Fresh food delivery emergency guarantee method

Abstract

The invention provides a fresh food distribution emergency guarantee method, which comprises the following steps of: if temperature distribution does not meet a preset temperature uniformity threshold requirement and a temperature difference exceeds a threshold, identifying a local area of which the temperature gradient exceeds a preset gradient threshold and is close to the periphery of a vehicle door according to a three-dimensional temperature distribution cloud picture; taking the area as a target area for emphatically adjusting the sealing performance of the fan and the vehicle door; performing airflow field numerical simulation on geometric structure characteristics, fan arrangement positions, different fan blade angles and different opening areas and angles of a vehicle door in a target area and a temperature control cabin by adopting a computational fluid dynamics method; the time, angle and frequency data of opening and closing of a vehicle door are collected in real time through a vehicle door sensor on a fresh delivery vehicle, and whether the current vehicle is in a frequent door opening and closing working condition or not is judged by combining a pre-established vehicle door opening and closing state judgment model.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and in particular to an emergency guarantee method for fresh food delivery. Background Art

[0002] In the temperature-controlled cabin of fresh food delivery vehicles, the angle and speed of the air circulation fan blades are key factors affecting the uniform temperature distribution. However, since the vehicle needs to frequently open and close the doors to load and unload goods during the delivery process, the air flow state in the cabin is constantly changing, and it is difficult for the fan to adjust in time to adapt to this dynamic change. If the fan speed is too low, the temperature difference between different areas in the cabin will be too large; if the speed is too high, the temperature fluctuation when the door is opened and closed will be aggravated. At the same time, if the fan blade angle is unreasonable, it will cause the air flow in the cabin to short-circuit, and the temperature in the local area will be difficult to control. Especially during the peak period of fresh food delivery, such as the eve of holidays, the vehicle opens and closes the door more frequently, which will cause the temperature of the goods in the area around the door to fluctuate greatly, affecting the quality of fresh food. Frequent opening and closing of the door causes a large amount of cold air to be lost, and the temperature in the car compartment rises sharply. After closing the door, it takes a long time for the fan to lower the temperature. In this process, the quality of the goods is easily damaged. In addition to the frequency of door opening and closing, the compression of the door seal is also an important factor affecting the temperature distribution in the area around the door. Insufficient compression of the door seal will cause a gap between the door and the body, causing rapid loss of cold air. However, over-compression of the sealing strip will increase the resistance to opening the door and affect the efficiency of loading and unloading. Therefore, it is necessary to dynamically adjust the compression of the door sealing strip according to different operating conditions, while reducing the loss of cold air and taking into account the convenience of opening the door. How to dynamically optimize the fan blade angle and speed parameters according to the real-time operating status of the vehicle, and coordinately control the compression of the door sealing strip to minimize the temperature disturbance caused by frequent opening and closing of the door and achieve rapid temperature balance in the cabin is a technical problem that needs to be solved urgently. Summary of the invention

[0003] The present invention provides a fresh food delivery emergency guarantee method, which mainly includes:

[0004] Acquire the temperature data collected in real time by multiple temperature sensors arranged in different areas of the temperature control cabin of the fresh food delivery vehicle, obtain a three-dimensional temperature distribution cloud map inside the entire temperature control cabin, and determine whether the temperature distribution of different cargo storage areas in the current temperature control cabin meets the preset temperature uniformity threshold requirements;

[0005] If the temperature distribution does not meet the preset temperature uniformity threshold requirement and the temperature difference exceeds the threshold, the local area near the door where the temperature gradient exceeds the preset gradient threshold is identified based on the three-dimensional temperature distribution cloud map, and it is used as the target area for key adjustment of the fan and door sealing performance;

[0006] Computational fluid dynamics methods are used to numerically simulate the airflow field of the target area, the geometric structure characteristics of the temperature control cabin, the fan layout, different fan blade angles, and different door opening areas and angles;

[0007] Through numerical simulation of the airflow field, the airflow streamline distribution diagram inside the temperature control cabin under different fan blade angles and door opening areas and angles is obtained, the fan angle range and door opening angle threshold that cause airflow short circuit and external air infiltration are identified, and they are excluded from the adjustable range of the fan and door, and the initial speed of the fan and the compression amount of the door sealing strip are determined;

[0008] The door sensors on the fresh delivery vehicles collect the time, angle and frequency data of the door opening and closing in real time, and combine the pre-established door opening and closing state judgment model to determine whether the current vehicle is in a frequent door opening and closing condition;

[0009] If the vehicle is in a condition of frequent door opening and closing, the optimal adjustment parameters corresponding to the current vehicle door opening condition are obtained based on the determined initial speed of the fan and the compression of the door sealing strip, as well as the pre-established mapping model of the fan blade angle, speed and the temperature gradient around the door, and the fan blade angle, speed and the compression of the door sealing strip are adjusted in real time;

[0010] During the fan and door seal adjustment process, the variance of the temperature data in the temperature control chamber near the door before and after the adjustment is calculated in real time. According to the variance change trend, the fan blade angle, speed and door seal compression are optimized online to find the optimal adjustment parameter combination. The fan and door sealing performance are adjusted simultaneously using the optimal adjustment parameter combination.

[0011] A door ambient temperature prediction model based on a long short-term memory network is used to predict the temperature change trend of the area around the door in the future according to the historical opening and closing state data of the door, the thermal conductivity of the door material and the current temperature gradient around the door. The fan and door sealing strip are pre-adjusted in advance according to the predicted temperature change trend to reduce the temperature fluctuation of the cargo around the door caused by frequent opening and closing of the door.

[0012] The technical solution provided by the embodiment of the present invention may have the following beneficial effects:

[0013] The present invention discloses an emergency guarantee method for fresh food delivery. The method collects temperature data inside a temperature control cabin through multiple temperature sensors, generates a three-dimensional temperature distribution cloud map, and determines whether the temperature uniformity meets the requirements. If the temperature distribution is uneven, the target area where the temperature gradient exceeds the threshold is identified, and the numerical simulation of the airflow field is performed using the computational fluid dynamics method to determine the fan and door adjustment parameters. Combined with the door switch state judgment model, the fan blade angle, rotation speed and door seal compression are adjusted in real time. The gradient descent algorithm is used to optimize the adjustment parameters online, and the temperature change trend around the door is predicted based on the long short-term memory network, and pre-adjustment is performed in advance. The present invention effectively improves the temperature uniformity and stability of the temperature control cabin of the fresh food delivery vehicle through multi-dimensional data collection, intelligent algorithm analysis and dynamic adjustment, reduces the temperature fluctuation caused by frequent door opening and closing, and ensures the quality of fresh products. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 The present invention is a flowchart of a fresh food delivery emergency guarantee method.

[0015] Figure 2 It is a schematic diagram of a fresh food delivery emergency guarantee method of the present invention.

