Method for air container control

Abstract

The present invention relates to a method for controlling an air container, and the technical scheme comprises: integrating a multi-dimensional sensor network, including temperature, humidity, pressure, vibration, and position sensors, to sense the state of the air container in real time; the real-time sensing comprises monitoring the internal and external environment of the container, external force impact or vibration; the sensor data is transmitted in real time to a control unit ICU integrated in the container, and the control unit ICU triggers a signal instruction including adjusting the temperature or issuing an alarm; during the loading and transportation process, by monitoring the center of gravity position and load distribution of the container, the sensor data is calculated in real time and instructions are issued to control mechanical hydraulic equipment to adjust the placement position of the goods; during transportation, if it is detected that the posture or load distribution of the container is abnormal, the internal cargo position is adjusted or the balance state is restored by adjusting the counterweight device inside the container; multiple smart containers on the same flight are interconnected through wireless communication technology to form a dynamic balance network.

Description

Technical Field

[0001] The present invention relates to a method for controlling a container, in particular to a method for controlling an aviation container. Background Art

[0002] Chinese invention patent 202210765684.9 discloses a method and device for air container control. Its main technical solution is to detect the smoke concentration in the heat dissipation duct, shut down the heat dissipation fan according to the detection result, or cut off the power supply of the air container, so as to avoid the heat dissipation process of the air container diluting the smoke concentration in the cargo hold, improve the accuracy of smoke detection in the cargo hold, and timely detect and extinguish the fire. This method has certain advantages in improving the safety of air container cargo transportation, but there are still some shortcomings and disadvantages in practical applications, which affect its comprehensive performance and application effect. First, the method mainly relies on detecting the smoke concentration in the heat dissipation duct to judge the fire risk, and prevents smoke dilution by shutting down the heat dissipation fan or cutting off the power supply of the air container, so as to ensure the accuracy of fire detection. This control strategy is too single, only responding to specific situations of fire detection, and lacking comprehensive monitoring and multi-dimensional management of the overall state of the air container. For example, this method does not involve the monitoring and control of key parameters such as temperature, humidity, pressure, and center of gravity position in the container, and cannot fully perceive the cargo state, resulting in limited response capabilities when facing other non-fire emergencies. In addition, simply turning off the cooling fan or cutting off the power supply to prevent smoke dilution may cause the temperature inside the container to rise rapidly in an emergency, which may lead to other risks, especially when carrying temperature-sensitive goods (such as medicines, food or chemicals), which may cause damage to the goods.

[0003] Secondly, this method lacks the ability of dynamic adjustment and intelligent control. Its main control method is to trigger simple switch operations (such as turning off the fan or cutting off the power supply) through preset thresholds, without considering the complex changes and diversity of the internal environment of the container. For example, the temperature, humidity, pressure and cargo weight distribution in the cargo hold will change with changes in flight altitude, route and external environmental conditions. Single smoke concentration monitoring and fan control cannot fully cope with these complex situations. At the same time, this method fails to make full use of integrated sensors and advanced algorithms (such as artificial intelligence or machine learning) to achieve real-time monitoring and intelligent adjustment of container status, so it shows low flexibility and adaptability when dealing with changing transportation environments and diverse cargo types.

[0004] Third, the method's deficiencies in data transmission and collaborative control also limit its application effect. According to the description in the patent document, the method does not mention how to exchange data and work collaboratively with other containers on the same flight, and lacks an effective data sharing and collaborative control mechanism. In modern air transportation, multiple air containers often need to work together to ensure overall transportation safety and efficiency. For example, in the event of an emergency, the status information of a single container (such as temperature rise, smoke detection, etc.) can be quickly shared with other containers and ground control centers through wireless communication technology, so that a more effective response plan can be formulated. However, the patented method is limited to operational control within a single container, and lacks data synchronization and collaborative control mechanisms across containers, making it impossible to fully utilize the overall information of the container group in actual operations, which may lead to improper local operations and affect overall safety and transportation efficiency.

[0005] In view of the above shortcomings, it can be seen that although the method for air container control proposed in Patent 202210765684.9 improves the accuracy and response speed of fire detection under specific circumstances, it still has great limitations and disadvantages as a whole. This method cannot comprehensively monitor and dynamically adjust the multi-dimensional state inside the container, lacks intelligent control means and cross-container collaborative control mechanism, the control strategy is single and passive, the energy consumption management efficiency is not high, and it lacks long-term data analysis and optimization capabilities. Therefore, this method may be subject to greater restrictions in practical applications, and it is difficult to cope with the complex and changeable situations in the modern air transportation environment and the growing safety, efficiency and environmental protection requirements. In contrast, modern air container control methods should be more intelligent, autonomous and integrated, and can better ensure the safety and transportation efficiency of goods, while reducing operating costs and energy consumption, and providing more efficient solutions for the development of the aviation logistics industry. Summary of the invention

[0006] The object of the present invention is to provide a method for air container control, thereby solving some of the drawbacks and deficiencies pointed out in the background technology.

