Building mortar layering degree detection system and detection process

Abstract

The invention discloses a building mortar layering degree detection system and a detection process, and relates to the technical field of building material detection, the detection system comprises a dynamic pressurization unit used for receiving a control instruction and feeding back a working state, applying pressure and vibration to a mortar sample, and accelerating a layering process; the multi-mode sensor array is used for monitoring physical parameter changes in the mortar layering process in real time by arranging different sensors; the rapid sample loading device is used for loading a mortar sample and automatically cleaning a mortar container; the data processing unit is electrically connected with the dynamic pressurization unit, the multi-mode sensor array and the rapid sample loading device and used for controlling operation of the detection system and processing, analyzing and predicting sensor data. And the detection time is compressed, so that constructors can obtain the detection result of the layering degree of the mortar in a short time.

Description

Technical Field

[0001] The present invention belongs to the technical field of building material testing, and particularly relates to a building mortar bleeding rate detection system and a detection process. Background Art

[0002] During the building construction process, the bleeding rate of mortar is one of the key indicators to measure its quality. The traditional method for detecting the bleeding rate of mortar uses the static method. It is necessary to load the mortar into a bleeding rate cylinder and let it stand for 30 minutes, and then determine the bleeding rate by measuring the consistency difference between the upper and lower layers of mortar. This method has a long detection time and is difficult to meet the requirements of rapid quality monitoring at the construction site, which may lead to delays in the construction progress. The traditional detection of the mortar bleeding layer does not fundamentally solve the problem of long standing time. With the development of the construction industry, the requirements for the efficiency and accuracy of mortar quality detection are increasing day by day. Therefore, we propose a building mortar bleeding rate detection system and a detection process. Summary of the Invention

[0003] To solve the above technical problems, the present invention is realized through the following technical solutions:

[0004] The present invention is a building mortar bleeding rate detection system, and the detection system includes:

[0005] Dynamic pressure application unit: used to receive control instructions and feedback the working state, and used to apply pressure and vibration to the mortar sample to accelerate the layering process;

[0006] Multimodal sensor array: used to set different sensors to monitor the changes in physical parameters during the mortar layering process in real time;

[0007] Quick sample loading device: used to load the mortar sample and automatically clean the mortar container;

[0008] Data processing unit: The data processing unit is electrically connected to the dynamic pressure application unit, the multimodal sensor array and the quick sample loading device respectively, and is used to control the operation, process, analyze and predict the sensor data of the detection system.

[0009] As a preferred technical solution, the dynamic pressure application unit specifically includes:

[0010] Pressure regulation module: used to control the hydraulic device to apply controllable pressure, adjust the pressure according to the instruction, and the calculation formula of the adjusted pressure P is as follows:

[0011]

[0012] In the formula, the initial pressure is set as P0, the current mass of the mortar sample is M, the current ambient temperature is T, the standard ambient temperature is T0, and the pressure adjustment coefficients are k1 and k2 respectively;

[0013] Vibration assistance module: It is used to integrate low-frequency micro-vibrations of 5 - 20 Hz to simulate the transportation vibration environment. The calculation formula for the vibration frequency f is as follows:

[0014] f = f0 + α × (LI - LI0);

[0015] In the formula, the initial set frequency is f0, the current mortar fluidity index is LI, the standard fluidity index is LI0, and the frequency adjustment coefficient is α.

[0016] As a preferred technical solution, the multi-modal sensor array specifically includes:

[0017] Distributed capacitance sensor: It is used to monitor the change of mortar dielectric constant to reflect the moisture distribution. The formula for the moisture content W is as follows:

[0018] W = a0 + a1 + ∈ + a2 × ∈2;

[0019] The conversion from the dielectric constant ∈ to the moisture content W is achieved through the moisture content W formula, and the experimental calibration coefficients are a0, a1, and a2;

[0020] Ultrasonic probe: It is used to vertically emit ultrasonic pulses and calculate the density gradient at different heights through the echo time difference. The calculation formula for the density change Δρ is as follows:

[0021]

[0022] In the formula, the coefficient related to the mortar characteristics is β, the propagation speed of ultrasonic waves in the mortar is υ, the echo time difference is Δt, and the measurement height interval is h;

[0023] Fiber Bragg grating strain sensor: It is used to detect the container deformation and invert the internal stress distribution.

