Thermostatting method for a constant temperature PCR amplifier

Abstract

The application discloses a temperature control method of a constant-temperature PCR amplification instrument, and belongs to the technical field of PCR amplification instruments, which comprises the following steps: a four-dimensional temperature prediction model is established; a temperature sensor periodically samples the temperature of a temperature adjusting module, the ambient temperature and the temperature of a hot cover; temperature data is input into the four-dimensional temperature prediction model; the internal temperature of reagents is pre-tested; based on historical data, the thermal inertia coefficient and the temperature drift compensation factor of the current cycle are calculated; the target temperature is pre-compensated; according to the deviation and the change rate of the deviation between the pre-compensated target temperature and the temperature of the reagents, the input-output domain and the rule base weight of a fuzzy controller are adjusted; and a power control signal is output. The four-dimensional temperature prediction model is established in advance, so that the reagent temperature can be accurately controlled; meanwhile, the target temperature is pre-compensated, the influence of temperature drift after multiple cycles is reduced, the input-output domain and the rule base weight of the fuzzy controller are dynamically adjusted, and the adaptability to the ambient temperature is improved.

Description

Technical Field

[0001] This invention belongs to the field of PCR amplification instrument technology, and relates to a temperature control method for an isothermal PCR amplification instrument. Background Technology

[0002] Isothermal PCR amplification technology is a molecular biology technique that amplifies nucleic acids at a constant temperature. Compared with traditional variable temperature PCR, it has advantages such as fast reaction speed, simple equipment, and no need for a complex temperature control system. It is widely used in fields such as point-of-care testing (POCT), clinical diagnosis, and food safety testing.

[0003] During use, maintaining a suitable temperature is crucial, requiring high precision and stability in temperature control. Isothermal PCR amplification instruments primarily detect temperature using temperature sensors and adjust it via Peltier modules. To ensure temperature uniformity across all locations, current methods primarily involve using multiple independent Peltier modules, temperature sensors, and thermally conductive materials with better conductivity. However, this still cannot accurately reflect the true temperature of the reagents within the reaction tubes, hindering precise temperature control. Furthermore, heat accumulation after multiple cycles can lead to temperature drift, further impacting the accuracy of temperature regulation. Summary of the Invention

[0004] To address the above problems, this invention proposes a temperature control method for an isothermal PCR amplification instrument, which effectively solves the problems in the prior art.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] A temperature control method for an isothermal PCR amplification instrument, comprising:

[0007] S1. Establish a four-dimensional temperature prediction model of "temperature control module-reagent-environment-heated lid", and initialize the cyclic thermal inertia database and fuzzy PID controller parameters;

[0008] S2. The temperature sensor periodically samples the temperature of the temperature control module, the ambient temperature, and the temperature of the hot cover, and removes noise interference through a moving average filtering algorithm.

[0009] S3. Input the filtered temperature data into the four-dimensional temperature prediction model to predict the internal temperature of the reagent.

[0010] S4, based on the previous Based on historical data from each cycle, calculate the thermal inertia coefficient and temperature drift compensation factor for the current cycle, and pre-compensate for the target temperature.

[0011] S5. Based on the deviation and rate of change of the pre-compensated target temperature and the predicted reagent temperature, adjust the input and output domains and rule base weights of the fuzzy controller, and output the power control signal.

[0012] S6. Based on the predicted reagent temperature and ambient temperature, dynamically adjust the overall temperature of the hot cap and perform independent temperature control compensation for the hot cap area corresponding to the edge holes.

[0013] S7. After each loop ends, update the weights of the cyclic thermal inertia database and the fuzzy rule base.

[0014] Optionally, the establishment of the four-dimensional temperature prediction model includes:

[0015] S11, Settings One calibration temperature point, covering the commonly used temperature range for isothermal PCR reactions;

[0016] S12. Hold at each calibration temperature point for T minutes, and synchronously collect the temperature of the temperature control module. reagent temperature Ambient temperature and the temperature of the hot cap ;

[0017] S13. The parameters of the four-dimensional temperature prediction model are obtained by fitting using the least squares method:

[0018] ;

[0019] in, For model parameters, This represents the temperature change rate of the temperature control module.