[0007] The present invention solves the above-mentioned technical problems by adopting the following technical solution, which includes the following steps:

[0008] S1. Intelligent state perception and control:

[0009] S1.1. Integrate a multi-dimensional sensor network, including temperature, humidity, pressure, vibration, and position sensors, to sense the state of the air container in real time; the real-time sensing includes monitoring the internal and external environment of the container, external force impact or vibration;

[0010] S1.2, the sensor data is transmitted in real time to the control unit ICU integrated in the container, and the control unit ICU triggers signal instructions including adjusting the temperature or issuing an alarm;

[0011] S2. Dynamic load and balancing management:

[0012] S3.1. During loading and transportation, by monitoring the center of gravity and load distribution of the container, real-time sensor data is calculated and instructions are issued to control the mechanical hydraulic equipment to adjust the cargo placement;

[0013] S3.2. During transportation, if the posture or load distribution of the container is detected to be abnormal, the position of the internal cargo or the counterweight device inside the container is adjusted to restore the balance state;

[0014] S3. Container group collaborative control:

[0015] S3.1. Multiple smart containers on the same flight are interconnected through wireless communication technology to form a dynamic balancing network; loads and environmental conditions between containers are balanced through data exchange in the dynamic balancing network;

[0016] S3.2. Share data among containers, including temperature, pressure, and location. Based on the shared data, the container group jointly adjusts their respective environmental control parameters.

[0017] Further, the dynamic load and balance management includes a data acquisition system, a dynamic balance algorithm and an automatic control system for mechanical hydraulic equipment;

[0018] The data acquisition system is composed of weight sensors, inertial measurement units (IMUs), laser rangefinders and ultrasonic sensors installed on the bottom, walls and top of the air container.

[0019] The data acquisition system monitors the weight distribution, center of gravity position, cargo height change and space occupancy of each area in the container in real time; including the use of a data fusion algorithm: assuming that the weight distribution function of any point (x, y, z) in the container is W(x, y, z, t), where x, y, z are three-dimensional spatial coordinates, and t is time, indicating the time point of the current acquisition, by solving the multidimensional integral:

[0020]

[0021] Where C(t) is the coordinate vector of the center of gravity of the container at time t, expressed as C(t) = (C x (t),C y (t),C z(t)), Ω represents the three-dimensional spatial volume of the container, and W(x, y, z, t) is the weight density function at the three-dimensional coordinate point (x, y, z) at time t.

[0022] Furthermore, the dynamic balance algorithm includes designing and autonomous control decisions and issuing instructions, wherein a dynamic balance control model is established using a reinforcement learning algorithm of adaptive learning, and the future trend of center of gravity changes is predicted and the cargo adjustment strategy is optimized through a combination of historical data and real-time data; the optimization path of the model is to achieve a minimum value for the nonlinear objective function:

[0023]

[0024] Among them, J is the optimization objective function of dynamic balance control, which represents the comprehensive index of the overall performance of the system in the time interval from 0 to T; is the rate of change of the center of gravity position C(t) relative to time t, reflecting the moving speed and direction of the center of gravity, expressed as the modulus of the vector; P(t) is the balance factor of the load distribution, indicating the uniformity of the positions of the cargoes in the container, and its value is calculated by the current load state; E(t) is the energy consumption function of the hydraulic equipment during cargo adjustment, indicating the energy consumed in performing each adjustment operation; α and β are adjustment parameters, representing the weight coefficients of load balance and energy consumption in the optimization objective, respectively;

[0025] When it is detected that the center of gravity deviates from the preset safety range or the load distribution is unbalanced, an adjustment instruction is generated according to the algorithm. The instruction includes the direction, distance, speed of movement and the operating details of the required hydraulic equipment.