[0024] As a preferred technical solution, the rapid sample loading device specifically includes:

[0025] Quantitative filling module: It is used to precisely control the amount of mortar loaded through a screw conveyor. The formula for the screw conveyor speed n is as follows:

[0026]

[0027] In the formula, the preset filling amount is V0, the current filled amount is V, the remaining filling time is t, and the proportionality coefficient is k;

[0028] Self-cleaning container: The inner wall is coated with a hydrophobic coating and is automatically cleaned after the detection is completed.

[0029] As a preferred technical solution, the data processing unit specifically includes:

[0030] Stratification degree prediction model: Used to predict the stratification degree of mortar by running a model based on time series analysis, normalize the sensor data, and normalize the processed data The formula is as follows:

[0031]

[0032] In the formula, the sensor data is xi; the minimum value in the sensor data is min(x), and the maximum value in the sensor data is max(x);

[0033] According to the normalized processed data The formula for predicting the stratification degree L is as follows:

[0034]

[0035] In the formula, the weight is wj, the bias term is b, the number of data items participating in the calculation is n, and the weights and bias terms are obtained through training with experimental data;

[0036] Adaptive calibration module: Used to dynamically adjust the pressure parameters according to the mortar type. The formula for the adjusted pressure pn is as follows:

[0037] pn = po × F(T);

[0038] In the formula, the pressure before adjustment is po, and the pressure adjustment factor determined according to the mortar type T is F(T);

[0039] Sample type identification module: Electrically connected to the adaptive calibration module, used to identify the type of mortar sample, and provide data for the adaptive calibration algorithm module through AI image recognition technology; Sample identification type module:

[0040] Data acquisition module: Electrically connected to the multi-modal sensor array, responsible for receiving the data collected by the distributed capacitance sensor, ultrasonic probe, and fiber Bragg grating strain sensor, and providing data for subsequent analysis and prediction;

[0041] Result analysis module: Electrically connected to the data processing module, used to receive the prediction results of the stratification degree prediction model, combine the sensor data, calculate the confidence level of the prediction results, and compare the deviation between the predicted value and the historical similar data.

[0042] The present invention is a detection process for the stratification degree of building mortar. The detection process includes the following steps:

[0043] Step S1: Put the mortar sample into the rapid loading device. The data processing unit identifies the mortar type, matches the pressure, vibration frequency, and detection time threshold from the pre-stored parameter table, sends control instructions to the dynamic pressure application module and the rapid loading device. The quantitative filling module of the rapid loading device controls the amount of mortar loaded according to the instructions, and uses pressure feedback closed-loop control to control the filling density;

[0044] Step S2: The dynamic pressure unit starts according to the instructions of the data processing unit, applies a set pressure and vibration frequency, and the multi-modal sensor array synchronously collects capacitance, ultrasonic, and strain data at a sampling frequency of 100 Hz and transmits it to the data processing unit;

[0045] Step S3: After receiving the sensor data, the data processing unit optimizes the weights and bias terms of the slump prediction model through the particle swarm algorithm;

[0046] Initialize the particle swarm: Set the number, position, and velocity of the particles. Each particle represents a set of model parameters, the position represents the value of the parameters, and the velocity represents the update direction and step size of the parameters;

[0047] Calculate the fitness: Substitute each set of parameters into the slump prediction model, combine with the currently collected sensor data for prediction, and use the error between the prediction result and the actual slump of the known sample as the fitness function value. The smaller the error, the higher the fitness;

[0048] Update the particle position and velocity: According to the historical best position of the particle itself and the global best position of the group, update the formula to adjust the position and velocity of the particle. The velocity update formula υi(t + 1) is as follows:

[0049] υi(t + 1) = w×υi(t) + c1r1(t)[pi - xi(t)] + c2r2(t)[pg - xi(t)];

[0050] The position update formula xi(t + 1) is as follows:

[0051] xi(t + 1) = xi(t) + υi(t + 1);

[0052] In the formula, the current position and velocity of particle i are υi(t) and xi(t) respectively, the inertia weight is w, the learning factors are c1 and c2 respectively, the random numbers within [0, 1] are r1(t) and r2(t) respectively, the historical best position of the particle is pi, and the global best position of the group is pg;

[0053] Iteration termination: Repeat the above steps until the preset number of iterations is reached or the fitness value converges, and finally output the optimized model parameters;

[0054] Step S4: Use the optimized slump prediction model, receive the sensor data in the first 3 minutes for prediction, calculate the confidence of the prediction result. When the confidence is greater than 95%, terminate the detection in advance and directly output the predicted slump result; if the confidence is insufficient, continue to apply pressure;

[0055] Step S5: The data processing unit generates a test report, which is displayed on a display device and stored on a storage device. After the test is completed, the self-cleaning container of the rapid sample loading device is automatically cleaned, and the data processing unit calibrates the multimodal sensor array to prepare for the next test.