[0016] Figure 3 This is another schematic diagram of a fresh food delivery emergency support method of the present invention. DETAILED DESCRIPTION

[0017] In order to further understand the content of the present invention, the present invention is described in detail in conjunction with the accompanying drawings and embodiments. The present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It is understood that the specific embodiments described herein are only used to explain the relevant inventions, rather than to limit the invention. It is also necessary to explain that, for ease of description, only the parts related to the invention are shown in the accompanying drawings.

[0018] like Figure 1-3 In this embodiment, a fresh food delivery emergency guarantee method may specifically include:

[0019] S101. Acquire temperature data collected in real time by multiple temperature sensors arranged in different areas of the temperature control cabin of the fresh food delivery vehicle, obtain a three-dimensional temperature distribution cloud map inside the entire temperature control cabin, and determine whether the temperature distribution of different cargo storage areas in the current temperature control cabin meets the preset temperature uniformity threshold requirement.

[0020] Real-time temperature data collected by multiple temperature sensors in the temperature control cabin are obtained, and a three-dimensional temperature distribution matrix inside the temperature control cabin is constructed according to the temperature data; a trilinear interpolation algorithm is used to estimate the temperature values ​​between sampling points in the three-dimensional temperature distribution matrix to obtain a continuous three-dimensional temperature distribution cloud map inside the temperature control cabin; multiple temperature monitoring units are divided according to different cargo storage areas in the temperature control cabin, and the average temperature value and standard deviation in the temperature monitoring unit are calculated; if the temperature standard deviation of the temperature monitoring unit exceeds a preset temperature uniformity threshold, it is determined that the temperature distribution of the temperature monitoring unit is uneven; the cooling or heating power is calculated according to the temperature deviation and change rate of the temperature monitoring unit, and an adjustment instruction for the temperature control system is generated.

[0021] Exemplarily, the real-time temperature data collected by multiple temperature sensors in the temperature control cabin are obtained, and the three-dimensional temperature distribution matrix inside the temperature control cabin is constructed according to the arrangement position of the sensors, the acquisition timestamp, the driving status of the vehicle and the ambient temperature. The temperature values ​​between the sampling points are estimated by the trilinear interpolation algorithm to obtain a continuous three-dimensional temperature distribution cloud map inside the entire temperature control cabin. According to the different cargo storage areas in the temperature control cabin, multiple temperature monitoring units are divided, and the average temperature value and standard deviation in each monitoring unit are calculated. If the temperature standard deviation of a monitoring unit exceeds the preset temperature uniformity threshold, which is set to 1°C, it is determined that the temperature distribution in the area is uneven, and the location information and temperature fluctuation range of the abnormal area are recorded. According to the cargo type and storage location information, combined with the temperature distribution status of each monitoring unit, the actual temperature deviation value of the environment in which each cargo is located is calculated. The PID control strategy is adopted to calculate the cooling or heating power according to the temperature deviation and the rate of change, generate the adjustment instructions of the temperature control system, and perform directional cooling or heating on the abnormal area until the temperature distribution uniformity of each monitoring unit meets the preset threshold requirements. According to the driving status of the delivery vehicle and the change of ambient temperature, the temperature data collection frequency and the response speed of the temperature control system are dynamically adjusted to ensure the accuracy of temperature monitoring while reducing energy consumption. The LZ77 lossless compression algorithm is used to compress the temperature data, reduce the transmission delay, and realize the real-time monitoring and precise adjustment of the temperature distribution in the temperature control cabin. 20 temperature sensors are arranged in the temperature control cabin to collect temperature data every 5 seconds. According to the sensor coordinates, collection time, vehicle speed and external temperature, a 60x60x60 temperature distribution matrix is ​​constructed. The temperature between the sampling points is estimated using the trilinear interpolation algorithm to generate a continuous three-dimensional temperature cloud map with a resolution of 0.1℃. The temperature control cabin is divided into 100 monitoring units of 10x10x10cm, and the average temperature and standard deviation of each unit are calculated. The temperature uniformity threshold is set to 1℃. If the standard deviation of a unit exceeds 1℃, the temperature distribution in the area is determined to be uneven, and the abnormal location and temperature range are recorded. Assume that three types of goods are stored in the temperature control cabin: Class A 0-4℃, Class B 4-8℃, and Class C 8-12℃. According to the cargo location and temperature distribution, the actual temperature deviation of each cargo is calculated. The PID control strategy is adopted, and the proportional coefficient Kp=0.5, the integral coefficient Ki=0.1, and the differential coefficient Kd=0.05 are set. According to the temperature deviation e(t) and the rate of change de(t) / dt, the refrigeration or heating power u(t)=Kp*e(t)+Ki*∫e(t)dt+Kd*de(t) / dt is calculated. The adjustment instructions are generated, such as "the power of refrigeration unit 1 increases by 20%", and the abnormal area is precisely adjusted. According to the vehicle speed and external temperature changes, the acquisition frequency is dynamically adjusted. For example, when the vehicle speed>60km / h or the external temperature changes>5℃ / h, the acquisition interval is shortened to 3 seconds. Using the LZ77 lossless compression algorithm, the temperature data compression rate is increased to 50%, reducing transmission delays.

[0022] S102. If the temperature distribution does not meet the preset temperature uniformity threshold requirement and the temperature difference exceeds the threshold, a local area where the temperature gradient exceeds the preset gradient threshold and is close to the door is identified based on the three-dimensional temperature distribution cloud map, and is used as a target area for focusing on adjusting the fan and door sealing performance.

[0023] Obtain a three-dimensional temperature distribution cloud map and use the central difference method to calculate the temperature gradient between adjacent grid points. Determine whether the temperature gradient exceeds the preset threshold and is located in the area around the door; if the conditions are met, mark the area as a local temperature abnormality area. Use a clustering algorithm to group the marked local temperature abnormality areas; calculate the average temperature, area, and center of gravity position of each abnormal area. According to the internal structure information of the car, determine the target area for key adjustments to the fan and door sealing performance;

