A Quantification Method for Dynamic Consistency of Temperature Field Based on Thermal Field Analysis

Abstract

This invention discloses a quantitative method for the dynamic consistency of temperature field based on thermal field analysis. Temperature measurement points are set up upstream and downstream of the filter. During the isothermal stage, a set disturbance is applied and multiple temperature sequences are collected. The preprocessed temperature data is obtained after detrending and bandpass filtering. Frequency domain analysis is used to calculate the average phase value of the upstream and downstream temperature sequences within a preset frequency band, and the results are grouped into the principal value region to form a downstream phase set. Based on this set, the absolute value of the phase mean and circular statistical parameters are obtained. By calculating the complementary value of the absolute value of the phase mean and the exponential decay result of the circular statistical parameters, a comprehensive score reflecting the temperature uniformity of the extinguishing effect is obtained, thus achieving a quantitative evaluation of the dynamic consistency of the extinguishing temperature field.

Description

Technical Field

[0001] This invention relates to the field of thermal field analysis technology, and in particular to a quantitative method for the dynamic consistency of temperature field based on thermal field analysis. Background Technology

[0002] Microbial incubators are widely used in laboratory and biopharmaceutical environments for the cultivation, preservation, and periodic sterilization of microorganisms. Sterilization typically involves high-temperature dry heat sterilization (e.g., 140–180 degrees Celsius), moist heat decontamination (approximately 90 degrees Celsius), and chemical gas decontamination (e.g., hydrogen peroxide vapor). Existing equipment often employs single-point sensors and a chamber circulation heating system for temperature control, using heaters, fans, and air deflectors to maintain a stable temperature distribution. However, under different sterilization modes, the airflow velocity, humidity, pressure, and thermal boundary conditions within the system vary significantly. Especially in systems with filters, the airflow direction and heat exchange efficiency are not constant, resulting in a complex time-varying temperature field within the isothermal zone.

[0003] Temperature uniformity is typically assessed by setting up multiple measuring points and calculating the instantaneous temperature difference at each point. This static assessment method is suitable for steady-state heating equipment, but its reliability decreases significantly in incubators with air circulation and filtration devices. Filter materials change air resistance under high temperature or humid conditions; when fans start or stop, or when filter units are installed or removed, the airflow direction and heat transfer path change, causing time lags and fluctuations in temperature distribution. Because traditional uniformity tests only record instantaneous temperatures and do not consider this dynamic delay, they often result in superficially uniform measurements while local areas have not yet reached the kill temperature, leading to inaccurate judgments.

[0004] Existing methods for analyzing the sterilization temperature in incubators neglect the impact of air filtration units on the temperature field under different operating conditions. The filtration device not only alters the airflow circulation but also possesses a certain heat capacity and hygroscopicity, causing its physical state to change with the switching of sterilization modes. For example, in the humid heat decontamination mode, the filter material absorbs moisture and responds slowly to the conduction and release of hot air; in the high-temperature dry heat mode, the filter material is removed or in an inert state, creating entirely different heat transfer pathways within the chamber. These state changes cause the temperature responses in different areas inside the chamber to become asynchronous, resulting in the temperature uniformity index measured by traditional methods failing to reflect the actual heat distribution during the sterilization phase. Summary of the Invention

[0005] Purpose of the invention: The present invention aims to provide a quantitative method for the dynamic consistency of temperature field based on thermal field analysis, which is used to quantify the dynamic consistency of temperature field during the sterilization process in a microbial incubator.

[0006] Technical solution: The quantitative method for dynamic consistency of temperature field based on thermal field analysis described in this invention includes:

[0007] (1) Set at least one temperature measuring point upstream of the filter of the microbial incubator and set multiple temperature measuring points downstream of the filter;

[0008] (2) During the constant temperature stage of the sterilization process, a set disturbance is applied to the incubator, and the temperature sequence of each temperature measuring point is collected according to the preset sampling parameters;

[0009] (3) Perform detrending and bandpass filtering on each temperature sequence to obtain the preprocessed temperature sequence;

[0010] (4) Using frequency domain analysis, calculate the average phase of the preprocessed temperature sequence of the upstream temperature measuring point and the preprocessed temperature sequence of each downstream temperature measuring point in the preset frequency band, and classify each average phase into the preset main value area to form a downstream phase set.

[0011] (5) Based on the downstream phase set, the average absolute value of the phase and the circular statistical parameters are calculated;

[0012] (6) Calculate the complementary value of the phase average absolute value, perform attenuation processing on the circular statistical parameters, and calculate the extermination effect score reflecting the temperature uniformity based on the complementary value and the attenuation processing result, thus completing the quantification of the dynamic consistency of the extermination temperature field.

[0013] Furthermore, in step (1), the direction of air flowing from the filter to the downstream is set as positive, based on the neutral plane of the filter;

[0014] The directed distance from the temperature measuring point to the neutral plane is defined as the shortest distance with a sign, where a positive value indicates that it is on the positive side and a negative value indicates that it is on the negative side.

[0015] If the directed distance from the temperature measuring point to the neutral plane is negative, then the temperature measuring point is designated as the upstream temperature measuring point; if the directed distance from the temperature measuring point to the neutral plane is positive, then the temperature measuring point is designated as the downstream temperature measuring point.

[0016] Furthermore, step (2) is as follows:

[0017] Determine the sampling time interval and the number of sampling points;

[0018] The total sampling duration shall be no less than three times the preset disturbance period;

[0019] The set temperature is the sum of the constant temperature setpoint and the preset disturbance that varies according to a sinusoidal law; where the preset disturbance period is the period that varies according to a sinusoidal law.

[0020] Apply the set temperature during the constant temperature phase;

[0021] Determine the discrete sampling time according to the sampling time interval;

[0022] The total number of temperature measuring points is the sum of the number of upstream temperature measuring points and the number of downstream temperature measuring points;

[0023] The temperature of each temperature measuring point is collected at each discrete sampling time to form a temperature sequence corresponding to each temperature measuring point.

[0024] Furthermore, step (3) is as follows:

[0025] Perform cubic polynomial least squares fitting on the temperature sequence corresponding to each temperature measurement point to obtain the trend term, and subtract the corresponding trend term from the temperature sequence to obtain the detrended temperature sequence.

[0026] The lower bandpass cutoff frequency, upper bandpass cutoff frequency, and filter order are set, and the detrended temperature sequence is processed using a zero-phase bidirectional Butterworth bandpass filter to obtain the preprocessed temperature sequence.

[0027] Furthermore, in step (4), the frequency domain analysis method is used to calculate the average phase of the preprocessed temperature sequence of the upstream temperature measuring point and the preprocessed temperature sequence of each downstream temperature measuring point within a preset frequency band, as follows:

[0028] The upstream and downstream temperature measurement point sets are determined based on the positive and negative directions of the directed distances from the temperature measurement points to the neutral plane of the filter.

[0029] Calculate the average value of the preprocessed temperature sequence corresponding to all upstream temperature measurement points to obtain the upstream reference preprocessed temperature sequence;

[0030] Discrete Fourier transforms were performed on the upstream reference preprocessed temperature sequence and the preprocessed temperature sequence corresponding to each downstream temperature measurement point, respectively.

[0031] The cross spectrum corresponding to each downstream temperature measuring point is obtained by conjugate operation, and the argument of each cross spectrum is extracted as the phase of the corresponding downstream temperature measuring point.