[0020] Optionally, step S4, which involves pre-compensating for the target temperature, includes:

[0021] S41. Establish a cyclic thermal inertia database and record the previous... The rate of temperature change for each temperature segment in each cycle Overshoot With adjustment time ;

[0022] S42. Calculate the thermal inertia coefficient for the current cycle. :

[0023] ;

[0024] in, For the first Overshoot of each cycle, For the first The rate of temperature change in each cycle;

[0025] S43. Calculate the temperature drift compensation factor for the current cycle. :

[0026] ;

[0027] in, For the first The actual stable temperature of each cycle, For the first The target temperature for each cycle;

[0028] S44. Perform pre-compensation for the target temperature:

[0029] .

[0030] Optionally, step S5 includes:

[0031] S51, Calculate temperature deviation and rate of change of deviation :

[0032] ;

[0033] ;

[0034] S52, according to and The absolute value of the value is used to adjust the scaling factor of the input universe of discourse of the fuzzy controller. and ;

[0035] S53. Based on the adjusted input universe of discourse, and The data is blurred into 7 fuzzy subsets: ;

[0036] S54. Perform fuzzy inference based on the fuzzy rule base to obtain the adjustment amount of the PID parameters. , , ;

[0037] S55, according to the preceding The control effect of each loop automatically adjusts the weight of each rule in the fuzzy rule base. ;

[0038] S56. Calculate the final PID parameters. , , ;

[0039] S57, Output power control signal:

[0040] .

[0041] Optionally, the input universe scaling factor in step S52 is:

[0042] ;

[0043] ;

[0044] in, and These are the maximum and minimum thresholds for temperature deviation. and The maximum and minimum thresholds for the rate of change of deviation, , This is the proportionality coefficient.

[0045] Optionally, the update of the fuzzy rule base weights in step S55 is as follows:

[0046] ;

[0047] in, For the first The first cycle The weight of each rule For learning rate, For the first The first cycle The activation level of the rule, This represents the temperature deviation of the current wheel.

[0048] Optionally, step S6 includes:

[0049] S61, The overall temperature of the hot cover is set as follows:

[0050] ;

[0051] in, Based on the basic temperature difference;

[0052] S62. Additional compensation is added to the temperature of the heat cover area corresponding to the edge hole:

[0053] ;

[0054] in, Standard ambient temperature, The compensation coefficient;

[0055] S63. The final temperature setting for the edge hole heat cover area is:

[0056] .

[0057] Optionally, step S7 includes:

[0058] S71. After each cycle, record the rate of temperature change, overshoot, settling time, and steady-state temperature for that cycle.

[0059] S72. Update the cyclic thermal inertia database, retaining the most recent data. Data in a loop;

[0060] S73, each A loop, based on the most recent The control effect of each cycle updates the initial parameters of the fuzzy PID controller. , , ;

[0061] S74. Once the set number of uses or usage duration is reached, the parameters of the four-dimensional temperature prediction model are refitted using the latest calibration data.

[0062] Optionally, step S73 includes:

[0063] After the isothermal stabilization phase of each cycle is completed, the steady-state error of this cycle is automatically extracted. Overshoot With adjustment time The comprehensive score is obtained by normalizing the three indicators and then weighting and summing them. :

[0064] ;

[0065] in, For the maximum allowable steady-state error, The maximum allowable overshoot, Maximum allowable adjustment time , , These are the weighting coefficients;

[0066] Update PID initial parameters:

[0067] ;

[0068] in, The learning rate for updating the initial parameters of the PID controller. , , The partial derivatives of the overall score with respect to the three PID parameters;

[0069] when At this time, the control effect is good, and there is no need to update the PID initial parameters (0.8). At that time, the control effect was mediocre, and the learning rate was halved to update the parameters. When the control effect is poor, the PID initial parameters need to be updated normally.

[0070] Compared with existing technologies, this invention has the following advantages: By pre-establishing a four-dimensional temperature prediction model of "temperature control module-reagent-environment-heated cap," the temperature of reagents that are inconvenient to test directly can be predicted during use, facilitating precise control of reagent temperature. Simultaneously, by setting the thermal inertia coefficient and temperature drift compensation factor based on historical data from the previous N cycles, the target temperature is pre-compensated, reducing the impact of temperature drift after multiple cycles. Furthermore, by adjusting the input-output domain and rule base weights of the fuzzy controller based on the deviation and rate of change between the target temperature and the predicted reagent temperature, the adaptability to ambient temperature is improved. Detailed Implementation

[0071] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0072] A temperature control method for an isothermal PCR amplification instrument disclosed in this embodiment of the invention includes:

[0073] S1. Establish a four-dimensional temperature prediction model of "temperature control module-reagent-environment-heated lid", and initialize the cyclic thermal inertia database and fuzzy PID controller parameters;

[0074] S2. The temperature sensor periodically samples the temperature of the temperature control module, the ambient temperature, and the temperature of the hot cover, and removes noise interference through a moving average filtering algorithm.