[0024] Exemplarily, according to the acquired three-dimensional temperature distribution cloud map, the temperature gradient between adjacent grid points is calculated by the central difference method. If the temperature gradient exceeds the preset threshold and is located in the area around the door, the area is marked as a local temperature abnormal area. The range of the area around the door is dynamically determined according to the size of the car body, usually 10% of the length of the car body extending inward from the edge of the door. For the marked local temperature abnormal area, the K-means clustering algorithm is used to group the abnormal points, and the average temperature, area and center of gravity position of each abnormal area are calculated. Combined with the internal structure information of the car body, the target area for key adjustment of the fan and door sealing performance is determined. According to the temperature abnormality characteristics of the target area, the fan adjustment model based on the decision tree algorithm is called, which is trained by historical temperature data and adjustment effects. Input parameters such as temperature gradient, area and position of the abnormal area, and output adjustment instructions such as fan speed and angle. At the same time, a preliminary adjustment suggestion for the door sealing strip pressure is generated. The fan adjustment instruction is executed by the temperature control system to continuously monitor the temperature change in the target area. If the temperature gradient still exceeds the threshold, the door sealing performance optimization program is triggered. The program first detects the pressure distribution of the sealing strip, and then adjusts the pressure of each section of the sealing strip according to the size and direction of the temperature gradient. The specific adjustment method is to increase the sealing strip pressure in the area with a larger temperature gradient and reduce the sealing strip pressure in the area with a smaller temperature gradient. This process is repeated until the temperature gradient in the target area drops below the threshold or reaches the preset upper limit of the number of adjustments. 50 temperature sensors are arranged in a refrigerated compartment with a length of 10 meters, a width of 3 meters, and a height of 2.5 meters. Temperature data is collected every 2 seconds to construct a 100x30x25 three-dimensional temperature distribution cloud map. The temperature gradient is calculated using the central difference method, and the gradient threshold is set to 2℃ / m. The area around the door is defined as within 1 meter of the door. If the temperature gradient of a point exceeds 2℃ / m and is within 1 meter of the door, it is marked as an abnormal point. The K-means clustering algorithm is used, where K=3 to group the abnormal points, and the average temperature, area, and center of gravity of each group are calculated. For example, three abnormal areas are identified as Area A (4.5℃, 0.5㎡, 0.3m from the door), Area B (6.2℃, 0.3㎡, 0.7m from the door), and Area C (3.8℃, 0.4㎡, 0.5m from the door). The fan adjustment model based on the decision tree inputs these parameters and outputs adjustment instructions including fan speed of 1500rpm and angle of 30° in Area A, 1800rpm and 45° in Area B, and 1200rpm and 20° in Area C. At the same time, the sealing strip pressure adjustment recommendation is generated to increase by 20% in Area A, 15% in Area B, and 10% in Area C. After the fan adjustment is executed, the temperature change is continuously monitored. After 5 minutes, the temperature gradient of Area A and Area C drops below the threshold, and Area B still exceeds the standard, triggering the sealing performance optimization program. The sealing strip pressure in Area B is detected to be 0.2MPa, which is lower than the standard 0.3MPa. The program gradually increases the sealing strip pressure in Area B to 0.35MPa, increasing by 0.05MPa each time and waiting for 30 seconds to observe the effect.After two adjustments, the temperature gradient in zone B dropped to 1.8°C / m, which was lower than the threshold, and the optimization was completed.

[0025] S103. Use computational fluid dynamics methods to perform numerical simulation of the airflow field for the target area, geometric structural features inside the temperature control cabin, fan layout positions, different fan blade angles, and different door opening areas and angles.

[0026] The internal geometric structure characteristics of the temperature control cabin, the fan layout position and the door opening state parameters are obtained, and a digital model of the temperature control cabin is constructed according to the parameters, and local mesh encryption is performed on the target area identified in the digital model; the calculation domain of the digital model is discretized, and if there is a complex geometric shape in the calculation domain, the complex geometric shape is adaptively processed; the boundary conditions of the calculation domain are set, and the boundary conditions include inlet wind speed, outlet pressure and wall no-slip condition; the control equation is discretized by using the finite volume method to solve the velocity field and the pressure field, and the solution of the pressure field is achieved by solving the Poisson equation; the airflow field distribution under different working conditions is simulated according to the set fan blade angle and the door opening state parameters; it is judged whether the simulation result converges, and if the residual of the simulation result is less than a preset threshold, it is determined that the simulation result converges; the converged simulation result is post-processed to generate an airflow velocity vector diagram, a temperature value line diagram, a streamline diagram and a pressure cloud diagram; the airflow velocity, temperature distribution and pressure gradient data of the target area are extracted from the post-processing results; the extracted data is subjected to dimensionality reduction processing to obtain the main factors affecting the airflow field distribution in the target area.

[0027] For example, according to the internal geometric structure characteristics of the temperature control cabin, the fan layout position, the door opening state and other parameters, the digital model of the temperature control cabin is constructed using AutoCAD software, and the local mesh is encrypted for the identified target area. The entire calculation domain is discretized using tetrahedron and hexahedron hybrid grids to adapt to complex geometric shapes. The boundary conditions of the calculation domain are set, including inlet wind speed, outlet pressure, wall no-slip conditions, etc. The κ-ε turbulence model and the k-ωSST model are selected to describe the airflow motion characteristics in the high Reynolds number and low Reynolds number regions, respectively. The control equations are discretized using the finite volume method, and the velocity field and pressure field are solved using the SIMPLE algorithm, where the pressure correction process is realized by solving the Poisson equation. Through parameterized settings, the airflow field distribution under different fan blade angles of 0° to 90°, intervals of 10°, and different door opening areas of 0% to 100%, intervals of 10% and angles of 0° to 90°, intervals of 10° are simulated. Each working condition is iterated until convergence, and the convergence standard is set to a residual error of less than 10^-4. The simulation results were post-processed, and visual graphics such as airflow velocity vector diagrams, temperature contour diagrams, streamline diagrams, and pressure cloud diagrams were generated using ParaView software. Key data such as airflow velocity, temperature distribution, and pressure gradient in the target area were extracted using data mining technology. The principal component analysis method was used to reduce the dimension of multiple sets of simulation results and identify the main factors affecting the airflow field distribution in the target area. An airflow field feature database was generated to provide a quantitative basis for subsequent fan control strategy optimization and door opening management. In a refrigerated compartment with a length of 12 meters, a width of 2.5 meters, and a height of 3 meters, an accurate three-dimensional digital model was constructed using AutoCAD software, including 6 fans and 1 double door. For the target area within 1 meter of the door, the grid size was set to 5 mm, and for other areas to 20 mm, a total of about 5 million grid cells were generated. The inlet wind speed was set to 5 m / s, the outlet pressure was standard atmospheric pressure, and the wall surface adopted a no-slip condition. The k-ωSST model was used near the target area, and the κ-ε model was used in other areas. The relaxation factor of the pressure correction equation in the SIMPLE algorithm was set to 0.3. Ten fan blade angles of 0° to 90°, with an interval of 10°, and 11 door opening states of 0% to 100%, with an interval of 10%, for a total of 110 working conditions were simulated. Each working condition was iterated 2000 times or stopped when the residual was less than 10^-4. ParaView software was used to generate visualization graphics, such as velocity vector diagrams on 100×100×100 uniform grid points in the target area. Key parameters were extracted through data mining, such as the average wind speed range of 0.5 to 2.5 m / s in the target area, the standard deviation of temperature uniformity <1℃, and the pressure gradient <10Pa / m. The principal component analysis method was used to reduce the dimensionality of the 110 sets of simulation results to 3 principal components, explaining 95% of the total variance. Finally, an airflow field feature database containing 11,000 data points was generated, each of which contained 15 key parameters of fan angle, door opening, and target area.

[0028] S104. By numerically simulating the airflow field, the airflow streamline distribution diagram inside the temperature control cabin under different fan blade angles and door opening areas and angles is obtained, the fan angle range and door opening angle threshold that cause airflow short circuit and external air infiltration are identified, and they are excluded from the adjustable range of the fan and the door, and the initial speed of the fan and the compression amount of the door sealing strip are determined.