[0032] The preset frequency band is determined based on the lower pass threshold frequency and the upper pass threshold frequency;

[0033] Filter out the set of discrete frequency points within the preset frequency band;

[0034] The average phase of each downstream temperature measuring point is obtained by averaging all phases within the discrete frequency set.

[0035] Furthermore, in step (4), the average value of each phase is assigned to a preset principal value region to form a downstream phase set, as follows:

[0036] The default principal value area is the range from negative pi to positive pi, and the default principal value area includes positive pi;

[0037] For each downstream temperature measurement point, the phase average value is divided by the sum of the phase average value and pi by twice the pi value, and then rounded down. The product of twice the pi value and the rounded down value is subtracted from the phase average value to obtain the result mapped to the preset principal value area.

[0038] If the mapped result is a negative pi, then the result is adjusted to a positive pi to obtain the principal value phase corresponding to each downstream temperature measurement point;

[0039] Collect the principal-valued phases corresponding to all downstream temperature measurement points to form a downstream phase set.

[0040] Furthermore, step (5) is as follows:

[0041] Calculate the average of the absolute values ​​of all principal-valued phases to obtain the average absolute value of the phases;

[0042] Calculate the average cosine value of each principal-valued phase to obtain the circular statistical cosine average.

[0043] Calculate the average of the sine values ​​for each principal-valued phase to obtain the circular statistical sine average.

[0044] The square root of the sum of the squares of the cosine mean and the squares of the sine mean of the circular statistics is obtained to obtain the consistency of the circular statistics.

[0045] Subtracting the circular statistical consistency from one gives the circular statistical dispersion.

[0046] Furthermore, step (6) is as follows:

[0047] The complementary value of the phase mean absolute value is calculated, and the circular statistical correlation parameters are attenuated. Based on the complementary value and the attenuation result, a kill effect score reflecting temperature uniformity is calculated, including:

[0048] Subtract the ratio of the average absolute value of the phase to pi from 1 to obtain the complementary value of the average absolute value of the phase.

[0049] The results are obtained by applying an exponential decay to the circular statistical dispersion, with the natural constant as the base and the negative value of the circular statistical dispersion as the exponent.

[0050] Multiplying the complementary value of the phase average absolute value with the result after exponential decay yields a kill effect score that reflects temperature uniformity, thus quantifying the dynamic consistency of the kill temperature field.

[0051] Beneficial Effects: Compared with existing technologies, the significant advantages of this invention are: By performing frequency domain analysis on the temperature response upstream and downstream of the air filter structure during the sterilization isothermal stage of the microbial incubator, this invention achieves dynamic temperature uniformity quantification based on phase relationships. This invention extracts the phase average between upstream and downstream by performing detrending and zero-phase bandpass filtering on each temperature sequence, and comprehensively evaluates the spatial distribution consistency using circular statistical parameters. This converts the dynamic temperature difference during the sterilization process into a continuous sterilization effect score, thus avoiding the shortcomings of traditional instantaneous temperature difference methods that cannot reflect the effects of time delay and airflow disturbance. This scoring result can stably characterize the true uniformity of the temperature field within the chamber under different sterilization modes, providing a repeatable and quantifiable analytical basis for incubator sterilization verification, and significantly improving the accuracy and reliability of sterilization effect evaluation. Attached Figure Description

[0052] Figure 1 This is a schematic diagram illustrating a specific implementation scenario of the present invention;

[0053] Figure 2 This is a flowchart of the quantification method for dynamic consistency of temperature field based on thermal field analysis according to the present invention. Detailed Implementation

[0054] The quantitative method for the dynamic consistency of temperature field based on thermal field analysis described in this invention includes:

[0055] (1) such as Figure 1 As shown, at least one temperature measuring point D is set upstream of the filter A in the microbial incubator, and multiple temperature measuring points E are set downstream of the filter C. Using the neutral plane B of the filter as a reference, the direction F of airflow from the HEPA filter G downstream is defined as positive.

[0056] The directed distance from the temperature measuring point to the neutral plane is defined as the shortest distance with a sign, where a positive value indicates that it is on the positive side and a negative value indicates that it is on the negative side.

[0057] If the directed distance from the temperature measuring point to the neutral plane is negative, then the temperature measuring point is designated as the upstream temperature measuring point; if the directed distance from the temperature measuring point to the neutral plane is positive, then the temperature measuring point is designated as the downstream temperature measuring point.

[0058] (2) For example Figure 2 As shown, during the isothermal stage of the sterilization process, a set disturbance is applied to the incubator, and the temperature sequence of each temperature measuring point is collected according to the preset sampling parameters.

[0059] Determine the sampling time interval and the number of sampling points;

[0060] The total sampling duration shall be no less than three times the preset disturbance period;

[0061] The set temperature is the sum of the constant temperature setpoint and the preset disturbance that varies according to a sinusoidal law; where the preset disturbance period is the period that varies according to a sinusoidal law.

[0062] Apply the set temperature during the constant temperature phase;

[0063] Determine the discrete sampling time according to the sampling time interval;

[0064] The total number of temperature measuring points is the sum of the number of upstream temperature measuring points and the number of downstream temperature measuring points;

[0065] The temperature of each temperature measuring point is collected at each discrete sampling time to form a temperature sequence corresponding to each temperature measuring point.

[0066] (3) Perform detrending and bandpass filtering on each temperature sequence to obtain the preprocessed temperature sequence.

[0067] Perform cubic polynomial least squares fitting on the temperature sequence corresponding to each temperature measurement point to obtain the trend term, and subtract the corresponding trend term from the temperature sequence to obtain the detrended temperature sequence.

[0068] The lower bandpass cutoff frequency, upper bandpass cutoff frequency, and filter order are set, and the detrended temperature sequence is processed using a zero-phase bidirectional Butterworth bandpass filter to obtain the preprocessed temperature sequence.

[0069] (4) Using frequency domain analysis, calculate the average phase of the preprocessed temperature sequence of the upstream temperature measuring point and the preprocessed temperature sequence of each downstream temperature measuring point in the preset frequency band, and classify each average phase into the preset main value area to form a downstream phase set.

[0070] Using frequency domain analysis, the average phase of the preprocessed temperature sequences from upstream and downstream temperature measuring points within a preset frequency band is calculated, as follows:

[0071] The upstream and downstream temperature measurement point sets are determined based on the positive and negative directions of the directed distances from the temperature measurement points to the neutral plane of the filter.

[0072] Calculate the average value of the preprocessed temperature sequence corresponding to all upstream temperature measurement points to obtain the upstream reference preprocessed temperature sequence;

[0073] Discrete Fourier transforms were performed on the upstream reference preprocessed temperature sequence and the preprocessed temperature sequence corresponding to each downstream temperature measurement point, respectively.

[0074] The cross spectrum corresponding to each downstream temperature measuring point is obtained by conjugate operation, and the argument of each cross spectrum is extracted as the phase of the corresponding downstream temperature measuring point.

[0075] The preset frequency band is determined based on the lower pass threshold frequency and the upper pass threshold frequency;

[0076] Filter out the set of discrete frequency points within the preset frequency band;

[0077] The average phase of each downstream temperature measuring point is obtained by averaging all phases within the discrete frequency set.

[0078] The average values ​​of each phase are assigned to a preset principal value region to form a downstream phase set, as follows:

[0079] The default principal value area is the range from negative pi to positive pi, and the default principal value area includes positive pi;

[0080] For each downstream temperature measurement point, the phase average value is divided by the sum of the phase average value and pi by twice the pi value, and then rounded down. The product of twice the pi value and the rounded down value is subtracted from the phase average value to obtain the result mapped to the preset principal value area.

