module; Step S2, the probabilistic temperature field uniformity prediction module starts to collect temperature detection values ​​of different areas inside the incubator, organizes and converts the collected temperature data, inputs it into the internal probabilistic prediction algorithm for calculation, generates temperature field uniformity evaluation data, and transmits the evaluation data to the fuzzy neural coupling compensation control module; Step S3, the respiratory metabolic parameter quantification module starts the gas sampling unit, extracts gas samples from the incubator, and sends them to the component analysis unit for gas component detection. The detected concentration data is transmitted to the parameter conversion unit to be converted into metabolic quantification parameters, filtered by the data filtering unit, and then transmitted to the fuzzy neural coupling compensation control module. The neural coupling compensation control module; Step S4, the fuzzy neural coupling compensation control module receives temperature field assessment data from the probabilistic temperature field uniformity prediction module and metabolic parameter signals from the respiratory metabolic parameter quantification module, and combines them with the target parameters set by the system, and performs calculations through an internal coupling compensation algorithm to generate control commands for the incubator gradient fusion oscillation control module, carbon dioxide concentration regulation module, and constant temperature shaker drive module; Step S5, the incubator gradient fusion oscillation control module, carbon dioxide concentration regulation module, and constant temperature shaker drive module respectively receive the control commands from the fuzzy neural coupling compensation control module, and, combined with the current working status and detection data of their respective modules, adjust the oscillation frequency, gradient distribution, carbon dioxide supply, shaker motion state, and temperature parameters through an internal control algorithm; Step S6, each module feeds back the adjusted working parameters and the actual detected state data to the fuzzy neural coupling compensation control module, the fuzzy neural coupling compensation control module compares the deviation between the feedback data and the set target parameters, performs coupling compensation calculations again, generates new control commands and sends them to each execution module to continuously adjust and stabilize the system parameters.

[0043] This intelligent feedback-type high-precision incubator with a fully automated carbon dioxide constant-temperature shaker system utilizes six interconnected modules. The fuzzy neural coupling compensation control module integrates temperature field prediction data and metabolic parameters, sending precise control commands to other modules to achieve coordinated regulation of temperature, carbon dioxide concentration, and oscillation state. The probabilistic temperature field uniformity prediction module comprehensively collects temperature data from different areas within the incubator, generating reliable temperature field assessment data. The incubator gradient fusion oscillation control module flexibly adjusts the oscillation frequency and gradient distribution. The respiratory metabolic parameter quantification module accurately converts gas composition data into metabolic parameters. Combined with the carbon dioxide concentration control and constant-temperature shaker drive modules, this system forms a fully automated control process, reducing human intervention, adapting to multi-sample culture scenarios, and significantly improving the stability and controllability of the culture process.

[0044] To address the issue of insufficient coordination among control modules, this system uses a fuzzy neural coupling compensation control module at its core. This module deeply integrates temperature field prediction, metabolic quantification, oscillation control, and concentration regulation, transmitting data and commands in real time to avoid parameter adjustment lags, rapidly respond to environmental changes, and ensure overall stability of the culture environment. To address the limited accuracy of temperature field prediction and metabolic parameter quantification, the probabilistic temperature field uniformity prediction module fully considers the influence of spatial gradient differences within the chamber and sample metabolic dynamics on the temperature field. The respiratory metabolic parameter quantification module, through sampling, analysis, conversion, and filtering, reduces data interference, accurately reflects the sample's metabolic state, and provides reliable support for control command generation, meeting the requirements of high-precision culture.