[0075] S3. Input the filtered temperature data into the four-dimensional temperature prediction model to predict the internal temperature of the reagent.

[0076] S4. Based on the historical data of the previous N cycles, calculate the thermal inertia coefficient and temperature drift compensation factor of the current cycle, and pre-compensate for the target temperature.

[0077] S5. Based on the deviation and rate of change of the pre-compensated target temperature and the predicted reagent temperature, adjust the input and output domains and rule base weights of the fuzzy controller, and output the power control signal.

[0078] S6. Based on the predicted reagent temperature and ambient temperature, dynamically adjust the overall temperature of the hot cap and perform independent temperature control compensation for the hot cap area corresponding to the edge holes.

[0079] S7. After each loop ends, update the weights of the cyclic thermal inertia database and the fuzzy rule base.

[0080] Specifically, by pre-establishing a four-dimensional temperature prediction model encompassing the "temperature control module, reagent, environment, and heated lid," the system predicts the temperature of reagents that are difficult to measure directly during use, facilitating precise temperature control. Simultaneously, by using thermal inertia coefficients and temperature drift compensation factors set based on historical data from the previous N cycles, the system pre-compensates for the target temperature, reducing the impact of temperature drift after multiple cycles. Furthermore, by analyzing the deviation between the target temperature and the predicted reagent temperature, and the rate of change of that deviation, the system adjusts the input-output domains and rule base weights of the fuzzy controller, improving its adaptability to ambient temperatures.

[0081] Among the feasible approaches, the establishment of a four-dimensional temperature prediction model includes:

[0082] S11, Settings One calibration temperature point, covering the commonly used temperature range for isothermal PCR reactions;

[0083] S12. Hold at each calibration temperature point for T minutes, and synchronously collect the temperature of the temperature control module. reagent temperature Ambient temperature and the temperature of the hot cap ;

[0084] S13. The parameters of the four-dimensional temperature prediction model are obtained by fitting using the least squares method:

[0085] ;

[0086] in, For model parameters, This represents the temperature change rate of the temperature control module.

[0087] Specifically, by setting up test reagents and test sensors to measure reagent temperature during experiments, and fitting this data with the temperature control module temperature, ambient temperature, and hot cap temperature, model parameters are obtained. In subsequent experiments, the model parameters are used to predict reagent temperature until the set number of uses or usage duration is reached, at which point the model parameters are updated again using test reagents and test sensors.

[0088] After acquiring the model parameters and initializing the fuzzy PID controller parameters according to the device settings, the temperature of the temperature control module, ambient temperature, and hot cap temperature are collected through multiple sensors. The sampling period can be set to 1ms, and the sliding filter window is set to 5. The collected signals are filtered, and the filtered data is stored. Based on the filtered data, the temperature change rate of the temperature control module is calculated, and then the current temperature of the reagent is calculated through the model parameters.

[0089] Calling multiple temperature change rates from the cyclic thermal inertia database Overshoot With adjustment time Calculate the thermal inertia coefficient of the current cycle. With temperature drift compensation factor :

[0090] ;

[0091] ;

[0092] in, For the first Overshoot of each cycle, For the first The rate of temperature change per cycle For the first The actual stable temperature of each cycle, For the first The target temperature for each cycle. Therefore, the pre-compensation for the target temperature is as follows:

[0093] .

[0094] Then the temperature deviation was calculated. and rate of change of deviation :

[0095] ;

[0096] ;

[0097] pass and The absolute value of the value is used to adjust the scaling factor of the input universe of discourse of the fuzzy controller. and :

[0098] ;

[0099] ;

[0100] in, and These are the maximum and minimum thresholds for temperature deviation. and The maximum and minimum thresholds for the rate of change of deviation, , This is the proportionality coefficient.