[0081] If the mapped result is a negative pi, then the result is adjusted to a positive pi to obtain the principal value phase corresponding to each downstream temperature measurement point;

[0082] Collect the principal-valued phases corresponding to all downstream temperature measurement points to form a downstream phase set.

[0083] (5) Based on the downstream phase set, the average absolute value of the phase and the circular statistical parameters are calculated.

[0084] Calculate the average of the absolute values ​​of all principal-valued phases to obtain the average absolute value of the phases;

[0085] Calculate the average cosine value of each principal-valued phase to obtain the circular statistical cosine average.

[0086] Calculate the average of the sine values ​​for each principal-valued phase to obtain the circular statistical sine average.

[0087] The square root of the sum of the squares of the cosine mean and the squares of the sine mean of the circular statistics is obtained to obtain the consistency of the circular statistics.

[0088] Subtracting the circular statistical consistency from one gives the circular statistical dispersion.

[0089] (6) Calculate the complementary value of the phase average absolute value, perform attenuation processing on the circular statistical parameters, and calculate the extermination effect score reflecting the temperature uniformity based on the complementary value and the attenuation processing result, thus completing the quantification of the dynamic consistency of the extermination temperature field.

[0090] The complementary value of the phase mean absolute value is calculated, and the circular statistical correlation parameters are attenuated. Based on the complementary value and the attenuation result, a kill effect score reflecting temperature uniformity is calculated, including:

[0091] Subtract the ratio of the average absolute value of the phase to pi from 1 to obtain the complementary value of the average absolute value of the phase.

[0092] The results are obtained by applying an exponential decay to the circular statistical dispersion, with the natural constant as the base and the negative value of the circular statistical dispersion as the exponent.

[0093] Multiplying the complementary value of the phase average absolute value with the result after exponential decay yields a kill effect score that reflects temperature uniformity, thus quantifying the dynamic consistency of the kill temperature field.

[0094] Preferably, step (1) involves classifying measurement points based on the neutral plane of the filter and the positive direction of the airflow, including:

[0095] (a) Definition of coordinate datum and positive direction:

[0096] The plane containing the geometric center of the filter is taken as the neutral plane, denoted as . Set the direction of airflow from the filter to the downstream as positive.

[0097] (b) Formula for calculating directed distance:

[0098] At any temperature measuring point within the extermination space (in ), calculate its distance to the neutral plane The shortest distance with signs The calculation formula is as follows:

[0099] ;

[0100] in, The sign function is taken when the measuring point is on the positive side. When located on the opposite side, take .

[0101] (c) Logic for determining upstream and downstream measuring points:

[0102] Based on the calculated directed distance The positive and negative values ​​of the temperature measurement points are used to divide the temperature measurement points into a set of upstream temperature measurement points. and downstream temperature measurement point set for:

[0103] ;

[0104] ;

[0105] like If so, the measuring point is determined to be an upstream temperature measuring point; if If so, then the measuring point is determined to be a downstream temperature measuring point.

[0106] in, The neutral plane of the filter. This represents the total number of temperature measurement points. The index number of the temperature measurement point. For the first Spatial location vector of each temperature measuring point For the first Temperature measuring points to the neutral plane The Euclidean shortest distance (non-negative real number). For the first The directed distance from each temperature measuring point to the neutral plane. This is a set of indices for upstream temperature measurement points. This is a set of indices for downstream temperature measurement points.

[0107] Preferably, step (2) involves applying a set perturbation and acquiring a temperature sequence, including:

[0108] (a) Determination of sampling parameters:

[0109] Determine the sampling time interval and number of sampling points This makes the total sampling time The following constraint formula is:

[0110] ;

[0111] in The preset disturbance period varies according to a sinusoidal law.

[0112] (b) Setting the temperature calculation formula (superposition of isothermal and disturbance):

[0113] During the isothermal phase of the extermination process, a preset perturbation is applied according to a sinusoidal law. The set temperature at each discrete sampling time The calculation formula is as follows:

[0114] ;

[0115] in, The constant temperature setting is H, which is the amplitude of the preset disturbance, and the sine term is the applied preset disturbance.

[0116] (c) Discrete sampling and sequence formation:

[0117] Determine each discrete sampling time according to the sampling time interval. And collect temperature data at each measurement point at each time point. The temperature values ​​form a corresponding temperature sequence. for:

[0118] ;

[0119] ;

[0120] Total number of all temperature measuring points This is the sum of the number of upstream temperature measuring points and the number of downstream temperature measuring points.

[0121] in, Total sampling time The number of sampling points. The sampling time interval, To preset the disturbance period, For the first Each discrete sampling time, For the first The set temperature at each discrete sampling time. For constant temperature setpoint, The initial phase of the preset disturbance, For the first At time 1, temperature measuring points The measured temperature For the first Discrete temperature sequence corresponding to each temperature measurement point This represents the total number of temperature measurement points.

[0122] Preferably, step (3) involves preprocessing the temperature sequence, including:

[0123] (a) Detrending using cubic polynomial least squares:

[0124] For each temperature measuring point Corresponding temperature sequence Establish information about discrete sampling times cubic polynomial trend model for:

[0125] ;

[0126] Determining the coefficient set using the least squares method This minimizes the sum of squared residuals;

[0127] Subtract the corresponding trend term from the temperature series to obtain the detrended temperature series. for:

[0128] ;

[0129] (b) Zero-phase bidirectional Butterworth bandpass filter:

[0130] Set the lower bandpass frequency Bandpass upper limit frequency and filter order Constructing Butterworth bandpass filter operators The detrended temperature series was processed using a zero-phase bidirectional filter. The process was performed to obtain the preprocessed temperature sequence. for:

[0131] ;

[0132] in, For the first The original temperature sequence of each temperature measurement point For discrete sampling times, To fit the obtained trend term sequence, The coefficients of the cubic polynomial are determined by least squares fitting. This is the detrended temperature sequence (residual sequence). To be the lower bandpass frequency, To limit the bandpass frequency, Let the filter order be . Design operators for Butterworth bandpass filters. It is a zero-phase bidirectional filtering operator (including forward filtering and reverse filtering). This is the temperature sequence after preprocessing.

[0133] Preferably, step (4) uses a frequency domain analysis method to calculate the phase average value, including:

[0134] (a) Construction of upstream reference preprocessed temperature sequence:

[0135] The set of upstream temperature measuring points is determined based on the positive or negative sign of the directed distance from the temperature measuring point to the neutral plane of the filter. and downstream temperature measurement point set ;

[0136] Calculate the preprocessed temperature sequence corresponding to all upstream temperature measurement points. The average value is used to obtain the upstream reference preprocessed temperature sequence. for:

[0137] ;

[0138] (b) Discrete Fourier Transform and Cross-Spectrum Calculation:

[0139] The upstream reference preprocessed temperature sequence was processed separately. and each downstream temperature measuring point Corresponding preprocessed temperature sequence Perform a Discrete Fourier Transform (DFT) to obtain the spectrum. and ;

[0140] The cross spectrum corresponding to each downstream temperature measuring point was calculated using conjugate operations. for:

[0141] ;

[0142] ;

[0143] in This indicates the complex conjugate operation.