[0045] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," "link," and "fix" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal communication between two components. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0046] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various equivalent changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A fully automatic intelligent feedback-type high-precision incubator / carbon dioxide constant temperature shaker system, characterized in that: include: Fuzzy neural coupling compensation control module, probabilistic temperature field uniformity prediction module, incubator gradient fusion oscillation control module, respiratory metabolic parameter quantification module, carbon dioxide concentration regulation module, and constant temperature shaker drive module; The fuzzy neural coupling compensation control module receives temperature distribution data from the probabilistic temperature uniformity prediction module and metabolic parameter signals from the respiratory metabolic parameter quantification module. After internal coupling compensation calculation, it sends control commands to the incubator gradient fusion oscillation control module, the carbon dioxide concentration regulation module, and the constant temperature shaker drive module. The probabilistic temperature uniformity prediction module collects temperature detection values ​​from different areas inside the incubator, generates temperature uniformity assessment data through a probabilistic prediction algorithm, and transmits it to the fuzzy neural coupling compensation control module. The incubator gradient fusion oscillation control module receives commands from the fuzzy neural coupling compensation control module and adjusts the oscillation frequency and gradient distribution inside the incubator based on its own gradient oscillation parameters. The respiratory metabolic parameter quantification module collects data on the gas composition and metabolic product concentrations generated by the respiratory metabolism of samples inside the incubator, converts them into quantified parameters, and transmits them to the fuzzy neural coupling compensation control module. The carbon dioxide concentration regulation module receives commands from the fuzzy neural coupling compensation control module and adjusts the carbon dioxide supply and flow rate to maintain a stable carbon dioxide concentration inside the incubator. The constant temperature shaker drive module receives instructions from the fuzzy neural coupling compensation control module to control the shaking amplitude, frequency, and working status of the constant temperature heating element, thereby coordinating the control of the shaker temperature and motion state.

2. The fully automatic intelligent feedback high-precision incubator carbon dioxide constant temperature shaker system according to claim 1, characterized in that, The coupling compensation control model expression used by the fuzzy neural coupling compensation control module is as follows: ； In the formula, The output is controlled by fuzzy neural coupling compensation. These are the weight coefficients of the neural fuzzy network. It is the Sigmoid activation function. This is the predicted temperature value output by the probabilistic temperature field uniformity prediction module. For metabolic quantification parameters, For network bias terms, This is the integral coefficient for temperature deviation. For time variables, Set the temperature value for the incubator. This represents the coefficient of change in metabolic parameters. The time rate of change of metabolic quantification parameters; the temperature field prediction model expression used in the probabilistic temperature field uniformity prediction module is: ； In the formula, This represents the probability value of temperature field uniformity. The variance of the temperature readings inside the incubator. For the first The measured values ​​at each temperature detection point This is the average value of all temperature measurement points. This refers to the allowable temperature fluctuation range. This refers to the number of detection points whose temperature readings fall within the set temperature fluctuation range. This represents the total number of temperature detection points.

3. The fully automatic intelligent feedback high-precision incubator carbon dioxide constant temperature shaker system according to claim 1, characterized in that, The temperature field dynamic prediction model expression used by the probabilistic temperature field uniformity prediction module is as follows: In the formula, for Predicted temperature value at time of day for Predicted temperature value at time of day To predict the correction factor, for Average temperature measurement at time [time]. The spatial gradient influence coefficient. This refers to the number of temperature detection points. for Time of the first Temperature measurements at each detection point For the first The distance from each detection point to the center of the incubator; the metabolic coupling regulation model expression used by the fuzzy neural coupling compensation control module is: ； In the formula, This is the metabolic coupling regulatory coefficient. For adjustment coefficients, The hyperbolic tangent activation function is used. This is the carbon dioxide concentration value detected by the carbon dioxide concentration control module. This represents the rate of change of carbon dioxide concentration over time.

4. The fully automatic intelligent feedback high-precision incubator carbon dioxide constant temperature shaker system according to claim 1, characterized in that, The gradient oscillation control model expression used by the gradient fusion oscillation control module of the incubator is as follows: ； In the formula, The output value is the oscillation frequency. Based on the oscillation frequency, The coefficient representing the influence of temperature field uniformity. This refers to the probability value of temperature field uniformity output by the probabilistic temperature field uniformity prediction module. To compensate for the control influence coefficient, The output is controlled by fuzzy neural coupling compensation. To compensate for the maximum limit of the control output, The oscillation angular frequency, For time variables, The oscillation phase angle is used; the temperature-oscillation cooperative model expression used by the fuzzy neural coupling compensation control module is: ； In the formula, The target value for temperature control, To set the temperature value, To compensate for the control deviation coefficient, To compensate for the average value of the control output, This is the oscillation frequency deviation coefficient. This represents the average oscillation frequency.