[0056] The present invention has the following beneficial effects:

[0057] The present invention uses a dynamic pressurizing unit and a hydraulic device to apply controllable pressure and integrate low-frequency micro-vibration to simulate natural settlement and accelerate the mortar stratification process, thereby compressing the detection time. This allows construction personnel to obtain the test results of the mortar stratification degree in a short time and evaluate the mortar quality in time, thereby quickly adjusting the construction plan, significantly improving construction efficiency, ensuring that the project can be smoothly carried out according to the scheduled schedule, and effectively reducing the additional costs caused by excessively long detection time.

[0058] The present invention adopts a multimodal sensor array, and distributed capacitance sensors, ultrasonic probes and fiber Bragg grating strain sensors work together to monitor the changes in physical parameters of mortar during the stratification process in real time from different angles. The distributed capacitance sensor can accurately monitor the changes in the dielectric constant of the mortar, thereby reflecting the moisture distribution; the ultrasonic probe calculates the density gradient at different heights by measuring the echo time difference; the fiber Bragg grating strain sensor detects the deformation of the container and inverts the internal stress distribution. These multi-source data are transmitted to the intelligent data processing terminal, combined with the stratification prediction model, and optimized to improve the detection accuracy and control the error within a smaller range.

[0059] The adaptive calibration module of the present invention can automatically adjust the detection parameters according to the mortar type determined by the sample type identification module. For high-fluidity mortar, due to its own poor stability, the adaptive calibration algorithm module will automatically adjust the pressure control module to limit the pressure within an appropriate range to avoid damage to the mortar structure caused by overpressure; for mortar containing special additives, the vibration frequency parameters can also be adjusted accordingly to ensure the accuracy of the detection. This high adaptability makes the present invention suitable for various construction scenarios.

[0060] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0062] Figure 1A building mortar slump test system and a test process flow chart according to the present invention. Specific embodiments

[0063] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0064] Please refer to Figure 1 As shown, the present invention is a building mortar slump test system, and the test system includes:

[0065] A dynamic pressure application unit: used to receive control instructions and feedback the working state, and used to apply pressure and vibration to the mortar sample to accelerate the layering process;

[0066] A multi-modal sensor array: used to set different sensors to monitor the changes of physical parameters in the mortar layering process in real time;

[0067] A rapid sample loading device: used to load the mortar sample and automatically clean the mortar container;

[0068] A data processing unit: The data processing unit is electrically connected to the dynamic pressure application unit, the multi-modal sensor array and the rapid sample loading device respectively, and is used to control the operation, processing, analysis and prediction of the sensor data of the test system.

[0069] The dynamic pressure application unit specifically includes:

[0070] A pressure regulation module: used to control the hydraulic device to apply a controllable pressure, adjust the pressure according to the instruction, and the calculation formula of the adjusted pressure P is as follows:

[0071]

[0072] In the formula, the initial pressure is set as P0, the current mass of the mortar sample is M, the current ambient temperature is T, the standard ambient temperature is T0, and the pressure adjustment coefficients are k1 and k2 respectively;

[0073] A vibration assistance module: used to integrate low-frequency micro-vibrations of 5-20HZ to simulate the transportation vibration environment, and the calculation formula of the vibration frequency f is as follows:

[0074] f = f0 + α × (LI - LI0);

[0075] In the formula, the initial set frequency is f0, the current mortar fluidity index is LI, the standard fluidity index is LI0, and the frequency adjustment coefficient is α.