[0029] The numerical simulation results of the airflow field carrying the fan blade angle and the door opening state are received, and streamlines are traced according to the numerical simulation results to obtain an airflow streamline distribution map; the Canny edge detection algorithm is used for the airflow streamline distribution map to obtain streamline feature data including streamline density, curvature and length; based on the streamline feature data, a clustering algorithm is used to determine the characteristic mode of airflow short circuit and external air infiltration, wherein the airflow short circuit judgment standard is that the streamline closure degree is greater than a preset closure threshold and the duration exceeds a preset time threshold, and the external air infiltration judgment standard is that the streamline density near the door changes suddenly and the direction is reversed; based on the characteristic mode of airflow short circuit and external air infiltration, a multivariate linear regression method is used to determine the mathematical model of the constraint relationship between the fan angle and the door opening angle; based on the mathematical model of the constraint relationship, a random forest algorithm is used to determine the prediction model of the initial speed of the fan and the compression amount of the door sealing strip, and the input parameters of the prediction model include the temperature difference between the inside and outside of the temperature control cabin, the type and load of cargo, and the internal structural characteristics of the compartment, and the internal structural characteristics of the compartment include the location of obstacles and the distribution of cargo.

[0030] For example, according to the numerical simulation results of the airflow field, the Runge-Kutta method is used to track the streamlines, and the distribution diagrams of the airflow streamlines under different fan blade angles and door opening states are generated. The streamline features, including streamline density, curvature and length, are extracted by the Canny edge detection algorithm. The streamline features are classified by the K-means clustering algorithm to identify the characteristic patterns of airflow short circuit and external air infiltration. The airflow short circuit judgment standard is set as the streamline closure degree is greater than 80% and the duration is more than 5 seconds. The external air infiltration judgment standard is the sudden change of streamline density near the door and the reversal of direction. Through traversal analysis, the fan angle range that causes airflow short circuit and the door opening angle threshold that causes external air infiltration are determined, and the multivariate linear regression method is used to establish the mathematical model of the constraint relationship between the fan angle and the door opening angle. The constraint relationship is applied to the optimization of the adjustable range to generate a list of optimized fan angle and door opening angle combinations. Based on the optimized adjustable range, the random forest algorithm is used to train the prediction model of the initial speed of the fan and the compression of the door seal. The input parameters include the temperature difference between the inside and outside of the temperature control cabin, the type and load of cargo, and the internal structural characteristics of the compartment, such as the location of obstacles and the distribution of cargo. The initial speed of the fan and the compression of the door seal are output to achieve the initial optimization configuration of the internal environment of the temperature control cabin. In a refrigerated compartment with a length of 12 meters, a width of 2.5 meters, and a height of 3 meters, the Runge-Kutta fourth-order method is used for streamline tracking. The time step is 0.01 seconds, and a total of 10,000 steps are calculated to generate a streamline distribution map on a 100×100×100 grid point. The Canny edge detection algorithm is used with the low threshold set to 50 and the high threshold set to 150 to extract streamline features. The K-means clustering algorithm is set with k=5 and the maximum number of iterations to 100, and the streamline features are divided into 5 categories. Airflow short circuit is determined when more than 80% of the streamlines return to the vicinity of the starting point within 5 seconds, and the distance is less than 0.5 meters; external air infiltration is determined when the density of streamlines within 1 meter of the door increases by 50% and the direction is opposite to the main flow direction. Traversing the fan angle of 0°-90°, step size 5° and the door opening angle of 0°-60°, step size 5°, the constraint relationship is Y=0.8X+15, where Y is the maximum allowable door opening angle and X is the fan angle. The adjustable range after optimization is 15°-75° for the fan angle and 0°-45° for the door opening angle. The random forest algorithm uses 100 decision trees with a maximum depth of 10, inputs 8 features, including internal and external temperature difference, 3 types of cargo, loading rate, 2 obstacle position parameters, cargo distribution uniformity, and outputs the initial fan speed range of 500-2000rpm and the sealing strip compression range of 2-10mm. The model is trained on 1000 sets of historical data, with an average absolute error of less than 50rpm and 0.5mm.

[0031] S105. The door sensor on the fresh delivery vehicle collects the time, angle and frequency data of the door opening and closing in real time, and combines it with the pre-established door opening and closing state judgment model to judge whether the current vehicle is in a frequent door opening and closing condition.

[0032] The door opening and closing time and angle data collected by the door sensor are obtained, and the collection frequency of the door sensor is a preset value; the Kalman filter algorithm is applied to the collected data, and the measurement noise covariance of the Kalman filter algorithm is set to a preset value, and the process noise covariance is set to a preset value; the number of door openings, the average opening time and the average opening angle are determined, and the number of door openings, the average opening time and the average opening angle are calculated within a time window dynamically adjusted according to the vehicle operation status; the door opening and closing state is judged by inputting the number of door openings, the average opening time and the average opening angle into a pre-trained support vector machine model, and the support vector machine model adopts a radial basis function kernel; if the probability value output by the support vector machine model exceeds a preset threshold for a preset number of time windows after exponential moving average smoothing, it is determined that the vehicle is in a frequent door opening and closing condition.

[0033] Exemplarily, the door sensor is used to collect the time and angle data of the door opening and closing in real time, and the data collection frequency is set to 10 times per second. The collected raw data is processed by Kalman filtering, and the measurement noise covariance R is set to 0.1, and the process noise covariance Q is set to 0.01, so as to obtain a smooth door state change curve. According to the processed door state change curve, the number of door openings, the average opening time and the average opening angle per unit time are calculated, and the time window length is dynamically adjusted according to the vehicle operation state, which is 3 minutes when parking, 5 minutes when driving at low speed, and 10 minutes when driving at high speed. The time window is slid once every 30 seconds to obtain the characteristic vector of the door opening and closing behavior. The calculated characteristic vector is input into the pre-trained door opening and closing state judgment model, which adopts the support vector machine algorithm, uses the radial basis function kernel, the kernel parameter γ is set to 0.1, and the penalty parameter C is set to 1, and the probability value of the vehicle being in the frequent door opening and closing condition in the current time window is output. The probability value output by the support vector machine is smoothed by exponential moving average, the smoothing coefficient α is set to 0.2, and the judgment threshold of frequent door opening and closing conditions is set to 0.8. If the smoothed probability value exceeds the threshold for three consecutive time windows, it is determined that the current vehicle is in a frequent door opening and closing condition, triggering the corresponding temperature control strategy adjustment mechanism. A high-precision angle sensor is installed on a distribution refrigerated truck, and the acquisition frequency is set to 10Hz to record the door opening and closing status. Assume that in a 30-minute distribution process, the original data shows that the door is opened and closed 35 times, and the average opening angle is 75 degrees. The Kalman filter algorithm is used to process the data, and the measurement noise covariance R is set to 0.1, and the process noise covariance Q is set to 0.01. After filtering, the number of effective switches is reduced to 32 times, and the average opening angle is corrected to 72 degrees. The vehicle operation status is judged based on GPS data, and it is found that the distribution process includes 5 minutes of parking and 15 minutes of low-speed driving.