5. The fully automatic intelligent feedback high-precision incubator carbon dioxide constant temperature shaker system according to claim 1, characterized in that, The metabolic quantification model expression used in the respiratory metabolic parameter quantification module is as follows: ； In the formula, For metabolic quantification parameters, For the number of gas samples, For the first Oxygen concentration in the second sample. For the first The carbon dioxide concentration of the second sample. For the first Gas flow rate at the time of the next sampling The sampling time interval, For sample quality; the gradient distribution model expression used by the gradient fusion oscillation control module of the incubator is: ; Coordinates inside the incubator gradient distribution value at, Based on the gradient value, This is the temperature gradient coefficient. coordinates The predicted temperature value at that location, This is the average predicted temperature value. To predict the standard deviation of temperature, The spatial attenuation coefficient, These are the spatial coordinates within the incubator.

6. The fully automatic intelligent feedback high-precision incubator carbon dioxide constant temperature shaker system according to claim 1, characterized in that, The concentration control model expression used by the carbon dioxide concentration regulation module is as follows: ； Carbon dioxide supply flow rate Basic supply flow, For control coefficients, Set the concentration for carbon dioxide. To measure the concentration of carbon dioxide, For time variables, For metabolic quantification parameters; the temperature-oscillation cooperative model expression used in the isothermal shaker drive module is: ； In the formula, The oscillation speed of the shaking table, Based on the swing speed, This is the temperature deviation coefficient. The target value for temperature control, This is the temperature measurement value of the shaker. The oscillation frequency coefficient is... This is the output value for the oscillation frequency.

7. The fully automatic intelligent feedback high-precision incubator carbon dioxide constant temperature shaker system according to claim 1, characterized in that, The incubator gradient fusion oscillation control module includes an oscillation frequency adjustment unit, a gradient distribution control unit, a multi-region collaborative unit, and a feedback calibration unit. The oscillation frequency adjustment unit receives control commands from the fuzzy neural coupling compensation control module, acquires the current oscillation frequency signal of the shaker, compares the deviation between the command frequency and the current frequency using an internal frequency adjustment algorithm, generates a frequency adjustment signal, and transmits it to the shaker drive component. This adjusts the motor speed and transmission ratio of the drive component to change the oscillation frequency. The gradient distribution control unit acquires temperature and gas concentration distribution data from different regions within the incubator, combines this data with the frequency signal output from the oscillation frequency adjustment unit, determines the gradient adjustment target for each region using a gradient calculation algorithm, generates a gradient control signal, and transmits it to the airflow guiding device within the incubator. The system adjusts the gradient distribution by changing the airflow direction, velocity, and power distribution of the heating element in conjunction with the heating element. The multi-region collaborative unit receives the output signals from the oscillation frequency adjustment unit and the gradient distribution control unit, collects the state feedback data of each region in the incubator, coordinates the oscillation and gradient parameters of different regions through a collaborative algorithm to avoid parameter conflicts between regions, generates collaborative control signals, and feeds them back to the oscillation frequency adjustment unit and the gradient distribution control unit respectively. The feedback calibration unit collects the output parameters of the gradient fusion oscillation control module of the incubator and the actual detected oscillation and gradient state data, calculates the parameter deviation value, generates a calibration signal through a calibration algorithm, and transmits it to the oscillation frequency adjustment unit, the gradient distribution control unit, and the multi-region collaborative unit to correct the output parameters of each unit to reduce the deviation.

8. The fully automatic intelligent feedback high-precision incubator carbon dioxide constant temperature shaker system according to claim 1, characterized in that, The respiratory metabolic parameter quantification module includes a gas sampling unit, a component analysis unit, a parameter conversion unit, and a data filtering unit. The gas sampling unit receives a system start signal or a timed sampling command, controls the sampling pump and valves to extract quantitative gas samples from different sample areas within the incubator, and delivers the gas samples to the component analysis unit through sampling pipelines. Simultaneously, it records the sampling time, sampling location, and sampling volume data. The component analysis unit receives the gas samples delivered by the gas sampling unit, activates its internal gas sensors and detection circuits, detects the concentrations of oxygen, carbon dioxide, and metabolic products in the sample, converts the detected analog signals into digital signals, generates concentration detection data, and transmits it to the parameter conversion unit. The parameter conversion unit receives the concentration detection data output by the component analysis unit, combines it with the sampling parameters recorded by the gas sampling unit, and uses a metabolic quantification algorithm to convert the concentration data into quantitative parameters reflecting the respiratory metabolic intensity of the sample. Simultaneously, it calculates the time-varying rate of change of the parameters, generates metabolic quantification parameters and rate-varying rate data, and transmits them to the data filtering unit. The data filtering unit receives metabolic quantification parameters and rate of change data output by the parameter conversion unit, uses a filtering algorithm to remove noise interference signals in the data, retains the effective parameter change trend, generates filtered metabolic parameter signals, and transmits them to the fuzzy neural coupling compensation control module.