[0076] The multimodal sensor array specifically includes:

[0077] Distributed capacitance sensor: used to monitor the change of dielectric constant of mortar to reflect the moisture distribution. The formula for moisture content W is as follows:

[0078] W = a0 + a1 + ∈ + a2×∈2;

[0079] The conversion from dielectric constant ∈ to moisture content W is achieved through the moisture content W formula, and the experimental calibration coefficients are a0, a1, and a2;

[0080] Ultrasonic probe: used to vertically emit ultrasonic pulses, and calculate the density gradient at different heights through the time difference of echoes. The calculation formula for the density change Δρ is as follows:

[0081]

[0082] In the formula, the coefficient related to the mortar characteristics is β, the propagation speed of ultrasonic waves in the mortar is υ, the time difference of echoes is Δt, and the measurement height interval is h;

[0083] Fiber Bragg grating strain sensor: used to detect the deformation of the container and invert the internal stress distribution.

[0084] The rapid sample loading device specifically includes:

[0085] Quantitative filling module: used to precisely control the amount of mortar loaded through a screw conveyor. The rotation speed n of the screw conveyor is as follows:

[0086]

[0087] In the formula, the preset filling amount is V0, the current filled amount is V, the remaining filling time is t, and the proportionality coefficient is k;

[0088] Self-cleaning container: the inner wall is coated with a hydrophobic coating and is automatically cleaned after the detection is completed.

[0089] The data processing unit specifically includes:

[0090] Stratification degree prediction model: used to predict the stratification degree of mortar by running a model based on time series analysis, and perform normalization processing on the sensor data. The normalized data The formula is as follows:

[0091]

[0092] In the formula, the sensor data is xi; the minimum value in the sensor data is min(x), and the maximum value in the sensor data is max(x);

[0093] According to the normalized data The formula for predicting the stratification degree L is as follows:

[0094]

[0095] Wherein, the weight is wj, the bias term is b, the number of data items participating in the calculation is n, and the weight and the bias term are obtained through training with experimental data;

[0096] Adaptive calibration module: used to dynamically adjust the pressure parameter according to the mortar type. The adjusted pressure pn formula is as follows:

[0097] pn = po × F(T);

[0098] Wherein, the pressure before adjustment is po, and the pressure adjustment factor determined according to the mortar type T is F(T);

[0099] Sample type identification module: electrically connected to the adaptive calibration module, used to identify the type of mortar sample, and provide data for the adaptive calibration algorithm module through AI image recognition technology;

[0100] Data acquisition module: electrically connected to the multi-modal sensor array, responsible for receiving the data collected by the distributed capacitance sensor, ultrasonic probe and fiber Bragg grating strain sensor, and providing data for subsequent analysis and prediction;

[0101] Result analysis module: electrically connected to the data processing module, used to receive the prediction result of the slump prediction model, combine the sensor data, calculate the confidence level of the prediction result, and compare the deviation between the predicted value and the historical similar data.

[0102] A specific application of this embodiment is: The present invention is a building mortar slump detection process, and the detection process includes the following steps:

[0103] Step S1: Put the mortar sample into the rapid loading device. The data processing unit identifies the mortar type, matches the pressure, vibration frequency and detection time threshold from the pre-stored parameter table, and sends control instructions to the dynamic pressure application module and the rapid loading device. The quantitative filling module of the rapid loading device controls the mortar loading amount according to the instructions, and uses pressure feedback closed-loop control to fill the density;

[0104] Step S2: The dynamic pressure application unit is started according to the instructions of the data processing unit, applies the set pressure and vibration frequency, and the multi-modal sensor array synchronously collects capacitance, ultrasonic and strain data at a sampling frequency of 100 Hz and transmits it to the data processing unit;

[0105] Step S3: After receiving the sensor data, the data processing unit optimizes the weights and bias terms of the slump prediction model through the particle swarm algorithm;

[0106] Initialize the particle swarm: Set the number, position, and velocity of the particles. Each particle represents a set of model parameters. The position represents the value of the parameters, and the velocity represents the update direction and step size of the parameters;

[0107] Calculate the fitness: Substitute each set of parameters into the stratification degree prediction model, and combine the currently collected sensor data for prediction. Use the error between the prediction result and the actual stratification degree of the known samples as the fitness function value. The smaller the error, the higher the fitness;

[0108] Update the particle position and velocity: According to the historical best position of the particle itself and the global best position of the group, use the update formula to adjust the position and velocity of the particle. The formula for updating the velocity υi(t+1) is as follows:

[0109] υi(t+1) = w×υi(t) + c1r1(t)[pi - xi(t)] + c2r2(t)[pg - xi(t)];

[0110] The formula for updating the position xi(t+1) is as follows:

[0111] xi(t+1) = xi(t) + υi(t+1);

[0112] In the formula, the current position and velocity of particle i are υi(t) and xi(t) respectively, the inertia weight is w, the learning factors are c1 and c2 respectively, the random numbers within [0,1] are r1(t) and r2(t) respectively, the historical best position of the particle is pi, and the global best position of the group is pg;

[0113] Iteration termination: Repeat the above steps until the preset number of iterations is reached or the fitness value converges. Finally, output the optimized model parameters;

[0114] Step S4: Use the optimized stratification degree prediction model to receive the sensor data of the first 3 minutes for prediction, calculate the confidence level of the prediction result. When the confidence level is greater than 95%, terminate the detection in advance and directly output the predicted stratification degree result; if the confidence level is insufficient, continue to pressurize;

[0115] Step S5: The data processing unit generates a detection report, which is displayed through the display device and saved by the storage device. After the detection is completed, the self-cleaning container of the rapid sample loading device is automatically cleaned, and the data processing unit calibrates the multi-modal sensor array to prepare for the next detection.

[0116] It should be noted that in the above system embodiments, the included units are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for easy distinction and do not limit the protection scope of the present invention.

[0117] In addition, those of ordinary skill in the art can understand that all or part of the steps in the methods of the above embodiments can be completed by instructing relevant hardware through a program, and the corresponding program can be stored in a computer-readable storage medium.

[0118] The preferred embodiments of the present invention disclosed above are only used to help illustrate the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific embodiments described. Obviously, many modifications and variations can be made according to the content of this specification. These embodiments are selected and specifically described in this specification in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can well understand and utilize the present invention. The present invention is only limited by the claims and their full scope and equivalents.

Claims

1. A building mortar stratification detection system, characterized in that: The detection system includes: Dynamic pressurization unit: used to receive control instructions and feedback working status, used to apply pressure and vibration to the mortar sample to accelerate the stratification process; Multimodal sensor array: used to monitor the changes in physical parameters during the mortar stratification process in real time by setting different sensors; Rapid sample loading device: used for filling mortar samples and automatically cleaning mortar containers; Data processing unit: The data processing unit is electrically connected to the dynamic pressurization unit, the multimodal sensor array and the rapid sample loading device, and is used to control the operation of the detection system, process, analyze and predict sensor data.

2. A building mortar stratification detection system according to claim 1, characterized in that: The dynamic pressurization unit specifically comprises: Pressure control module: used to control the hydraulic device to apply controllable pressure and adjust the pressure according to the command. The calculation formula of the adjusted pressure P is as follows: In the formula, the initial pressure is set to P0, the current mortar sample mass is M, the current ambient temperature is T, the standard ambient temperature is T0, and the pressure adjustment coefficients are k1 and k2 respectively; Vibration auxiliary module: used to integrate 5-20HZ low-frequency micro-vibration to simulate the transportation vibration environment. The vibration frequency f is calculated as follows: f = f0 + α × (LI - LI0); In the formula, the initial setting frequency is f0, the current mortar fluidity index is LI, the standard fluidity index is LI0, and the frequency adjustment coefficient is α.

3. A building mortar stratification detection system according to claim 1, characterized in that: The multimodal sensor array specifically includes: Distributed capacitance sensor: used to monitor the change of mortar dielectric constant to reflect the moisture distribution. The moisture content W formula is as follows: W=a0+a1+∈+a2×∈2; The conversion of dielectric constant ∈ to moisture content W is realized through the moisture content W formula, and the experimental calibration coefficients are a0, a1, and a2; Ultrasonic probe: used to emit ultrasonic pulses vertically, and calculate the density gradient at different heights by echo time difference. The density change Δρ is calculated as follows: In the formula, the coefficient related to the mortar characteristics is β, the propagation speed of ultrasound in the mortar is υ, the echo time difference is Δt, and the measurement height interval is h; Fiber Bragg grating strain sensor: used to detect container deformation and invert internal stress distribution.