[0034] <30km / h and 10 minutes of high-speed driving ≥30km / h. Accordingly, three time windows are set as 3 minutes for parking, 5 minutes for low speed, and 10 minutes for high speed. Calculate the feature vector, including the number of door openings per minute, the average opening time, and the average opening angle. Input the feature vector into the pre-trained support vector machine model, which uses a radial basis function kernel with a kernel parameter γ of 0.1 and a penalty parameter C of 1. The model output shows that the probability of frequent door opening and closing during parking is 0.9, 0.7 when driving at low speed, and 0.3 when driving at high speed. Apply exponential moving average to these probability values ​​with a smoothing coefficient α of 0.2 to obtain a smoothed probability sequence. Set the judgment threshold to 0.8, and find that the probability values ​​of 9 minutes in 3 consecutive time windows during parking exceed the threshold. It is determined that the vehicle is in a frequent door opening and closing condition, triggering the adjustment of the temperature control strategy.

[0035] S106. If the vehicle is in a condition of frequent door opening and closing, the optimal adjustment parameters corresponding to the current vehicle door opening condition are obtained based on the determined initial fan speed and the compression of the door sealing strip, as well as the pre-established mapping model of the fan blade angle, speed and the temperature gradient around the door, and the fan blade angle, speed and the compression of the door sealing strip are adjusted in real time.

[0036] The characteristic parameters of the vehicle under the condition of frequent door opening and closing are obtained, and the characteristic parameters include the frequency of door opening and closing, the duration of door opening and the rate of change of cabin temperature; the fan speed, blade angle and the compression of the door sealing strip are collected, and the temperature sensors distributed around the door collect temperature data; the temperature gradient is calculated according to the characteristic parameters, fan speed, blade angle, sealing strip compression and temperature data; the characteristic parameters, fan speed, blade angle, sealing strip compression and temperature gradient are input into a pre-trained multi-layer perceptron neural network; the neural network outputs the optimal fan blade angle, optimal fan speed and optimal door sealing strip compression under the current vehicle's door opening condition; the fan blade angle is adjusted according to the optimal fan blade angle, the fan speed is adjusted according to the optimal fan speed, and the door sealing strip compression is adjusted according to the optimal door sealing strip compression.

[0037] Exemplarily, characteristic parameters of the current vehicle under frequent door opening and closing conditions, including door opening and closing frequency, door opening duration and cabin temperature change rate, are obtained as basic data for subsequent adjustments. The current fan speed, blade angle and door seal compression are collected. At the same time, temperature sensors distributed around the door are used to collect temperature data once per second, and the temperature gradient around the door is calculated using the central difference method. The collected characteristic parameters, fan parameters, seal parameters and temperature gradient data are used as input variables and input into a pre-trained multi-layer perceptron neural network, which includes an input layer (10 nodes), two hidden layers (20 nodes per layer) and an output layer (3 nodes). The neural network outputs the optimal fan blade angle, speed and door seal compression under the current working condition based on the input parameters and the mapping relationship between the fan blade angle, speed and temperature gradient around the door obtained by historical data training. According to the optimal parameters output by the neural network, the fan blade angle is adjusted in real time by the actuator, with an accuracy of 0.1 degrees, the speed is 10rpm, and the compression of the door seal is 0.1mm. The actuator response time does not exceed 100ms. After the adjustment is completed, the temperature data around the door is collected again at a frequency of 2Hz, the adjusted temperature gradient is calculated, and the new data is fed back to the neural network to form a closed-loop control and continuously optimize the adjustment parameters. During the mission of a refrigerated delivery vehicle, the on-board system detected that the number of door openings and closings reached 25 times within 30 minutes, the average door opening time was 45 seconds, and the temperature fluctuation amplitude of the compartment was 3℃, and it was determined that the door was frequently opened and closed. The current fan speed is immediately collected as 1200rpm, the blade angle is 30°, and the compression of the door seal is 5mm. 8 temperature sensors distributed around the door collect data at the same time, and the temperature gradient is calculated to be 0.5℃ / cm using the central difference method. These data were input into a pre-trained multi-layer perceptron neural network with 10 input nodes, two hidden layers of 20 nodes each, and 3 output nodes, using the ReLU activation function. The network was trained on 1000 sets of historical data for 500 rounds with a learning rate of 0.001. The network outputs the optimal fan blade angle of 35°, the speed of 1500rpm, and the compression of the seal strip of 6mm. The actuator then began to adjust the blade angle at a rate of 0.1° / s for 50 seconds; the speed increased at a rate of 60rpm / s for 5 seconds; and the compression of the seal strip increased at a rate of 0.2mm / s for 5 seconds. After the adjustment was completed, the temperature sensor re-collected data at a frequency of 2Hz for 15 seconds, and the new temperature gradient was calculated to be 0.3℃ / cm. This new data was fed back to the neural network, and the network output fine-tuning suggestions: the blade angle remained unchanged, the speed was reduced to 1450rpm, and the compression of the seal strip was increased to 6.2mm.

[0038] S107. During the fan and door seal adjustment process, the variance of the temperature data of the area near the door in the temperature control cabin before and after the adjustment is calculated in real time. According to the variance change trend, the fan blade angle, rotation speed and door seal compression are optimized online to find the optimal adjustment parameter combination. The fan and door sealing performance are adjusted simultaneously using the optimal adjustment parameter combination.

[0039] Acquire temperature data collected by a temperature sensor, and use the temperature data to calculate temperature variance; construct an objective function using the temperature variance and energy consumption factor, and the objective function includes a temperature variance term and an energy consumption term; determine the fan blade angle, rotation speed, and sealing strip compression as optimization variables, and use the optimization variables to adjust the objective function f(x)=Σ(Ti-Tavg) 2 / n+λE, where Ti is the temperature of each point, Tavg is the average temperature, n is the number of sensors, E is the energy consumption factor, and λ is the weight coefficient; the optimization variables are optimized online by a gradient descent algorithm, and the gradient descent algorithm uses an adaptive step size for iteration; if the change amplitude of the temperature variance is less than a preset threshold or reaches a maximum number of iterations, the optimization process is stopped; it is determined whether the objective function value is the minimum, and if the objective function value is the minimum, the current parameter combination is determined to be the global optimal solution; the global optimal solution is applied to the adjustment of the fan blade angle, rotation speed, and sealing strip compression.