[0026] Furthermore, the automatic control system of the mechanical hydraulic equipment includes mechanical hydraulic equipment integration and intelligent control and real-time feedback closed loop;

[0027] Hydraulic push rods, adjustable support rods and roller platforms are configured at the bottom and inside of the container to move and redistribute the cargo according to the instructions issued by the control unit ICU. A closed-loop control system is formed by integrating the mechanical hydraulic equipment with the sensor network and ICU in the container. After the equipment executes the instructions, the sensor feeds back the adjusted center of gravity position and load status, and the adjustment effect is verified by an iterative formula:

[0028] C n+1 =C n +ΔC n =C n +λ(C target -C n )

[0029] Among them, C n+1 is the center of gravity position after the n+1th adjustment, indicating the new center of gravity coordinates after adjustment; C nis the center of gravity position before the nth adjustment, indicating the currently known center of gravity coordinates; ΔC n is the change in the center of gravity during the nth adjustment, indicating the change in the center of gravity position during this adjustment; λ is the step coefficient, indicating the control parameter of the adjustment amplitude, and its value is between 0<λ≤1; C target It is the preset target center of gravity position, which indicates the coordinates of the center of gravity when the cargo inside the container is in a balanced state. If the target balance state is still not reached after adjustment, a second or multiple iterative adjustments are performed according to the feedback data until a balance state is reached.

[0030] Furthermore, the dynamic balancing network construction includes the construction of a multi-container wireless communication network, the coordinated processing of data exchange and dynamic balancing algorithms, and the coordinated balancing and environmental optimization of multiple containers;

[0031] A wireless communication module is integrated in each air container, and the wireless communication module is configured with a unique identity identifier ID; self-organizing network technology is used to enable the communication module of each container to discover and connect to other smart containers on the same flight, forming a multi-hop dynamic balance network;

[0032] During the network formation process, each container announces its current status through a broadcast protocol, including the center of gravity position, temperature, humidity, and load information, and selects a communication path based on link quality, signal strength, and network topology. Suppose the communication radius of each container i is r i , the signal strength function is S i (d), where d is the distance between two containers, the communication topology of the multi-hop dynamic balance network is described as an integral equation:

[0033]

[0034] Where T is the communication topology quality function of the entire network, which indicates the overall communication coverage quality of the network; Ω represents the three-dimensional space where the container is located; B is the number of containers; Θ(x) is the Heaviside step function, which is used to determine whether the signal is valid. It is defined as: when x ≥ 0, Θ(x) = 1, otherwise it is 0; r i is the communication radius of the i-th container, indicating the maximum communication distance that the container can cover; S i (d) is the signal strength function, which indicates that the signal strength between two containers depends on the distance d between them.

[0035] Furthermore, the collaborative processing method of data exchange and dynamic balance algorithm includes: the data sharing process uses a distributed ledger based on blockchain technology, and each container runs a collaborative dynamic balance algorithm and calculates an adjustment strategy based on the shared data and status; assuming that at time t, the state vector of the container group is X(t)=[x1(t),x2(t),…,x N (t)], where x i (t) is the state variable of the i-th container, including the center of gravity position, temperature and humidity parameters. The optimization objective is to minimize the function:

[0036]

[0037] Among them, J is the optimization objective function, which represents the performance index of the entire container group in the time interval [0, T]. The goal is to minimize it to optimize the overall performance; is the rate of change of the i-th container state variable relative to time, reflecting the speed of change of the container state; Φ i (x i (t), X(t)) is the interaction function between the i-th container and other containers, which represents the interaction between containers due to changes in environment and load at time t; γ i is the corresponding weight coefficient.

[0038] Furthermore, the multi-container coordinated balancing and environmental optimization method includes:

[0039] When a container detects a center of gravity deviation or abnormal environmental conditions, an adjustment plan is calculated based on the collaborative control strategy and notified to other containers through the wireless network; after each container performs the adjustment action, new status data is collected in real time and fed back to the dynamic balance network, forming an adaptive learning closed-loop system;

[0040] Through data exchange and adjustment, the entire container group tends to a stable equilibrium state; suppose that after each adjustment, the state vector of the container is updated to X n+1 =X n +ΔX n , where X n is the state vector before the nth adjustment, representing the state set of all containers before adjustment, ΔX n is the adjustment amount, indicating the state change during the adjustment process, and satisfies the following iterative relationship:

[0041] ΔX n =η(X tre -X n )

[0042] Among them, η is the adjustment step coefficient, which controls the amplitude of each adjustment and its value range is between 0<η≤1; Xtre is the target state vector, which represents the state set when the container group reaches the equilibrium state. The adjustment process is iterated until the target state is reached.