[0144] (c) Phase extraction and bandwidth selection:

[0145] Extracting each cross spectrum The argument is used as the phase of the corresponding downstream temperature measuring point. for:

[0146] ;

[0147] Based on the lower pass threshold frequency and bandpass upper limit frequency Determine the preset frequency band and filter out the set of discrete frequency point indices within that frequency band. for:

[0148] ;

[0149] (d) Calculation of average phase value within the frequency band:

[0150] For each downstream temperature measuring point Calculate its discrete frequency set All phases within The average value is used to obtain the phase average value corresponding to the measuring point. for:

[0151] ;

[0152] in, This is a set of upstream temperature measurement points. This is a set of downstream temperature measurement points. The number of upstream temperature measurement points (i.e., the set) momentum ), For the first generation belonging to the upstream set Preprocessed temperature sequence of each measuring point The upstream reference preprocessed temperature sequence, For the first generation belonging to the downstream set Preprocessed temperature sequence of each measuring point This corresponds to the discrete frequency domain sequence (spectrum). The number of sampling points. The imaginary unit, For the first Cross-spectrum of downstream measuring points relative to the upstream reference, For the first Downstream measurement points at frequency points Phase at that point, These are the lower and upper cutoff frequencies of the bandpass filter. The sampling time interval, A set of discrete frequency point indices within a preset frequency band. For set The number of frequency points included For the first The average phase value of each downstream measuring point.

[0153] Preferably, in step (4), the average value of each phase is assigned to a preset principal value region to form a downstream phase set, including:

[0154] (a) Definition of principal value region:

[0155] Set the preset main value area Let be a semi-open interval between negative and positive pi, and let be the interval containing the positive pi:

[0156] ;

[0157] (b) Principalization mapping computation:

[0158] For each downstream temperature measuring point Corresponding phase average value Calculate the integer number of the corresponding circumference. for:

[0159] ;

[0160] Using this integer The phase average value is mapped to a preset principal value region to obtain the mapping result. for:

[0161] ;

[0162] in This indicates the floor function.

[0163] (c) Boundary adjustment and phase set formation:

[0164] Check mapping results If the result is negative pi, it is adjusted to positive pi to obtain the final principal-valued phase. for:

[0165] ;

[0166] Collect all downstream temperature measurement points (in The corresponding principal-valued phase Forming downstream phase sets for:

[0167] ;

[0168] in, Preset main value area , Pi For the first The original phase average value of each downstream measuring point This is the floor function. This represents the integer number of cycles of the phase shift. The phase value after initial mapping. This is the final principal-valued phase after boundary adjustment. This is a set of downstream temperature measurement points. This is the downstream phase set.

[0169] Preferably, step (5) involves calculating the phase mean absolute value and circular statistical parameters based on the downstream phase set, including:

[0170] (a) Calculation of the average absolute value of the phase:

[0171] Obtain downstream phase set Number of downstream temperature measurement points included The principal-valued phase in the set is denoted as ( );

[0172] Calculate the arithmetic mean of the absolute values ​​of all principal-valued phases to obtain the phase mean absolute value. for:

[0173] ;

[0174] (b) Calculation of the average of circular statistical vectors:

[0175] Calculate each principal-valued phase separately The average of the cosine and sine values ​​is used to obtain the circular statistical cosine mean. Sine average of circles for:

[0176] ;

[0177] ;

[0178] (c) Circular statistical consistency calculation:

[0179] The square root of the sum of the squares of the cosine mean and the squares of the sine mean of the circular statistics yields the circular statistical consistency. for:

[0180] ;

[0181] (d) Calculation of circular statistical dispersion:

[0182] Subtract the circular statistical consistency from 1 to obtain the circular statistical dispersion. for:

[0183] ;

[0184] in, The total number of downstream temperature measurement points (i.e., the downstream phase set) (number of elements) For set The Middle Principal phase of each downstream temperature measuring point The absolute value of the phase average. The circular statistical cosine mean, The sine mean of the circle is statistically analyzed. This represents the circular statistical consistency (its value ranges from 0 to 1, with a larger value indicating a more concentrated phase). This is the circular statistical dispersion (its value is between 0 and 1, and the smaller the value, the more concentrated the phase).

[0185] Preferably, in step (6), the complementary value of the phase mean absolute value is calculated, the circular statistical correlation parameters are attenuated, and the extermination effect score is calculated, including:

[0186] (a) Calculation of the complementary value of the average absolute value of the phase:

[0187] Calculate the absolute value of the phase mean With Pi The ratio of the two values ​​is used to subtract the ratio from the sum of the two values ​​to obtain the complementary value of the phase mean absolute value. for:

[0188] ;

[0189] (b) Exponential decay treatment of circular statistical dispersion:

[0190] With natural constant Using the base as the statistical dispersion, the circular distribution is used. The negative value is used as the exponent for calculation, resulting in the product after applying an exponential decay to the circular statistical dispersion. for:

[0191] ;

[0192] (c) Calculation of extermination effectiveness score:

[0193] The complementary value of the phase mean absolute value Results after exponential decay Multiplying the results yields a kill effectiveness score that reflects temperature uniformity. The quantification of the dynamic consistency of the extermination temperature field is completed as follows:

[0194] ;

[0195] in, The absolute value of the phase average. Pi It is the complementary value of the average absolute value of the phase (the larger the value, the smaller the overall phase lag). For circular statistical dispersion, It is a natural constant. This is the result after applying exponential decay to the circular statistical dispersion (the larger the value, the more uniform the spatial distribution). The extermination effect is scored (this score comprehensively reflects the dynamic consistency of the temperature field; the higher the value, the better the consistency).

[0196] The filter neutral plane is the reference plane where the filter's geometric center is located, used to delineate the upstream and downstream affiliation of measurement points.

[0197] The temperature measurement point index is a numbered identifier for each temperature measurement point, used to distinguish different measurement points.

[0198] The total number of temperature measurement points is the total number of all temperature measurement points set by humans. It is preferred that there be at least one upstream and at least two downstream, with a total of no less than three points, in order to ensure the effectiveness of the upstream reference signal and the reliability of the downstream statistical analysis.

[0199] The spatial position vector of the i-th temperature measuring point is the coordinate data of the i-th temperature measuring point in three-dimensional space, which can be obtained by measuring with a three-dimensional coordinate measuring instrument or by calibration on the equipment assembly drawings.

[0200] The sign function is an artificially defined identifier function used to distinguish the location and direction of the measuring point. It is preferred to take 1 when the measuring point is on the positive side and -1 when it is on the negative side, so as to clearly define the correspondence between the location of the measuring point and the direction of the airflow.

[0201] The Euclidean shortest distance from the i-th temperature measuring point to the neutral plane is the straight-line distance from the i-th measuring point to the neutral plane, and the value is non-negative.

[0202] The directed distance from the i-th temperature measuring point to the neutral plane is the signed distance obtained by multiplying the sign function by the Euclidean shortest distance, and is used to determine the upstream and downstream affiliation of the measuring point.

[0203] The upstream temperature measurement point set is an index summary of all directed temperature measurement points with negative distances, used for unified processing of upstream measurement point signals.

[0204] The downstream temperature measurement point set is an index summary of all directed temperature measurement points with positive distances, used for unified processing of downstream measurement point signals.

[0205] The method of dividing upstream and downstream measuring points based on the neutral plane where the filter's geometric center is located requires that the neutral plane be determined based on the midpoint of the filter frame's thickness direction, and that this plane be perpendicular to the airflow streamline direction. For example, if the filter frame thickness is 100 mm, the neutral plane is the plane perpendicular to the airflow at a point 50 mm in the thickness direction. Automatic classification using directional distance symbols avoids the subjectivity of traditional experience-based labeling of upstream and downstream points.