[0040] For example, during the fan and door seal adjustment process, 20 temperature sensors are evenly arranged within a 1-meter range around the door to collect real-time temperature data of the area near the door in the temperature control cabin. The sampling frequency is set to 10 times per second and the collection time is 30 seconds. The variance of the temperature data before and after adjustment is calculated to obtain the temperature uniformity index. The variance calculation formula is Σ(Ti-Tavg) 2 / n, where Ti is the temperature of each point, Tavg is the average temperature, and n is the number of sensors. According to the calculated temperature variance change trend, the objective function f(x) = Σ(Ti-Tavg) is constructed. 2 / n+λE, where E is the energy consumption factor and λ is the weight coefficient. The temperature variance and energy consumption are taken as optimization targets, the fan blade angle, speed and door seal compression are taken as optimization variables, and the gradient descent algorithm is used to optimize these three parameters online. During the optimization process, the adjustment range of the fan blade angle is set to 0°~90°, the speed range is 500~3000 rpm, and the door seal compression range is 2~10 mm. Adaptive step size is adopted, and the initial step size is 1°, 100 rpm and 0.5 mm respectively, which decreases by 1 / sqrt(k) with the number of iterations k. The single adjustment amplitude is set not to exceed 10% of the maximum range to avoid frequent and large adjustments. Through multiple iterative optimizations, when the change amplitude of the temperature variance is less than the preset threshold of 0.1℃2 or the maximum number of iterations is 50, the optimization process is stopped. In order to avoid falling into the local optimal solution, the method of multiple random initialization parameters is adopted, and three rounds of independent optimization are performed, and the parameter combination with the smallest objective function value is selected as the global optimal solution. Finally, this optimal parameter combination is simultaneously applied to the adjustment of the fan blade angle, speed and door seal compression. During the mission of a refrigerated delivery vehicle, temperature optimization control is started. 20 high-precision temperature sensors with an accuracy of ±0.1°C are evenly arranged within 1 meter around the door to form a 4x5 matrix. Data is collected every 0.1 seconds for 30 seconds to obtain 6000 temperature data points. In the initial state, the temperature variance is calculated to be 2.5°C2 and the average temperature is 4°C. Construct the objective function f(x)=Σ(Ti-4) 2 / 20+0.01E, where E is the normalized energy consumption factor. The initial values ​​of the optimization variables are set to a fan blade angle of 45°, a speed of 1500rpm, and a sealing strip compression of 5mm. The gradient descent algorithm is used, and the initial learning rate is set to 0.1, which decays by 10% every 10 iterations. In the first round of iterations, the partial derivatives of the objective function with respect to the three variables are calculated to obtain the gradient vector [-0.5, 20, 0.2]. Based on this, the parameters are updated, including the blade angle adjusted to 44.95°, the speed increased to 1502rpm, and the compression increased to 5.01mm. The single adjustment amplitude did not exceed 10% of the maximum range. After 30 iterations, the temperature variance dropped to 1.2℃2, but the change tended to be gentle. The second round of optimization was then started, and the parameters were randomly initialized to a blade angle of 60°, a speed of 2000rpm, and a compression of 7mm. After 40 iterations in the second round, the temperature variance was further reduced to 0.8℃2. After the third round of optimization, the optimal parameter combination was obtained as blade angle 52°, rotation speed 1750rpm, compression 6.5mm, and temperature variance 0.6℃2. This set of parameters was used for final adjustment to achieve the best balance between temperature uniformity and energy consumption.

[0041] S108. Use a door ambient temperature prediction model based on a long short-term memory network to predict the temperature change trend of the area around the door in the future based on the historical opening and closing state data of the door, the thermal conductivity of the door material, and the current temperature gradient around the door. Pre-adjust the fan and the door sealing strip in advance based on the predicted temperature change trend to reduce the temperature fluctuation of the cargo around the door caused by frequent opening and closing of the door.

[0042] Receive door opening and closing status data and temperature data, the data being collected by a door sensor; construct an input data set containing time series features based on the data, the input data set including rolling storage data; use a long short-term memory network to construct a temperature prediction model, the prediction model including preset LSTM layers, preset units in each layer, and a dropout layer added in the middle; input the input data set into the prediction model to obtain the temperature change trend of the area around the door in the future; calculate the interval in which a preset percentage of temperature changes in historical data fall, and define the changes exceeding the interval as a temperature change threshold; determine whether the temperature change trend exceeds the temperature change threshold, and if so, use a gradient descent algorithm to optimize the objective function; calculate the optimal pre-adjustment parameters for the fan blade angle, rotation speed, and door sealing strip compression according to the objective function; perform a pre-adjustment operation before the temperature change occurs, the pre-adjustment operation is performed based on the optimal pre-adjustment parameters; receive actual temperature change data, and calculate the deviation between the actual temperature change and the predicted value; dynamically adjust the prediction model parameters and the pre-adjustment strategy according to the deviation.

[0043] Exemplarily, the historical opening and closing state data of the car door are collected and stored, including the number of opening and closing times, opening and closing times, and opening angles. At the same time, the thermal conductivity of the car door material, the real-time temperature gradient data around the car door, and the ambient temperature data are recorded. The data collection frequency is set to once per minute, and the data of the last 24 hours are rolled and stored to construct an input data set containing time series features. The temperature prediction model is constructed using a long short-term memory network. The number of input layer nodes is set to be the same as the number of features. It contains 3 LSTM layers, 128 units per layer, and a dropout layer is added in the middle. The dropout rate is 0.2. The output layer is a fully connected layer, and the temperature prediction value for each minute in the next 30 minutes is output. The mean square error is used as the loss function, the Adam optimizer is used, the learning rate is set to 0.001, and 500 rounds of training are performed. The constructed input data set is input into the prediction model to obtain the temperature change trend of the area around the car door in the next 30 minutes. By calculating the interval in which 95% of the temperature changes in the historical data fall as the normal range, the changes beyond this range are defined as the temperature change threshold to determine whether pre-adjustment is required. If the predicted temperature change exceeds the threshold, the objective function is optimized using the gradient descent algorithm. The objective function takes into account temperature uniformity and energy consumption, and calculates the optimal pre-adjustment parameters for the fan blade angle, speed, and door seal compression. The pre-adjustment operation is performed 10 minutes before the temperature change occurs to suppress the temperature fluctuation in advance. After the pre-adjustment is performed, the deviation between the actual temperature change and the predicted value is compared, and the deviation data is fed back to the prediction model to dynamically adjust the model parameters and pre-adjustment strategy to continuously optimize the prediction accuracy and adjustment effect. During the mission of a refrigerated delivery vehicle, the on-board system collects data once a minute, recording an average of 10 door openings per hour, an average opening and closing time of 15 seconds each time, an opening angle between 60° and 90°, a thermal conductivity of 0.023W / (m·K) for the door material, an average temperature gradient of 0.5℃ / cm around the door, and an ambient temperature range of -5℃ to 35℃. The data of the last 24 hours is rolled and stored to form an input matrix of 14408. The long short-term memory network model contains 3 LSTM layers, each with 128 units, and a dropout layer in the middle with a learning rate of 0.2. The model training uses historical data from the past 3 months, a total of 130,000 samples, using the mean square error loss function and Adam optimizer, with a learning rate of 0.001. After 500 rounds of training, the mean absolute error on the validation set dropped to 0.3°C. The model predicts the temperature change around the door in the next 30 minutes and finds that the temperature may rise from 4°C to 7°C, which exceeds the 95% confidence interval of historical data.