4. A building mortar stratification detection system according to claim 1, characterized in that: The rapid sample loading device specifically comprises: Quantitative filling module: used to accurately control the amount of mortar loaded through the screw conveyor. The formula for the screw conveyor speed n is as follows: In the formula, the preset filling volume is V0, the current filling volume is V, the remaining filling time is t, and the proportional coefficient is k; Self-cleaning container: The inner wall is coated with a hydrophobic coating and automatically cleans after the test is completed.

5. A building mortar stratification detection system according to claim 1, characterized in that: The data processing unit specifically includes: Delamination prediction model: used to predict the degree of mortar delamination by running a model based on time series analysis, normalizing the sensor data, and normalizing the data The formula is as follows: Wherein, the sensor data is xi; the minimum value in the sensor data is min(x), and the maximum value in the sensor data is max(x); According to the normalization data The formula for predicting the stratification degree L is as follows: In the formula, the weight is wj, the bias term is b, the number of data items involved in the calculation is n, and the weight and bias term are obtained through experimental data training; Adaptive calibration module: used to dynamically adjust pressure parameters according to the mortar type. The pressure pn formula after adjustment is as follows: pn = po × F(T); In the formula, the pressure before adjustment is po, and the pressure adjustment factor determined according to the mortar type T is F(T); Sample type identification module: electrically connected to the adaptive calibration module, used to identify the type of mortar samples and provide data for the adaptive calibration algorithm module through AI image recognition technology; Data acquisition module: It is electrically connected to the multimodal sensor array and is responsible for receiving data collected by distributed capacitance sensors, ultrasonic probes, and fiber Bragg grating strain sensors to provide data for subsequent analysis and prediction; Result analysis module: electrically connected to the data processing module, used to receive the prediction results of the hierarchical prediction model, combine the sensor data, calculate the confidence of the prediction results, and compare the deviation between the prediction value and historical similar data.

6. A construction mortar stratification detection process according to claim 1, characterized in that: The following steps are involved: Step S1, placing the mortar sample into the rapid sample loading device, the data processing unit identifies the mortar type, matches the pressure, vibration frequency and detection time threshold from the pre-stored parameter table, sends a control instruction to the dynamic pressurization module and the rapid sample loading device, and the quantitative filling module of the rapid sample loading device controls the mortar loading amount according to the instruction, and uses the pressure feedback closed loop to control the filling density; Step S2, the dynamic pressurizing unit is started according to the instruction of the data processing unit, and the set pressure and vibration frequency are applied. The multimodal sensor array synchronously collects capacitance, ultrasonic wave, and strain data at a sampling frequency of 100 Hz, and transmits them to the data processing unit; Step S3: After receiving the sensor data, the data processing unit optimizes the weights and bias items of the stratification prediction model by using a particle swarm algorithm; Initialize the particle swarm: set the number, position and speed of particles. Each particle represents a set of model parameters. The position indicates the value of the parameter, and the speed indicates the update direction and step size of the parameter. Calculate fitness: Substitute each set of parameters into the stratification prediction model, combine the currently collected sensor data for prediction, and use the error between the prediction result and the actual stratification of the known sample as the fitness function value. The smaller the error, the higher the fitness. Update particle position and speed: According to the particle's own historical optimal position and the group's global optimal position, the update formula adjusts the particle's position and speed. The speed update formula υi(t+1) is as follows: υi(t+1)=w×υi(t)+c1r1(t)[pi-xi(t)]+c2r2(t)[pg-xi(t)]; The position update formula xi(t+1) is as follows: xi(t+1)=xi(t)+υi(t+1); Where, the current position and velocity of particle i are υi(t) and xi(t), the inertia weight is w, the learning factors are c1 and c2, the random numbers in [0,1] are r1(t) and r2(t), the historical optimal position of the particle is pi, and the global optimal position of the group is pg; Iteration termination: Repeat the above steps until the preset number of iterations is reached or the fitness value converges, and finally output the optimized model parameters; Step S4: using the optimized stratification prediction model, receiving the sensor data of the previous 3 minutes for prediction, calculating the confidence of the prediction result, and when the confidence is greater than 95%, terminating the detection in advance and directly outputting the predicted stratification result; if the confidence is insufficient, continue to pressurize; Step S5: The data processing unit generates a test report, which is displayed on a display device and stored on a storage device. After the test is completed, the self-cleaning container of the rapid sample loading device is automatically cleaned, and the data processing unit calibrates the multimodal sensor array.

Images