[0043] The method for air container control of the present invention has the following beneficial effects:

[0044] First, by integrating a multi-dimensional sensor network and an intelligent control unit, the internal and external status of the container can be monitored in real time, including parameters such as the center of gravity, temperature, humidity, pressure, and vibration, thereby achieving accurate state perception and dynamic monitoring of the container. This real-time perception and monitoring improves the safety and stability of cargo transportation, and can quickly respond to abnormal situations and take timely measures to prevent cargo damage or accidents.

[0045] The dynamic load and balance management system can automatically optimize the placement of cargo and ensure the balance of the container during flight by adjusting the center of gravity and load distribution inside the container through real-time calculation and control during loading and transportation. This method realizes dynamic adjustment of cargo position through automatic control of mechanical hydraulic equipment, reduces the risk of container overturning and cargo movement, and improves the stability and safety of air transportation.

[0046] By establishing a multi-container wireless communication network and using a distributed ledger using blockchain technology, each container can achieve data sharing and collaborative control. Using self-organizing network technology and wireless communication modules, a multi-hop dynamic balance network is formed to ensure that the status information of each container can be transmitted and exchanged in real time and reliably, so that the entire container group can jointly calculate and execute the optimal balance strategy and environmental control parameters, thereby optimizing overall transportation efficiency, reducing energy consumption, and improving logistics efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 The present invention is a flow chart of the method for air container control.

[0048] Figure 2 This is a flow chart of the dynamic load and balancing management method of the present invention.

[0049] Figure 3 This is a flow chart of the dynamic balance network construction method of the present invention. DETAILED DESCRIPTION

[0050] The specific implementation modes of the present invention will be described in detail below in conjunction with the accompanying drawings.

[0051] Combined with Figure 1, a method for air container control, including an intelligent state perception and control system, the system integrates a multi-dimensional sensor network, including temperature sensors, humidity sensors, pressure sensors, vibration sensors and position sensors, these sensors sense the state of the air container in real time, the real-time perception includes monitoring the environmental parameters inside and outside the container, such as the changes in temperature, humidity and pressure, and also includes detecting the intensity and frequency of external force impact or vibration applied to the container to ensure the safety of the goods during transportation. The sensor network transmits the collected data to an intelligent control unit (ICU) integrated inside the container in real time. After receiving the sensor data, the control unit ICU triggers corresponding signal instructions through built-in algorithms and logical judgments. These signal instructions include dynamically adjusting the temperature inside the container according to the data fed back by the sensor, such as starting or shutting down a heater or cooler device to maintain the optimal environmental conditions required for cargo storage, or when an abnormal condition is detected, such as an external impact or vibration exceeding a preset threshold, an alarm signal is issued to notify relevant personnel or a ground control center so that corresponding measures can be taken in time to ensure the safety of the goods and the stability of transportation.

[0052] During the loading and transportation process, the center of gravity position and load distribution of the container are monitored in real time through a variety of sensors integrated inside the container. These sensors can accurately measure the placement, weight distribution and relative position changes of the goods in the container. The sensor data is transmitted to the control unit in real time for calculation and analysis. The control unit calculates the current center of gravity position and load distribution status through a dedicated algorithm. If abnormal conditions such as center of gravity offset or uneven load distribution are detected, the system will immediately issue control instructions to adjust the mechanical hydraulic equipment. The mechanical hydraulic equipment includes hydraulic push rods, support rods and other actuators, which can move the goods or reallocate the goods according to the instructions to ensure that the container remains stable and balanced during transportation;

[0053] In addition, during transportation, the system will continuously monitor the posture data and load status of the container. If it detects that the posture of the container is abnormal, such as the tilt angle exceeds the safe range or the load redistribution in the container causes the center of gravity to change, the system will automatically trigger the adjustment program and first try to restore the balance by moving the position of the goods inside the container. If the movement of the goods cannot completely solve the problem, the system can also adjust the counterweight device inside the container. These counterweight devices are specially designed movable modules that can quickly adjust the position and weight inside the container, so that the overall center of gravity of the container returns to a stable state.

[0054] In the same flight, multiple smart containers are interconnected through wireless communication technology to form a dynamic balance network. These containers are interconnected using wireless communication modules such as Wi-Fi, 5G or LoRa, and can communicate and exchange data with each other in real time, forming a multi-hop dynamic balance network.