9. The fully automatic intelligent feedback high-precision incubator carbon dioxide constant temperature shaker system according to claim 1, characterized in that, The carbon dioxide concentration control module includes a concentration detection unit, a supply control unit, a flow regulation unit, and a leakage monitoring unit. The concentration detection unit collects carbon dioxide concentration signals from different locations within the incubator, converts the analog concentration signals into digital signals via a detection circuit, calculates the mean and stability of the digital signals, generates carbon dioxide concentration detection data, and transmits it to the supply control unit and the leakage monitoring unit. The supply control unit receives control commands from the fuzzy neural coupling compensation control module and the concentration detection data from the concentration detection unit, compares the difference between the commanded concentration and the detected concentration, calculates the required carbon dioxide supply amount using a supply algorithm, generates a supply control signal, and transmits it to the carbon dioxide supply unit. The device controls the valve opening and gas supply pressure of the supply device to adjust the supply volume; the flow regulation unit receives the supply control signal from the supply control unit, collects the airflow speed and pressure data in the incubator, determines the airflow circulation rate and flow path through the flow algorithm, generates a flow control signal and transmits it to the fan and airflow guide component in the incubator, and changes the fan speed and the angle of the guide component to adjust the airflow state; the leakage monitoring unit receives the concentration detection data from the concentration detection unit and the supply volume data from the supply control unit, calculates the deviation between the theoretical concentration change value and the actual concentration change value, determines whether there is a carbon dioxide leakage through the leakage judgment algorithm, generates a leakage monitoring signal and transmits it to the supply control unit.

10. The fully automatic intelligent feedback high-precision incubator carbon dioxide constant temperature shaker system according to any one of claims 1-9, characterized in that, The system operation includes the following steps: Step S1, starting each module of the system, enabling the fuzzy neural coupling compensation control module, probabilistic temperature field uniformity prediction module, incubator gradient fusion oscillation control module, respiratory metabolic parameter quantification module, carbon dioxide concentration regulation module, and constant temperature shaker drive module to enter the initial working state. Each module completes internal parameter initialization and self-check, and sends an initial state signal to the fuzzy neural coupling compensation control module; Step S2, the probabilistic temperature field uniformity prediction module starts collecting temperature detection values ​​from different areas inside the incubator, organizes and converts the collected temperature data, inputs it into the internal probabilistic prediction algorithm for calculation, generates temperature field uniformity evaluation data, and transmits the evaluation data to the fuzzy neural coupling compensation control module; Step S3, the respiratory metabolic parameter quantification module starts the gas sampling unit, extracts gas samples from the incubator, and transports them to the component analysis unit for gas component detection. The detected concentration data is transmitted to the parameter conversion unit to be converted into metabolic quantification parameters, filtered by the data filtering unit, and then transmitted to the fuzzy neural coupling compensation control module; Step S4 4. The fuzzy neural coupling compensation control module receives temperature field assessment data from the probabilistic temperature field uniformity prediction module and metabolic parameter signals from the respiratory metabolic parameter quantification module. Combined with the system's set target parameters, it performs calculations using an internal coupling compensation algorithm to generate control commands for the incubator gradient fusion oscillation control module, carbon dioxide concentration regulation module, and constant temperature shaker drive module. Step S5: The incubator gradient fusion oscillation control module, carbon dioxide concentration regulation module, and constant temperature shaker drive module each receive the control commands from the fuzzy neural coupling compensation control module. Combining their respective module's current operating status and detection data, they adjust the oscillation frequency, gradient distribution, carbon dioxide supply, shaker motion state, and temperature parameters using an internal control algorithm. Step S6: Each module feeds back the adjusted operating parameters and the actual detected state data to the fuzzy neural coupling compensation control module. The fuzzy neural coupling compensation control module compares the deviation between the feedback data and the set target parameters, performs a new coupling compensation calculation, generates new control commands, and sends them to each execution module for continuous adjustment and stable control of the system parameters.

Images