[0206] The definition of the signed shortest distance is to correlate the straight-line distance from the measuring point to the neutral plane with the airflow direction, with the measurement direction being the projection direction along the airflow axis. For example, if the airflow is from the filter to the extinguishing area in the positive direction, and a measuring point is on the positive side of the neutral plane with an axial projection distance of 120 mm, its directional distance is also 120 mm; another measuring point is on the opposite side with an axial projection distance of 50 mm, its directional distance is -50 mm. The assignment is directly determined by the positive or negative value.

[0207] The specific method for determining the neutral plane is to find the midpoint of the filter frame in the thickness direction on the equipment assembly drawing or the actual object, and draw a plane perpendicular to the airflow streamline through the midpoint. This plane is the neutral plane.

[0208] The directional distance is measured by projecting the distance along the airflow axis. The neutral plane is used as a reference during measurement. The projected distance along the positive direction of the airflow is taken as a positive value, and the distance in the opposite direction is taken as a negative value, ensuring that the distance is directly related to the airflow direction.

[0209] When the directed distance is 0, the measuring point is not included in the classification of upstream and downstream measuring points, nor is it included in subsequent phase calculations, thus avoiding analysis errors caused by classification ambiguity.

[0210] The sign function is determined by the direction of airflow from the filter to the extinguishing area. The sign function of the measuring point located on the side of this direction is 1, and the sign function of the measuring point located on the opposite side (i.e. the side closer to the fan or heating chamber) is -1, clearly distinguishing the directional boundaries.

[0211] The sampling time interval is a manually set time interval between two adjacent temperature acquisitions, preferably 0.1 seconds to 1 second, in order to control the total amount of data while ensuring data accuracy and avoiding redundancy.

[0212] The number of sampling points is a manually set total number of data points to be collected. It is preferably calculated based on the sampling time interval and the total sampling duration to meet the constraint that the total sampling duration is not less than three times the disturbance period.

[0213] The total sampling duration is the actual total data collection time calculated from the sampling time interval and the number of sampling points, and is used to ensure that the collected data covers a sufficient disturbance period.

[0214] The preset disturbance period is a manually set repetition period of a sinusoidal disturbance, preferably 30 to 120 seconds, to adapt to the dynamic response speed of the temperature field and ensure that the disturbance signal can be effectively captured.

[0215] The constant temperature setpoint is a manually set baseline temperature for the sterilization process, preferably 121 degrees Celsius, to meet the temperature requirements for the sterilization of common microorganisms.

[0216] The preset disturbance amplitude is a temperature change range of a sinusoidal disturbance that is set manually, preferably between 0.5 degrees Celsius and 2 degrees Celsius, so as to provide effective excitation without affecting the killing effect.

[0217] The preset initial phase of the disturbance is an initial angle of the sinusoidal disturbance that is set manually, preferably 0, to simplify calculations and not affect the excitation effect of the disturbance signal.

[0218] The nth discrete sampling time is the specific acquisition time point calculated by the sampling time interval and the sequence number, which is used to standardize the time node of data acquisition.

[0219] The measured temperature of the i-th temperature measuring point at time tn is the actual temperature data of the i-th measuring point at the corresponding discrete time, which can be acquired by a contact temperature sensor or a non-contact temperature measurement device.

[0220] The temperature sequence of the i-th temperature measurement point is a data set composed of the measured temperatures of the i-th measurement point at all discrete sampling times in chronological order, which is used for subsequent signal processing.

[0221] The set temperature at the nth moment is the target temperature calculated by superimposing the constant temperature setpoint and the sinusoidal disturbance, and is used to apply a controllable temperature excitation to the incubator.

[0222] The design incorporates sinusoidal perturbations during the isothermal phase of the extermination process. By artificially injecting controllable temperature changes, the originally stable temperature field generates a quantifiable dynamic response. For example, if the isothermal setpoint is 121 degrees Celsius, and a sinusoidal perturbation with an amplitude of 1 degree Celsius and a period of 60 seconds is superimposed, the setpoint temperature will change systematically between 120 and 122 degrees Celsius. This method can overcome the limitations of static temperature acquisition and capture the dynamic transmission characteristics of the temperature field.

[0223] The constraint rule that the total sampling time should not be less than three times the preset perturbation period ensures that the collected data contains sufficient perturbation periods, avoiding distortion in frequency domain analysis due to insufficient data. For example, if the perturbation period is 60 seconds, the total sampling time should be at least 180 seconds to fully cover three perturbation periods, providing a reliable data foundation for subsequent phase calculations.

[0224] Based on the logic of synchronously acquiring temperature sequences from multiple measurement points at discrete sampling times, the temperature data of all measurement points correspond one-to-one in the time dimension, avoiding the impact of time deviation on the analysis of upstream and downstream phase relationships.

[0225] The preset disturbance amplitude is constrained to ensure that the minimum set temperature is not lower than the lower limit of the extermination process. For example, if the lower limit of the extermination process is 118 degrees Celsius and the constant temperature setting is 121 degrees Celsius, the maximum amplitude shall not exceed 3 degrees Celsius to prevent the disturbance from causing the temperature to drop too low and affecting the extermination effect.

[0226] The sampling frequency should be set to be no less than 20 times the disturbance frequency. For example, if the disturbance period is 60 seconds and the disturbance frequency is 0.0167 Hz, the sampling frequency should be no less than 0.334 Hz, that is, the sampling time interval should be no more than 3 seconds, which meets the requirements of the sampling theorem.

[0227] The technical guarantee for multi-point synchronous acquisition is to use a unified clock signal to control all temperature sensors, or to correct deviations through timestamp alignment technology after acquisition, to ensure that the acquisition time difference between different measurement points at the same discrete moment does not exceed 1 millisecond.

[0228] The method for capturing the steady-state window during the isothermal phase is to wait for the temperature field to stabilize for 1 to 2 perturbation cycles after applying the perturbation before starting data acquisition. For example, if the perturbation cycle is 60 seconds, data acquisition is started 120 seconds after the perturbation is applied, and the initial transition period data is discarded.

[0229] The selection of the initial phase of the disturbance is to ensure that the initial phase value does not affect the quantization result of dynamic consistency. Choosing 0 is preferred to simplify the calculation. If there are multiple disturbance sources, the interference can be avoided by adjusting the initial phase.

[0230] The cubic polynomial fitting coefficients are four constants determined by the least squares method and are used to construct a trend model of the temperature series.

[0231] The trend term sequence is a set of temperature change trend data constructed based on cubic polynomial fitting coefficients and discrete sampling times.

[0232] The detrended temperature series is the set of residual data obtained by subtracting the corresponding trend term series from the original temperature series.

[0233] The lower bandpass cutoff frequency is a manually set low-frequency filtering threshold, preferably 0.8 times the perturbation frequency, to cover the low-frequency range of the main lobe of the perturbation signal.

[0234] The upper limit of the bandpass frequency is a manually set high-frequency filtering threshold, preferably 1.2 times the perturbation frequency, to cover the high-frequency range of the main lobe of the perturbation signal.

[0235] The filter order is a manually set order of the Butterworth filter, preferably between 2 and 4th order, to balance the filtering effect and computational complexity.

[0236] The Butterworth bandpass filter operator is a signal filtering operation rule based on the upper and lower bandpass frequencies and the order.

[0237] The zero-phase bidirectional filtering operator is an operation rule that includes both forward and reverse filtering, used to cancel out the phase shift caused by filtering.