[0055] Through the dynamic balancing network, each container can share and transmit its own status data in real time, including internal environmental parameters such as temperature, humidity, pressure, and information such as the container's geographical location and posture. Through these shared data, the system can dynamically monitor the load and environmental conditions of the entire container group. Based on these data, the control units of each container jointly calculate the optimal load balancing strategy and environmental control plan. The container group autonomously adjusts its own environmental control parameters according to the calculation results. For example, when a container detects that its internal temperature is too high, it can transmit this information to other containers through the dynamic balancing network. Other containers in the group may take corresponding measures according to their own temperature conditions, such as reducing their own heat output, or adjusting the internal air flow direction to jointly reduce the overall temperature. This coordinated adjustment ensures that the environmental conditions in all containers are maintained at the best state to ensure the safe transportation of goods.

[0056] At the same time, when it is detected that the load distribution of some containers is abnormal or the center of gravity is shifted, through the shared data, other containers in the group can adjust their respective loads or center of gravity positions to help restore the overall balance.

[0057] Embodiment 1:

[0058] Combined with Figure 2 Flowchart, a typical air container, whose dimensions are 3 meters long, 2.5 meters wide and 2.5 meters high (the corresponding three-dimensional space volume Ω is 18.75 cubic meters). A data acquisition system composed of multiple sensors is installed inside the container, including weight sensors, inertial measurement units IMU, laser rangefinders and ultrasonic sensors. These sensors are arranged on the bottom, walls and top of the container to monitor the weight distribution, center of gravity position, cargo height changes and space occupancy in real time.

[0059] Assume that the weight distribution function of any position point (x, y, Z) in the container is W(x, y, z, t), where x, y, z are three-dimensional spatial coordinates, and t is time, which indicates the time point of the current collection. At a certain time point (such as t = 10 minutes), the data acquisition system detects that there are multiple cargoes of different weights in the container. For example, cargoes weighing 200 kg, 150 kg, and 100 kg are placed at positions (1, 1, 0.5), (2, 1.5, 1), and (0.5, 2, 2), respectively. These weight data are obtained through weight sensors, and the weight density function of each position point can be expressed by dividing the mass of the cargo by the space volume it occupies. Assuming that the volume of each cargo is 0.5 cubic meters, 0.3 cubic meters, and 0.2 cubic meters, respectively, their weight densities at their respective positions are 400 kg / m3, 500 kg / m3, and 500 kg / m3, respectively.

[0060] In order to determine the current center of gravity coordinate vector C(t) of the container, the system uses a data fusion algorithm to solve the center of gravity position based on the data collected by all sensors in the container. The calculation formula is:

[0061]

[0062] Substituting the collected data into the formula, at time t = 10 minutes, the weight distribution function W (x, y, z, t) is expressed as the weight density of each position point, so the total mass M can be calculated as:

[0063]

[0064] Then, the component C of the coordinate vector C(t) of the center of gravity is x (t),C y (t),C z (t) can be expressed as:

[0065]

[0066] Substituting actual data into the calculation, we get:

[0067]

[0068] Therefore, the coordinate vector of the center of gravity position of the container at time t = 10 minutes is C(t) = (1.22, 1.39, 1) meters. Based on this center of gravity position, the system can use a dynamic balance algorithm to determine whether it is necessary to adjust the position of the cargo inside the container to maintain balance. For example, if it is detected that the center of gravity is offset beyond the safe range (the set safety range is that the center of gravity position does not exceed 0.5 meters), the system will immediately trigger the automatic control system of the mechanical hydraulic equipment to issue a command, start the hydraulic push rod or adjustable support rod, move the corresponding cargo or counterweight block, and return the center of gravity position to the safe range.

[0069] If the system needs to adjust the offset of the center of gravity position to no more than 0.2 meters, the adjustment step coefficient λ can be set in the range of 0.1 to 0.5. The adjustment strategy is implemented step by step according to the direction and magnitude of the center of gravity change. Dynamic balance is achieved through iterative adjustment to ensure the safety and stability of the container during flight.

[0070] In the air transport process, the system detects that the center of gravity position at time t = 10 minutes is C(t) = (1.22, 1.39, 1) meters, which exceeds the set safety range. At this time, the dynamic balance algorithm establishes a dynamic balance control model through the reinforcement learning algorithm of adaptive learning. The model predicts the future center of gravity change trend based on the combination of historical data and real-time data, and optimizes the cargo adjustment strategy. The optimization goal of the model is to minimize the nonlinear objective function J. The expression of the objective function is as follows:

[0071]

[0072] In this formula, J is the optimization objective function of dynamic balance control, which represents the comprehensive index of the overall performance of the system in the time interval 0 to T. The system hopes to minimize this function to ensure balance and stability; is the rate of change of the center of gravity position C(t) relative to time t, expressed as the modulus of a vector, and its value reflects the speed and direction of the center of gravity.