[0044] ±1.5℃. Immediately start the pre-adjustment calculation, and use the gradient descent algorithm to optimize the objective function f(x)=w1 T+w2*E, where T is the temperature uniformity, E is the energy consumption, and the weights w1=0.7 and w2=0.3. After 100 iterations, the optimal pre-adjustment parameters are obtained: the fan blade angle is adjusted from 25° to 35°, the speed is increased from 1500rpm to 1800rpm, and the sealing strip compression is increased from 6mm to 8mm. Perform pre-adjustment 8 minutes before the predicted temperature rises. After adjustment, the actual temperature only rises to 5.5℃, which is 1.5℃ lower than the predicted value. Feed this deviation back to the prediction model, and fine-tune the model parameters through online learning to reduce the average absolute error of the next prediction to 0.25℃.

[0045] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A method for emergency guarantee of fresh food delivery, characterized in that: The method comprises: Acquire the temperature data collected in real time by multiple temperature sensors arranged in different areas of the temperature control cabin of the fresh food delivery vehicle, obtain a three-dimensional temperature distribution cloud map inside the entire temperature control cabin, and determine whether the temperature distribution of different cargo storage areas in the current temperature control cabin meets the preset temperature uniformity threshold requirements; If the temperature distribution does not meet the preset temperature uniformity threshold requirement and the temperature difference exceeds the threshold, the local area near the door where the temperature gradient exceeds the preset gradient threshold is identified based on the three-dimensional temperature distribution cloud map, and it is used as the target area for key adjustment of the fan and door sealing performance; Computational fluid dynamics methods are used to numerically simulate the airflow field of the target area, the geometric structure characteristics of the temperature control cabin, the fan layout, different fan blade angles, and different door opening areas and angles; Through numerical simulation of the airflow field, the airflow streamline distribution diagram inside the temperature control cabin under different fan blade angles and door opening areas and angles is obtained, the fan angle range and door opening angle threshold that cause airflow short circuit and external air infiltration are identified, and they are excluded from the adjustable range of the fan and door, and the initial speed of the fan and the compression amount of the door sealing strip are determined; The door sensors on the fresh delivery vehicles collect the time, angle and frequency data of the door opening and closing in real time, and combine the pre-established door opening and closing state judgment model to determine whether the current vehicle is in a frequent door opening and closing condition; If the vehicle is in a condition of frequent door opening and closing, the optimal adjustment parameters corresponding to the current vehicle door opening condition are obtained based on the determined initial speed of the fan and the compression of the door sealing strip, as well as the pre-established mapping model of the fan blade angle, speed and the temperature gradient around the door, and the fan blade angle, speed and the compression of the door sealing strip are adjusted in real time; During the fan and door seal adjustment process, the variance of the temperature data in the temperature control chamber near the door before and after the adjustment is calculated in real time. According to the variance change trend, the fan blade angle, speed and door seal compression are optimized online to find the optimal adjustment parameter combination. The fan and door sealing performance are adjusted simultaneously using the optimal adjustment parameter combination. A door ambient temperature prediction model based on a long short-term memory network is used to predict the temperature change trend of the area around the door in the future according to the historical opening and closing state data of the door, the thermal conductivity of the door material and the current temperature gradient around the door. The fan and door sealing strip are pre-adjusted in advance according to the predicted temperature change trend to reduce the temperature fluctuation of the cargo around the door caused by frequent opening and closing of the door.

2. The method according to claim 1, characterized in that: The method of acquiring temperature data collected in real time by multiple temperature sensors arranged in different areas of the temperature control cabin of the fresh food delivery vehicle, obtaining a three-dimensional temperature distribution cloud map inside the entire temperature control cabin, and judging whether the temperature distribution of different cargo storage areas in the current temperature control cabin meets the preset temperature uniformity threshold requirement includes: Acquire real-time temperature data collected by multiple temperature sensors in the temperature control cabin, and construct a three-dimensional temperature distribution matrix inside the temperature control cabin according to the temperature data; Using a trilinear interpolation algorithm to estimate the temperature values ​​between the sampling points in the three-dimensional temperature distribution matrix, to obtain a continuous three-dimensional temperature distribution cloud map inside the temperature control cabin; According to different cargo storage areas in the temperature-controlled cabin, multiple temperature monitoring units are divided, and the average temperature value and standard deviation in the temperature monitoring unit are calculated; If the temperature standard deviation of the temperature monitoring unit exceeds a preset temperature uniformity threshold, it is determined that the temperature distribution of the temperature monitoring unit is uneven; The cooling or heating power is calculated according to the temperature deviation and the change rate of the temperature monitoring unit, and the adjustment instruction of the temperature control system is generated.

3. The method according to claim 1, characterized in that: If the temperature distribution does not meet the preset temperature uniformity threshold requirement and the temperature difference exceeds the threshold, a local area where the temperature gradient exceeds the preset gradient threshold and is close to the door is identified according to the three-dimensional temperature distribution cloud map, and is used as a target area for focusing on adjusting the fan and door sealing performance, including: Obtain a three-dimensional temperature distribution cloud map and use the central difference method to calculate the temperature gradient between adjacent grid points; Determining whether the temperature gradient exceeds a preset threshold and is located in the area around the door; If the conditions are met, the area is marked as a local temperature anomaly area; The marked local temperature anomaly areas are grouped using a clustering algorithm; Calculate the average temperature, area and center of gravity position of each abnormal area; Based on the internal structural information of the vehicle compartment, determine the target areas for key adjustments to the fan and door sealing performance.

4. The method according to claim 1, characterized in that: The computational fluid dynamics method is used to perform numerical simulation of the airflow field on the target area, the geometric structure characteristics inside the temperature control cabin, the fan layout, different fan blade angles, and different door opening areas and angles, including: Acquire the internal geometric structure characteristics of the temperature control cabin, the fan layout position and the door opening state parameters, build a digital model of the temperature control cabin according to the parameters, and perform local mesh encryption on the target area identified in the digital model; Discretizing the computational domain of the digital model, and if there are complex geometric shapes in the computational domain, adaptively processing the complex geometric shapes; Setting boundary conditions of the computational domain, wherein the boundary conditions include inlet wind speed, outlet pressure and a wall no-slip condition; The control equation is discretized by using the finite volume method to solve the velocity field and the pressure field. The pressure field is solved by solving the Poisson equation. According to the set fan blade angle and door opening state parameters, simulate the airflow field distribution under different working conditions; Determine whether the simulation result converges, and if the residual of the simulation result is less than a preset threshold, determine that the simulation result converges; Post-processing the converged simulation results to generate airflow velocity vector diagram, temperature value line diagram, streamline diagram and pressure cloud diagram; Extracting the airflow velocity, temperature distribution and pressure gradient data of the target area from the post-processing results; The extracted data are processed by dimensionality reduction to obtain the main factors affecting the airflow field distribution in the target area.