[0238] The preprocessed temperature sequence is the final signal set obtained by performing zero-phase bidirectional Butterworth bandpass filtering on the detrended temperature sequence.

[0239] The design employs cubic polynomial least squares fitting for detrending processing. Addressing the slow drift interference that is common in temperature field data, a cubic polynomial model is constructed to accurately capture the trend component. For example, if a temperature sequence slowly increases over time, fitting a cubic polynomial and subtracting the trend term yields a residual sequence containing only the disturbance and the effective signal. Compared to low-order polynomials or moving averages, this approach more comprehensively eliminates complex trends.

[0240] A combined scheme of zero-phase bidirectional Butterworth bandpass filtering overcomes the phase delay problem introduced by traditional unidirectional filtering. This method first performs forward filtering on the sequence, then reverse filtering on the result. The phase shifts from the two filters cancel each other out, ensuring that the output signal is in phase with the original valid signal. For example, in dynamic temperature field analysis, this avoids distortion of the phase relationship between upstream and downstream measuring points caused by filtering, ensuring the accuracy of subsequent phase calculations.

[0241] The setting logic of binding the bandpass frequency with the preset disturbance frequency allows the filter to focus on the main disturbance frequency band, accurately retaining the signal related to dynamic consistency, and filtering out irrelevant noise and high-frequency interference.

[0242] The specific method for solving cubic polynomial least squares fitting involves substituting the discrete sampling time and temperature sequences into the system of equations, minimizing the sum of squared residuals through matrix operations, and obtaining four fitting coefficients. This matrix solution process can be implemented using numerical computation software or programming.

[0243] The bandpass frequency is set based on a preset perturbation frequency, with 0.8 to 1.2 times the perturbation frequency used as the upper and lower bandpass limits. For example, if the perturbation frequency is 0.0167 Hz, the lower bandpass limit is 0.0134 Hz, and the upper bandpass limit is 0.0200 Hz.

[0244] When resources are plentiful, the filter order should be prioritized at 4th order to pursue better filter roll-off characteristics; when resources are limited, the 2nd order should be selected to reduce computational load.

[0245] The long sequence segmentation processing scheme divides the long sequence into segments of fixed length with 50% overlap between segments. After filtering each segment, 10% of the transient points at both ends are discarded, and only the stable middle part is spliced ​​together.

[0246] The specific implementation logic of zero-phase bidirectional filtering is to first apply the Butterworth filter to the detrended sequence in a forward order, then apply the same filter again to the filtering result in a reverse order, and finally output the reverse-filtered sequence.

[0247] The number of upstream temperature measurement points is the total number of measurement points included in the upstream temperature measurement point set.

[0248] The upstream reference preprocessed temperature sequence is a data set obtained by averaging the preprocessed temperature sequences of all upstream measuring points over time.

[0249] The preprocessed temperature sequence of the d-th downstream measuring point is the set of signals obtained from the d-th downstream measuring point after detrending and bandpass filtering.

[0250] The discrete frequency domain sequence of the upstream reference sequence is the spectral data obtained by performing a discrete Fourier transform on the temperature sequence after preprocessing the upstream reference.

[0251] The discrete frequency domain sequence of the d-th downstream measuring point is the spectral data obtained by performing a discrete Fourier transform on the preprocessed temperature sequence of the d-th downstream measuring point.

[0252] The cross spectrum of the d-th downstream measurement point is the result of multiplying the discrete frequency domain sequence of the d-th downstream measurement point with the complex conjugate of the upstream reference discrete frequency domain sequence.

[0253] The phase of the d-th downstream measuring point at frequency k is the argument of the cross spectrum of the d-th downstream measuring point.

[0254] Sampling frequency is the number of samples taken per unit of time.

[0255] The kth discrete frequency is the frequency value corresponding to the kth frequency point.

[0256] The set of discrete frequency points within the preset frequency band is a summary of the indices of all discrete frequency points within the preset frequency band.

[0257] The average phase value of the d-th downstream measurement point is the average result of all phases of the d-th downstream measurement point within the discrete frequency point set.

[0258] The imaginary unit is a constant used to represent the imaginary part of a complex number.

[0259] Complex conjugate operation is a mathematical operation that takes the conjugate of a complex number, that is, inverting the sign of the imaginary part of the complex number.

[0260] The method for constructing the upstream reference preprocessed temperature sequence eliminates single-point measurement errors by spatially averaging the preprocessed sequences of all upstream measurement points, thus forming a high signal-to-noise ratio reference. For example, if there are two upstream measurement points, and the preprocessed sequence values ​​at each time point are 121.2 and 121.3 respectively, the corresponding reference sequence value is 121.25, making the upstream excitation signal more representative.

[0261] By combining Discrete Fourier Transform and complex conjugate operation to calculate the cross spectrum, and then extracting the cross spectrum argument as the phase, the design accurately quantifies the phase relationship between upstream and downstream measurement points. This approach breaks through the limitations of traditional time-domain analysis, effectively separating signals and noise in the frequency domain and focusing on disturbance-related phase information.

[0262] Based on the logic of filtering discrete frequency points using upper and lower bandpass limits, phase calculations focus only on the main perturbation frequency band, eliminating noise interference from irrelevant frequency bands. For example, if the upper and lower bandpass limits correspond to 0.8 to 1.2 times the perturbation frequency, only frequency points within this range are selected for phase averaging, ensuring that the phase results are directly related to dynamic consistency.

[0263] When averaging phases within a frequency set, the complex unit vector averaging method is used to avoid arithmetic mean distortion caused by phases looping around the positive and negative pi. For example, when multiple phases are distributed around the positive and negative pi, the arithmetic mean will deviate from the true value, while the complex unit vector averaging obtains an accurate average phase by summing the cosine and sine values ​​and then taking the argument.

[0264] The engineering implementation of the Discrete Fourier Transform (DFT) prioritizes the Fast Fourier Transform (FFT) algorithm. If the number of sampling points is not a power of 2, it is padded with zeros to the nearest power of 2 before calculation. Zero padding only adjusts the frequency sampling grid and does not change the spectral structure of the original signal, thus improving computational efficiency.

[0265] The order of cross-spectral conjugation operations is determined to be the multiplication of the downstream discrete frequency domain sequence by the complex conjugation of the upstream reference discrete frequency domain sequence. This order ensures that the phase result reflects the hysteresis relationship between the downstream and the upstream, which is consistent with the logic of heat transfer.

[0266] The adaptive adjustment method of the preset frequency band is to calculate the amplitude of the upstream reference discrete frequency domain sequence within the preset frequency band, find the frequency point with the largest amplitude as the actual main frequency, and then extend 3 discrete frequency points to the left and right of this frequency point to form the final discrete frequency point set, which adapts to the small drift of the disturbance frequency.

[0267] To ensure the accuracy of phase extraction, a Hanning window is applied to the upstream and downstream preprocessed temperature sequences before performing the Discrete Fourier Transform. The window function formula is an amplitude equal to 0.5 minus 0.5 multiplied by a cosine function. The independent variable of the cosine function is twice pi multiplied by the time index divided by the number of sampling points minus 1, which reduces the impact of spectral leakage on phase calculation.

[0268] The floating-point tolerance for frequency filtering is set to 10 to the power of -6. When the difference between the discrete frequency and the upper and lower bandpass frequencies is less than this tolerance, the frequency is determined to belong to the preset frequency band.

[0269] The preset principal value region is a uniform phase range set by humans. It is preferably a half-open and half-closed interval from negative pi to positive pi and includes positive pi, so as to standardize the phase values ​​of different ranges.