[0101] Based on the scaling factor of the input universe and Adjust the input universe of discourse, and Multiply by respectively and Mapping to the standard universe of discourse, the mapped and The data is blurred into 7 fuzzy subsets: A triangular membership function is used. Then, fuzzy inference is performed based on the fuzzy rule base to obtain the adjustment amount of the PID parameters. , , Subsequently, according to the previous The control effect of each loop automatically adjusts the weight of each rule in the fuzzy rule base. :

[0102] ;

[0103] in, For the first The first cycle The weight of each rule For learning rate, For the first The first cycle The activation level of the rule, This represents the temperature deviation of the current wheel.

[0104] Thus, the final PID parameters are calculated. , , :

[0105] ;

[0106] in, This represents the total number of fuzzy rules. The membership degree corresponding to a single rule. , , These are the initial parameters.

[0107] Final control output power control signal for:

[0108] .

[0109] To better control temperature uniformity, the overall temperature of the heated cap is dynamically adjusted based on the predicted reagent temperature and ambient temperature, and independent temperature compensation is applied to the heated cap area corresponding to the edge holes.

[0110] The overall temperature setting for the heat cover is:

[0111] ;

[0112] in, The base temperature difference is the fixed temperature difference between the overall temperature of the hot cap and the predicted reagent constant temperature. It can be selected as 3-7℃. When the PCR reaction solution is heated, water vapor will evaporate. If the temperature of the hot cap is lower than the reagent temperature, the water vapor will condense on the top of the tube wall, which may cause changes in the volume of the reaction system and inconsistent amplification between wells. The temperature of the hot cap is moderately higher than the reagent temperature to lock in the water vapor and prevent condensation, without over-drying the reagent.

[0113] Additional compensation is added to the temperature of the heat cover area corresponding to the edge hole:

[0114] ;

[0115] in, The standard ambient temperature is the temperature of the environment in which the equipment is located. The compensation coefficient can be set to 0.1-0.3. The colder the environment, the more heat is lost from the edges. The system will automatically add more temperature compensation to smooth out the temperature difference between the edge holes and the center holes.

[0116] The final temperature setting for the edge hole heat cover area is:

[0117] .

[0118] To facilitate real-time temperature control, after each cycle, the rate of temperature change, overshoot, settling time, and steady-state temperature are recorded and updated to the cycle thermal inertia database. The cycle thermal inertia database retains the most recent data. Each loop of data, each A loop, based on the most recent The control effect of each cycle updates the initial parameters of the fuzzy PID controller. , , .

[0119] Specifically, after the isothermal stabilization phase of each cycle ends, the steady-state error of that cycle is automatically extracted. Overshoot With adjustment time The comprehensive score is obtained by normalizing the three indicators and then weighting and summing them. :

[0120] ;

[0121] in, For the maximum allowable steady-state error, The maximum allowable overshoot, Maximum allowable adjustment time , , The weighting coefficient can be 0.5, 0.3, or 0.2.

[0122] Score based on overall control effectiveness As the loss function, the initial parameters of the PID are updated in reverse:

[0123] ;

[0124] in, The learning rate for updating the initial parameters of the PID controller can be set to 0.001. , , This represents the partial derivative of the overall score with respect to the three PID parameters. When... At that time, the control effect was good, 0.8 At that time, the control effect was mediocre, and the learning rate was halved to update the parameters. When the control effect is poor, the PID initial parameters need to be updated normally.

[0125] Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A temperature control method for an isothermal PCR amplification instrument, characterized in that, include: S1. Establish a four-dimensional temperature prediction model of "temperature control module - reagent - environment - hot cap", and initialize the cyclic thermal inertia database and fuzzy PID controller parameters; S2. The temperature sensor periodically samples the temperature of the temperature control module, the ambient temperature, and the temperature of the hot cover, and removes noise interference through a moving average filtering algorithm. S3. Input the filtered temperature data into the four-dimensional temperature prediction model to predict the internal temperature of the reagent. S4, based on the previous Based on historical data from each cycle, calculate the thermal inertia coefficient and temperature drift compensation factor for the current cycle, and pre-compensate for the target temperature. S5. Based on the deviation and rate of change of the pre-compensated target temperature and the predicted reagent temperature, adjust the input and output domains and rule base weights of the fuzzy controller, and output the power control signal. S6. Based on the predicted reagent temperature and ambient temperature, dynamically adjust the overall temperature of the hot cap and perform independent temperature control compensation for the hot cap area corresponding to the edge holes. S7. After each loop ends, update the weights of the cyclic thermal inertia database and the fuzzy rule base.