[0073] In this case, if the center of gravity changes drastically in a short period of time, that is, If the load is larger, the system will tend to make corresponding adjustments to reduce the rate of change; P(t) is the load distribution balance factor, and its value is calculated by the current load state. It indicates the uniformity of the positions of the goods in the container. If the goods are evenly distributed in the container, P(t) is smaller, otherwise it is larger; E(t) is the energy consumption function of the hydraulic equipment during the cargo adjustment process, indicating the energy consumed in performing each adjustment operation. In the current system, the energy consumption of each hydraulic equipment operation can be expressed as a function, E(t) = k·number of operations, where k is the energy consumption coefficient per unit operation, ranging from 5 to 10 joules.

[0074] To achieve the best balance and minimum energy consumption, the model adjusts the parameters α and β to balance the weight coefficients of load distribution and energy consumption in the optimization target. When α=0.3 and β=0.7, the system is more inclined to reduce energy consumption rather than pursue the ultimate load balance. When the center of gravity position deviates from the preset safety range or the load distribution is uneven, the algorithm generates adjustment instructions. For example, when the center of gravity deviation is detected to exceed the set safety range of 0.5 meters, the system calculates m / s, indicating that the center of gravity moves quickly and exceeds the allowed rate of change. P(t) = 0.6 indicates that the current load distribution is slightly unbalanced. E(t) = 7 joules is the average energy consumption for each adjustment operation.

[0075] The algorithm substitutes the above parameters into the formula:

[0076]

[0077] The calculation results are:

[0078]

[0079] In order to minimize the optimization objective function J, the system will generate a set of adjustment instructions, which include the direction, distance, speed and operation details of the required hydraulic equipment. Specifically, if the system decides to adjust the center of gravity position in a way that minimizes energy consumption, the instruction may require moving a lighter cargo (such as a 100 kg cargo located at 0.5, 2, 2) to the left by 0.2 meters to reduce the center of gravity offset; at the same time, the instruction sets the movement speed of the hydraulic push rod to 0.1 meters per second to ensure a safe operating speed and reduce unnecessary energy consumption.

[0080] After executing the adjustment command, the system monitors the center of gravity position again and sets the new center of gravity position as C ′ (t) = (1.1, 1.2, 1) meters, which meets the safety range requirements and the center of gravity changes at a speed of m / s, energy consumption reduced to 4 joules, the system evaluates that the current status has been significantly improved compared to the previous one, the algorithm verifies that the adjustment strategy is effective and records the data as a reference for future decision optimization.

[0081] Next is how the automatic control system of the mechanical hydraulic equipment executes adjustment instructions and continuously optimizes the state of the container through a closed-loop feedback mechanism after detecting the deviation of the center of gravity. In this example, a variety of mechanical hydraulic equipment is configured inside and at the bottom of the container, including hydraulic push rods, adjustable support rods and roller platforms. These devices can automatically move and redistribute the goods according to the instructions issued by the control unit ICU to achieve the balance control of the container.

[0082] In the previous stage, the system detected that at time t = 10 minutes, the center of gravity position was C(t) = (1.22, 1.39, 1) meters, which exceeded the set safety range. According to the adjustment instructions generated by the dynamic balancing algorithm, the mechanical hydraulic equipment began to move the 100 kg cargo located at 0.5, 2, 2 to the left by 0.2 meters, and the movement speed of the hydraulic push rod was 0.1 meters per second. At this time, the intelligent control unit ICU monitors the center of gravity changes in real time according to the sensor data, and feedbacks the adjusted center of gravity position and load status through the sensor to form a closed-loop control system.

[0083] Assume the center of gravity before adjustment is C n =(1.22,1.39,1) meters, the target center of gravity is C target =(1.0,1.2,1.0) meters. After the initial adjustment, the sensor detects the new center of gravity position as C n+1 =(1.15,1.35,1) meters. In order to verify the adjustment effect, the system uses the following iterative formula for calculation:

[0084] C n+1 =C n +ΔC n =C n +λ(C target -C n )

[0085] In this formula, ΔC n is the change in the center of gravity during this adjustment process, and λ is the step coefficient, which ranges from 0<λ≤1 and controls the amplitude of the adjustment. When setting the initial adjustment, the system selects λ=0.5 as the step coefficient and substitutes it into the formula for calculation:

[0086] ΔC n =λ(C target -C n )=0.5×((1.0,1.2,1.0)-(1.22,1.39,1))

[0087] The calculation results are:

[0088] ΔC n =0.5×(-0.22,-0.19,0)=(-0.11,-0.095,0)

[0089] Therefore, the adjusted center of gravity position is:

[0090] C n+1 =(1.22,1.39,1)+(-0.11,-0.095,0)=(1.11,1.295,1)

[0091] At this time, the system once again detected that the center of gravity position was still beyond the target range, indicating that this adjustment has not yet fully reached the preset balance state. Therefore, the system continues to make secondary iterative adjustments based on the feedback data. The adjustment command is updated, the hydraulic push rod movement direction remains unchanged, but the step coefficient is increased to λ = 0.7 to speed up the convergence speed. The new adjustment amount is:

[0092] ΔC n+1 =0.7×((1.0,1.2,1.0)-(1.11,1.295,1))

[0093] The calculation results are:

[0094] ΔC n+1 =0.7×(-0.11,-0.095,0)=(-0.077,-0.0665,0)

[0095] Therefore, the center of gravity position after the second adjustment is:

[0096] C n+2 =(1.11,1.295,1)+(-0.077,-0.0665,0)=(1.033,1.2285,1)

[0097] After the second adjustment, the center of gravity is closer to the target position C. target = (1.0, 1.2, 1.0) meters, but there is still a slight deviation. According to the allowable error range set by the system (such as within 0.05 meters), if the current deviation is within the allowable range, the system determines that the equilibrium state has been reached; otherwise, the system will continue to perform fine-tuning operations. If the deviation of the current state is set to be within the allowable error range, the system will stop further adjustments and record the step length, direction, energy consumption and time consumption data of this adjustment for future optimization and adjustment strategy learning.

[0098] Embodiment 2:

[0099] Combined with Figure 3 Flowchart, a wireless communication module is integrated in each air container. The module is configured with a unique ID and uses self-organizing network technology to enable the communication module of each container to autonomously discover and connect to other smart containers on the same flight, forming a multi-hop dynamic balance network. During the network formation process, each container notifies its current status information, including the center of gravity position, temperature, humidity, and load information, through a broadcast protocol.

[0100] Suppose there are 5 smart containers (numbered 1 to 5) on a flight, each of which is equipped with a wireless communication module. The communication radius of the communication module is r1 = 100 meters, r2 = 120 meters, r3 = 80 meters, r4 = 90 meters, and r5 = 110 meters. All containers are distributed in the cargo hold at different positions. In the initial stage of networking, each container sends its own status information through a broadcast protocol, including the current center of gravity position, temperature, humidity, load information, etc., and selects the optimal communication path based on link quality, signal strength, and network topology. At this time, it is necessary to calculate the communication topology quality function T of the multi-hop dynamic equilibrium network to describe the overall communication coverage quality of the network.

[0101] Set the distance d between every two containers ij (i, j represent different container numbers), such as the distance d between container 1 and container 2 12 = 70 meters, the distance d between container 2 and container 3 23 = 60 meters, the distance d between container 3 and container 4 34 = 100 meters, the distance d between container 4 and container 5 45 = 80 meters, the distance d between container 1 and container 5 15 = 110 meters. The signal strength function for each container is defined as Among them A i is the initial value of the signal strength, γ i is the signal attenuation coefficient, ranging from 0.01 to 0.05. The initial signal strength of each container is set to A1=50, A2=60, A3=40, A4=55, A5=45, and the signal attenuation coefficients are γ1=0.02, γ2=0.03, γ3=0.01, γ4=0.025, γ5=0.015.

[0102] According to the set communication topology description formula:

[0103]

[0104] Where T is the communication topology quality function of the entire network, Ω represents the three-dimensional space area where the container is located, N = 5 is the number of containers, and Θ(x) is the Heaviside step function, which is defined as: when x ≥ 0, Θ(x) = 1, otherwise it is 0.

[0105] Substitute the data into the formula for calculation. First, calculate the signal strength function value between each pair of containers. Take the signal strength between container 1 and container 2 as an example:

[0106] S1(d 12 )=50·e -0.02×70 ≈50·e-1.4 ≈50·0.2466≈12.33

[0107] S2(d 12 )=60·e -0.03×70 ≈60·e -2.1 ≈60·0.1225≈7.35

[0108] Therefore, when the communication path is from container 1 to container 2, due to d 12 =70≤r1=100, so Θ(r1-d 12 )=1, the final signal strength is Θ(r1-d 12 )·S1(d 12 )=1·12.33=12.33.