[0270] Pi is a mathematical constant representing the ratio of a circle's circumference to its diameter.

[0271] The integer number of circumferences is obtained by dividing the sum of the phase average and the value of pi by twice the value of pi and then rounding down.

[0272] The phase value after initial mapping is the result of subtracting the product of twice pi and the integer number of circumferences from the average phase value.

[0273] The principal phase is the final phase value that falls into the preset principal value region after boundary adjustment.

[0274] The downstream phase set is a summary set of principal-valued phases corresponding to all downstream temperature measurement points.

[0275] The floor function is a mathematical operation that rounds the result down to the nearest integer in the direction of negative infinity.

[0276] A predefined half-open, half-closed interval, ranging from negative to positive pi and including positive pi, is designated as the principal value region. Phase range standardization is achieved by mapping any average phase value to this interval. For example, if the average phase value at a downstream measuring point is three times pi, by calculating the integer number of revolutions (1), the initial mapped phase value is three times pi minus two times pi, resulting in positive pi, which falls within the predefined principal value region, thus avoiding statistical errors caused by phase range dispersion.

[0277] The design adjusts phase values ​​that result in negative pi after mapping to positive pi to ensure the integrity of the principal value region and the consistency of phase data. For example, if the average value of a certain phase is negative pi after mapping, it is adjusted to positive pi according to the rules to avoid the impact of boundary value differences on subsequent statistical parameter calculations.

[0278] The phase value is accurately mapped to the principal value region by using an integer number of circumferences. The calculation logic of this integer can automatically adapt to phase average values ​​of different sizes, making the phase standardization process more efficient.

[0279] The floating-point tolerance for phase mapping is set to 10 to the power of -12. When the absolute value of the difference between the phase value after initial mapping and the negative pi is less than this tolerance, it is judged as negative pi and needs to be adjusted to positive pi to avoid misjudgment caused by floating-point calculation errors.

[0280] The association between the principal phase and the measurement point ID is to bind the corresponding downstream measurement point index number to each principal phase when forming the downstream phase set. For example, each element in the set is stored in the form of measurement point index plus principal phase, which makes it easy to trace the specific measurement point corresponding to each phase.

[0281] The rule for handling outliers in the phase set is that if a principal valued phase exceeds the preset principal value range or deviates significantly from other phases, it is judged as an outlier, removed, and the number of downstream measurement points is recounted before subsequent calculations are performed, in order to avoid outlier data affecting the analysis results.

[0282] The number of downstream temperature measurement points is the number of principal-valued phases contained in the downstream phase set.

[0283] The principal-valued phase of the m-th downstream measuring point is the final phase value of the m-th downstream measuring point after principal-valued mapping and boundary adjustment.

[0284] The phase mean absolute value is the arithmetic mean of the absolute values ​​of all principal-valued phases.

[0285] The circular statistical cosine mean is the arithmetic mean of the cosine values ​​of all principal-valued phases.

[0286] The circular statistical sine mean is the arithmetic mean of the sine values ​​of all principal-valued phases.

[0287] Circular statistical consistency is the square root of the sum of the squares of the cosine mean and the squares of the sine mean of the circular statistics.

[0288] The circular statistical dispersion is the result obtained by subtracting the circular statistical consistency from one.

[0289] Applying circular statistical theory to dynamic consistency analysis of temperature fields, this method calculates the cosine and sine averages of the principal-valued phases and then derives methods for circular statistical consistency and dispersion, adapting to the circular characteristics of phase data. For example, with four principal-valued phases representing 0.2, 0.3, 0.25, and 0.35 pi, their cosine and sine values ​​are calculated, averaged, and then the consistency is obtained by taking the square root of the sum of squares, accurately reflecting the concentration of phases.

[0290] By introducing the phase mean absolute value as a quantitative indicator, the overall lag level of downstream measuring points relative to upstream is intuitively characterized by averaging all principal-valued phase absolute values. This indicator, combined with circular statistical parameters, characterizes the dynamic properties of the temperature field from two dimensions: overall lag and spatial concentration, overcoming the limitations of a single indicator.

[0291] By establishing a complementary relationship between circular statistical consistency and dispersion, the greater the consistency, the more concentrated the phase, and the smaller the dispersion, the more uniform the temperature field. This two-way characterization allows for a more comprehensive quantification of the dynamic consistency of the temperature field.

[0292] The meaning of circular statistical parameters: The value of circular statistical consistency ranges from 0 to 1. When the value is close to 1, it indicates that all principal-valued phases are highly concentrated, and when the value is close to 0, it indicates that the phases are dispersed. Circular statistical dispersion is the opposite of consistency. When the value is close to 0, the phases are concentrated, and when it is close to 1, the phases are dispersed.

[0293] The method for handling missing or abnormal downstream measurement points is as follows: if a principal-valued phase is determined to be an outlier or the corresponding measurement point data is missing, the phase is removed and the number of valid downstream measurement points is recounted. Then, based on the valid phase, various statistical parameters are calculated to avoid invalid data affecting the results.

[0294] The accuracy control of the calculation of the absolute value of phase average is to not directly remove individual extreme principal values ​​of phase, but retain them to participate in the averaging. If the extreme values ​​cause the result deviation to be too large, a weighted average method can be used to give the extreme values ​​a lower weight, thus balancing accuracy and objectivity.

[0295] The accuracy requirement for trigonometric function calculations is to retain six decimal places when calculating the cosine and sine values ​​of the principal-valued phase, to ensure the accuracy of subsequent operations such as squaring, summing, and square root extraction, and to reduce the impact of numerical errors on the final parameters.

[0296] The phase mean absolute value is the arithmetic mean of the absolute values ​​of the principal-valued phases at all downstream measuring points.

[0297] Pi is a mathematical constant representing the ratio of a circle's circumference to its diameter.

[0298] The complementary value of the phase mean absolute value is obtained by subtracting the ratio of the phase mean absolute value to pi.

[0299] The circular statistical dispersion is the result obtained by subtracting the circular statistical consistency from one.

[0300] The natural constant is a constant used in mathematics for exponential operations, with a value of approximately 2.71828.

[0301] The exponential decay result of circular statistical dispersion is calculated with the natural constant as the base and the negative value of circular statistical dispersion as the exponent.

[0302] The extermination effectiveness score is a comprehensive quantitative value obtained by multiplying the complementary value of the phase average absolute value with the exponential decay result.

[0303] The compliance threshold is a scoring standard set by humans to determine whether the dynamic consistency of the temperature field meets the standard. It is preferably 0.85, which is a reasonable threshold determined by combining the performance of common extermination equipment, cavity volume and sensor layout.

[0304] The design uses the ratio of the average absolute value of the phase to pi to calculate the complementary value, normalizing the average absolute value of the phase to the range of 0 to 1, thus achieving standardized quantification of the overall phase lag. For example, if the average absolute value of the phase is 0.3 pi, the complementary value is 1 minus 0.3, which equals 0.7, intuitively reflecting the relative magnitude of the lag level.

[0305] The exponential decay method, which uses the natural constant as the base, applies a nonlinear penalty to the phase dispersion, with stronger decay resulting from larger dispersion. For example, when the circular statistical dispersion is 0.1, the exponential decay result is approximately 0.905; when the dispersion is 0.5, the decay result is approximately 0.607, which highlights the impact of uniformity differences on the score.