5. The method according to claim 1, characterized in that: The method of numerically simulating the airflow field to obtain the airflow streamline distribution diagram inside the temperature control cabin under different fan blade angles and door opening areas and angles, identifying the fan angle range and door opening angle threshold that cause airflow short circuit and external air infiltration, and excluding them from the adjustable range of the fan and the door, and determining the initial speed of the fan and the compression amount of the door sealing strip, includes: Receiving numerical simulation results of an airflow field including a fan blade angle and a door opening state, and performing streamline tracing according to the numerical simulation results to obtain an airflow streamline distribution diagram; Using the Canny edge detection algorithm on the airflow streamline distribution diagram to obtain streamline feature data including streamline density, curvature and length; According to the streamline characteristic data, a clustering algorithm is used to determine the characteristic patterns of airflow short circuit and external air infiltration, wherein the airflow short circuit determination criterion is that the streamline closure degree is greater than a preset closure degree threshold and the duration exceeds a preset time threshold, and the external air infiltration determination criterion is that the streamline density near the door changes suddenly and the direction reverses; Based on the characteristic patterns of airflow short-circuiting and external air infiltration, the mathematical model of the constraint relationship between the fan angle and the door opening angle is determined by using the multivariate linear regression method. According to the constraint relationship mathematical model, a random forest algorithm is used to determine a prediction model for the initial speed of the fan and the compression amount of the door sealing strip. The input parameters of the prediction model include the temperature difference between the inside and outside of the temperature-controlled cabin, the type and load of cargo, and the internal structural characteristics of the compartment. The internal structural characteristics of the compartment include the location of obstacles and the distribution of cargo.

6. The method according to claim 1, characterized in that The door sensor on the fresh delivery vehicle collects the time, angle and frequency data of the door opening and closing in real time, and combines the pre-established door opening and closing state judgment model to judge whether the current vehicle is in a frequent door opening and closing condition, including: Acquire the door opening and closing time and angle data collected by the door sensor, wherein the collection frequency of the door sensor is a preset value; Applying a Kalman filter algorithm to process the collected data, wherein the measurement noise covariance of the Kalman filter algorithm is set to a preset value, and the process noise covariance is set to a preset value; Determining the number of door openings, average opening duration, and average opening angle, wherein the number of door openings, average opening duration, and average opening angle are calculated within a time window dynamically adjusted according to the vehicle operation status; Determine the door opening and closing state by inputting the door opening times, average opening time and average opening angle into a pre-trained support vector machine model, wherein the support vector machine model adopts a radial basis function kernel; If the probability value output by the support vector machine model exceeds a preset threshold value for a preset number of consecutive time windows after being smoothed by an exponential moving average, it is determined that the vehicle is in a frequent door opening and closing condition.

7. The method according to claim 1, characterized in that If the vehicle is in a frequent door opening and closing condition, the optimal adjustment parameters corresponding to the current vehicle door opening condition are obtained according to the determined initial fan speed and the compression of the door sealing strip, as well as the pre-established mapping model of the fan blade angle, speed and the temperature gradient around the door, and the fan blade angle, speed and the compression of the door sealing strip are adjusted in real time, including: Obtain characteristic parameters of the vehicle under the condition of frequent door opening and closing, including door opening and closing frequency, door opening duration and cabin temperature change rate; The fan speed, blade angle and compression of the door seal are collected, and the temperature sensors distributed around the door collect temperature data; Calculate the temperature gradient based on characteristic parameters, fan speed, blade angle, seal strip compression and temperature data; The characteristic parameters, fan speed, blade angle, sealing strip compression and temperature gradient are input into a pre-trained multi-layer perceptron neural network; The neural network outputs the optimal fan blade angle, optimal fan speed and optimal door seal compression under the current vehicle door opening condition; The fan blade angle is adjusted according to the optimal fan blade angle, the fan speed is adjusted according to the optimal fan speed, and the door seal strip compression is adjusted according to the optimal door seal strip compression.

8. The method according to claim 1, characterized in that In the process of adjusting the fan and the door seal strip, the variance of the temperature data of the area near the door in the temperature control cabin before and after the adjustment is calculated in real time, and the fan blade angle, speed and door seal strip compression are optimized online according to the variance change trend to find the optimal adjustment parameter combination, and the fan and door sealing performance are adjusted simultaneously using the optimal adjustment parameter combination, including: Acquire temperature data collected by a temperature sensor, where the temperature data is used to calculate temperature variance; The temperature variance and energy consumption factor are used to construct an objective function, wherein the objective function includes a temperature variance term and an energy consumption term; Determine the fan blade angle, rotation speed and sealing strip compression as optimization variables, and the optimization variables are used to adjust the objective function f(x)=Σ(Ti-Tavg) 2 / n+λE, where Ti is the temperature of each point, Tavg is the average temperature, n is the number of sensors, E is the energy consumption factor, and λ is the weight coefficient; Performing online optimization of the optimization variables by a gradient descent algorithm, wherein the gradient descent algorithm uses an adaptive step size for iteration; If the change in temperature variance is less than the preset threshold or reaches the maximum number of iterations, the optimization process is stopped; Determine whether the objective function value is the minimum, and if the objective function value is the minimum, determine that the current parameter combination is the global optimal solution; The global optimal solution is applied to the adjustment of the fan blade angle, rotation speed and sealing strip compression.

9. The method according to claim 1, characterized in that: The door surrounding temperature prediction model based on the long short-term memory network is used to predict the temperature change trend of the area around the door in the future according to the historical opening and closing state data of the door, the thermal conductivity of the door material and the current temperature gradient around the door. The fan and the door sealing strip are pre-adjusted in advance according to the predicted temperature change trend to reduce the temperature fluctuation of the cargo around the door caused by the frequent opening and closing of the door, including: Receiving door opening and closing status data and temperature data, the data being collected by a door sensor; constructing an input data set containing time series features according to the data, wherein the input data set includes rolling storage data; A temperature prediction model is constructed using a long short-term memory network. The prediction model includes preset LSTM layers, each layer has preset units, and a dropout layer is added in the middle; Inputting the input data set into a prediction model to obtain a future temperature change trend of an area around a vehicle door; Calculating an interval within which a preset percentage of temperature changes in historical data fall, and defining changes beyond the interval as a temperature change threshold; Determine whether the temperature change trend exceeds a temperature change threshold, and if so, use a gradient descent algorithm to optimize the objective function; Calculating optimal pre-adjustment parameters of fan blade angle, rotation speed and door seal compression according to the objective function; Performing a pre-adjustment operation before the temperature change occurs, the pre-adjustment operation being performed based on the optimal pre-adjustment parameters; receiving actual temperature change data, and calculating the deviation between the actual temperature change and the predicted value; The prediction model parameters and pre-adjustment strategies are dynamically adjusted according to the deviation.

Images