2. The temperature control method for an isothermal PCR amplification instrument according to claim 1, characterized in that: The establishment of the four-dimensional temperature prediction model includes: S11, Settings One calibration temperature point, covering the commonly used temperature range for isothermal PCR reactions; S12. Hold at each calibration temperature point for T minutes, and synchronously collect the temperature of the temperature control module. reagent temperature Ambient temperature and the temperature of the hot cap ; S13. The parameters of the four-dimensional temperature prediction model are obtained by fitting using the least squares method: ； in, For model parameters, This represents the temperature change rate of the temperature control module.

3. The temperature control method for an isothermal PCR amplification instrument according to claim 1, characterized in that: The target temperature pre-compensation step S4 includes: S41. Establish a cyclic thermal inertia database and record the previous... The rate of temperature change for each temperature segment in each cycle Overshoot With adjustment time ; S42. Calculate the thermal inertia coefficient for the current cycle. : ； in, For the first Overshoot of each cycle, For the first The rate of temperature change in each cycle; S43. Calculate the temperature drift compensation factor for the current cycle. : ； in, For the first The actual stable temperature of each cycle, For the first The target temperature for each cycle; S44. Perform pre-compensation for the target temperature: 。 4. The temperature control method for an isothermal PCR amplification instrument according to claim 3, characterized in that: Step S5 includes: S51, Calculate temperature deviation and rate of change of deviation : ； ； S52, according to and The absolute value of the value is used to adjust the scaling factor of the input universe of discourse of the fuzzy controller. and ; S53. Based on the adjusted input universe of discourse, and The data is blurred into 7 fuzzy subsets: ; S54. Perform fuzzy inference based on the fuzzy rule base to obtain the adjustment amount of the PID parameters. , , ; S55, according to the preceding The control effect of each loop automatically adjusts the weight of each rule in the fuzzy rule base. ; S56. Calculate the final PID parameters. , , ; S57, Output power control signal: 。 5. The temperature control method for an isothermal PCR amplification instrument according to claim 4, characterized in that: The input universe scaling factor in step S52 is: ； ； in, and These are the maximum and minimum thresholds for temperature deviation. and The maximum and minimum thresholds for the rate of change of deviation, , This is the proportionality coefficient.

6. The temperature control method for an isothermal PCR amplification instrument according to claim 4, characterized in that: The update of the fuzzy rule base weights in step S55 is as follows: ； in, For the first The first cycle The weight of each rule For learning rate, For the first The first cycle The activation level of the rule, This represents the temperature deviation of the current wheel.

7. The temperature control method for an isothermal PCR amplification instrument according to claim 2, characterized in that: Step S6 includes: S61, The overall temperature of the hot cover is set as follows: ； in, Based on the basic temperature difference; S62. Additional compensation is added to the temperature of the heat cover area corresponding to the edge hole: ； in, Standard ambient temperature, The compensation coefficient; S63. The final temperature setting for the edge hole heat cover area is: 。 8. The temperature control method for an isothermal PCR amplification instrument according to claim 1, characterized in that: Step S7 includes: S71. After each cycle, record the rate of temperature change, overshoot, settling time, and steady-state temperature for that cycle. S72. Update the cyclic thermal inertia database, retaining the most recent data. Data in a loop; S73, each A loop, based on the most recent The control effect of each cycle updates the initial parameters of the fuzzy PID controller. , , ; S74. Once the set number of uses or usage duration is reached, the parameters of the four-dimensional temperature prediction model are refitted using the latest calibration data.

9. The temperature control method for an isothermal PCR amplification instrument according to claim 8, characterized in that: Step S73 includes: After the isothermal stabilization phase of each cycle is completed, the steady-state error of this cycle is automatically extracted. Overshoot With adjustment time The comprehensive score is obtained by normalizing the three indicators and then weighting and summing them. : ； in, For the maximum allowable steady-state error, The maximum allowable overshoot, Maximum allowable adjustment time , , These are the weighting coefficients; Update PID initial parameters: ； in, The learning rate for updating the initial parameters of the PID controller. , , The partial derivatives of the overall score with respect to the three PID parameters; when At this time, the control effect is good, and there is no need to update the PID initial parameters (0.8). At that time, the control effect was mediocre, and the learning rate was halved to update the parameters. When the control effect is poor, the PID initial parameters need to be updated normally.

Images