[0306] The logic of synthesizing the extinction effect score by multiplying the phase complement value and the exponential decay result integrates information from two dimensions: overall hysteresis and spatial uniformity, forming a single quantitative index. This comprehensive scoring method overcomes the limitations of single indicators, comprehensively reflects the dynamic consistency of the temperature field, and facilitates compliance judgment in engineering.

[0307] The compliance threshold is determined by combining the equipment model, the volume of the sterilization chamber, the number of downstream temperature measuring points, and historical validation data. For example, for small incubators with small chamber volumes and densely distributed measuring points, the threshold can be set to 0.8; for large incubators, it can be set to 0.75, ensuring that the threshold is suitable for different scenarios.

[0308] The engineering application logic for the extermination effectiveness score is as follows: when the score is greater than or equal to the compliance threshold, it is determined that the dynamic consistency of the temperature field meets the requirements, the equipment is unlocked and a test report is generated; when the score is less than the compliance threshold, an audible and visual alarm is triggered and the equipment is locked, prompting staff to investigate abnormal temperature fields.

[0309] The method for adapting and adjusting the scoring for different scenarios involves recalibrating the compliance threshold through multiple sets of experimental data when the extermination process temperature differs or the filter model is changed, ensuring the applicability of the scoring. For example, the threshold can be appropriately lowered for high-temperature extermination processes, while the threshold can be raised for low-temperature processes.

[0310] The reason for choosing the natural constant as the base for exponential decay is that the decay characteristics of the natural exponential function are gradual and continuous, which can avoid excessive or insufficient penalty caused by other bases, while meeting the engineering requirements for gradual evaluation of uniformity differences.

Claims

1. A method for quantifying dynamic consistency of temperature field based on thermal field analysis, characterized in that, include: (1) Set at least one temperature measuring point upstream of the filter of the microbial incubator and set multiple temperature measuring points downstream of the filter; (2) During the constant temperature stage of the sterilization process, a set disturbance is applied to the incubator, and the temperature sequence of each temperature measuring point is collected according to the preset sampling parameters; (3) Perform detrending and bandpass filtering on each temperature sequence to obtain the preprocessed temperature sequence; (4) Using frequency domain analysis, calculate the average phase of the preprocessed temperature sequence of the upstream temperature measuring point and the preprocessed temperature sequence of each downstream temperature measuring point in the preset frequency band, and classify each average phase into the preset main value area to form a downstream phase set. (5) Based on the downstream phase set, the average absolute value of the phase and the circular statistical parameters are calculated; (6) Calculate the complementary value of the phase average absolute value, perform attenuation processing on the circular statistical parameters, and calculate the extermination effect score reflecting the temperature uniformity based on the complementary value and the attenuation processing result, thus completing the quantification of the dynamic consistency of the extermination temperature field. In step (4), the frequency domain analysis method is used to calculate the average phase of the preprocessed temperature sequence of the upstream temperature measuring point and the preprocessed temperature sequence of each downstream temperature measuring point within the preset frequency band, as follows: The upstream and downstream temperature measurement point sets are determined based on the positive and negative directions of the directed distances from the temperature measurement points to the neutral plane of the filter. Calculate the average value of the preprocessed temperature sequence corresponding to all upstream temperature measurement points to obtain the upstream reference preprocessed temperature sequence; Discrete Fourier transforms were performed on the upstream reference preprocessed temperature sequence and the preprocessed temperature sequence corresponding to each downstream temperature measurement point, respectively. The cross spectrum corresponding to each downstream temperature measuring point is obtained by conjugate operation, and the argument of each cross spectrum is extracted as the phase of the corresponding downstream temperature measuring point. The preset frequency band is determined based on the lower pass threshold frequency and the upper pass threshold frequency; Filter out the set of discrete frequency points within the preset frequency band; The average phase value of each downstream temperature measuring point is obtained by averaging all phases within the discrete frequency point set for each downstream temperature measuring point. In step (4), the average value of each phase is assigned to a preset principal value region to form a downstream phase set, as follows: The default principal value area is the range from negative pi to positive pi, and the default principal value area includes positive pi; For each downstream temperature measurement point, the phase average value is divided by the sum of the phase average value and pi by twice the pi value, and then rounded down. The product of twice the pi value and the rounded down value is subtracted from the phase average value to obtain the result mapped to the preset principal value area. If the mapped result is a negative pi, then the result is adjusted to a positive pi to obtain the principal value phase corresponding to each downstream temperature measurement point; Collect the principal-valued phases corresponding to all downstream temperature measurement points to form a downstream phase set; Step (5) is as follows: Calculate the average of the absolute values ​​of all principal-valued phases to obtain the average absolute value of the phases; Calculate the average cosine value of each principal-valued phase to obtain the circular statistical cosine average. Calculate the average of the sine values ​​for each principal-valued phase to obtain the circular statistical sine average. The square root of the sum of the squares of the cosine mean and the squares of the sine mean of the circular statistics is obtained to obtain the consistency of the circular statistics. Subtracting the circular statistical consistency from one gives the circular statistical dispersion.

2. The method of claim 1, wherein, In step (1), the direction of airflow from the filter to the downstream is set as positive, based on the neutral plane of the filter; The directed distance from the temperature measuring point to the neutral plane is defined as the shortest distance with a sign, where a positive value indicates that it is on the positive side and a negative value indicates that it is on the negative side. If the directed distance from the temperature measuring point to the neutral plane is negative, then the temperature measuring point is designated as the upstream temperature measuring point; if the directed distance from the temperature measuring point to the neutral plane is positive, then the temperature measuring point is designated as the downstream temperature measuring point.

3. The method of claim 2, wherein the method further comprises: Step (2) is as follows: Determine the sampling time interval and the number of sampling points; The total sampling duration shall be no less than three times the preset disturbance period; The set temperature is the sum of the constant temperature setpoint and the preset disturbance that varies according to a sinusoidal law; where the preset disturbance period is the period that varies according to a sinusoidal law. Apply the set temperature during the constant temperature phase; Determine the discrete sampling time according to the sampling time interval; The total number of temperature measuring points is the sum of the number of upstream temperature measuring points and the number of downstream temperature measuring points; The temperature of each temperature measuring point is collected at each discrete sampling time to form a temperature sequence corresponding to each temperature measuring point.

4. The quantification method for dynamic consistency of temperature field based on thermal field analysis according to claim 3, characterized in that, Step (3) is as follows: Perform cubic polynomial least squares fitting on the temperature sequence corresponding to each temperature measurement point to obtain the trend term, and subtract the corresponding trend term from the temperature sequence to obtain the detrended temperature sequence. The lower bandpass cutoff frequency, upper bandpass cutoff frequency, and filter order are set, and the detrended temperature sequence is processed using a zero-phase bidirectional Butterworth bandpass filter to obtain the preprocessed temperature sequence.

5. The quantification method for dynamic consistency of temperature field based on thermal field analysis according to claim 1, characterized in that, Step (6) is as follows: The complementary value of the phase mean absolute value is calculated, and the circular statistical correlation parameters are attenuated. Based on the complementary value and the attenuation result, a kill effect score reflecting temperature uniformity is calculated, including: Subtract the ratio of the average absolute value of the phase to pi from 1 to obtain the complementary value of the average absolute value of the phase. The results are obtained by applying an exponential decay to the circular statistical dispersion, with the natural constant as the base and the negative value of the circular statistical dispersion as the exponent. Multiplying the complementary value of the phase average absolute value with the result after exponential decay yields a kill effect score that reflects temperature uniformity, thus quantifying the dynamic consistency of the kill temperature field.

Images