[0109] The signal strengths between the remaining containers are calculated in a similar manner to obtain the effective signal strengths between each pair of containers. All signal strengths are summed to obtain the numerical approximation of the integral equation, which is the value of the communication topology quality function T of the multi-hop dynamic equilibrium network:

[0110] T≈∫ Ω (12.33+7.35+…)dΩ

[0111] Through this calculation result, the system can analyze the communication coverage quality of the entire container group. If it finds that the signal quality of some paths is lower than the preset threshold, it will automatically adjust the communication path or enhance the robustness and stability of the network by adding relay nodes, thereby ensuring that during the transportation process, the status data of the container group can be transmitted and processed in real time and efficiently, ensuring the dynamic balance and environmental optimization of the containers.

[0112] Using a distributed ledger based on blockchain technology, each container runs a collaborative dynamic balance algorithm to calculate the optimal adjustment strategy based on shared data and status. In this system, each container will regularly broadcast its own status data (including center of gravity position, temperature, humidity and other parameters), and share this data securely with all other containers through blockchain technology to ensure data transparency and immutability.

[0113] Suppose there are 5 smart containers (numbered 1 to 5) on a flight, and the state vector of each container at time t is X(t) = [x1(t), x2(t), x3(t), x4(t), x5(t)], where x i(t) is the state variable of the i-th container, including the center of gravity position, temperature and humidity parameters. At time t = 0, the initial state of each container is: x1(0) = (1.0, 25, 40), x2(0) = (1.2, 26, 42), x3(0) = (1.1, 27, 43), x4(0) = (1.3, 24, 41), x5(0) = (1.15, 25, 40), where the first three components of each state vector represent the center of gravity position, temperature (unit: Celsius) and humidity (unit: percentage).

[0114] In order to optimize the overall performance of the container group, the system hopes to minimize the following optimization objective function J:

[0115]

[0116] In this formula, J is the optimization objective function, which represents the performance index of the entire container group in the time interval [0, T]. The goal is to minimize it to optimize the overall performance; is the rate of change of the i-th container state variable relative to time, reflecting the speed of change of the container state; Φ i (x i (t), X(t)) is the interaction function between the i-th container and other containers, which represents the interaction between containers due to changes in environment and load at time t; γ i is the corresponding weight coefficient, reflecting the importance of each container in the optimization objective, and the value range is set to 0.1 to 0.5.

[0117] To specify the optimization process, the weight coefficients of each container are set to γ1 = 0.2, γ2 = 0.3, γ3 = 0.1, γ4 = 0.4, and γ5 = 0.2. When the time t = 10 minutes, the state change rate is:

[0118] Mutual influence function: Φ i (x i The specific form of (t),X(t)) can be expressed as the sum of the squares of the temperature and humidity differences between containers, that is:

[0119]

[0120] At time t = 10 minutes, the temperature and humidity differences between containers are: The temperature difference matrix is:

[0121]

[0122] The humidity difference matrix is:

[0123]

[0124] Calculate the value of each interaction function by substituting, for example, for container 1:

[0125] Φ1(x1(t),X(t))=(1 2 +2 2 )+(2 2 +3 2 )+(1 2 +1 2 )=1+4+4+9+1+1=20

[0126] The interaction function values ​​of other containers are calculated similarly and then substituted into the optimization objective function:

[0127]

[0128] Calculation results:

[0129]

[0130] By minimizing the objective function J, the system decides to adjust the temperature and humidity settings of some containers, or reallocate the locations of some goods. The specific adjustment strategy is executed by the ICU control unit of each container. Set in the next time period, after adjustment, the state change rate of the container and the value of the mutual influence function are significantly reduced, and finally a better state balance is achieved.

[0131] When a container detects a center of gravity shift or abnormal environmental conditions, the system calculates the corresponding adjustment plan based on the collaborative control strategy and notifies other containers through the wireless network to ensure that all containers are adjusted in a coordinated manner. After each container performs the adjustment action, it will collect new status data in real time and feed it back to the dynamic balance network to form an adaptive learning closed-loop system. Through continuous data exchange and adjustment, the entire container group gradually tends to a stable balance state.

[0132] Assume that during the flight, the control unit ICU of container 1 (numbered 1) detects that the center of gravity position deviation exceeds the set safety threshold, and the internal temperature rises to 28°C, while the target temperature is 25°C. At this time, the state vector of container 1 is X1 = (1.1, 28, 40), where the state variables represent the center of gravity position, temperature (unit: Celsius) and humidity (unit: percentage). Container 1 notifies other containers 2 to 5 through the wireless network, and the current states of other containers are X2 = (1.0, 26, 42), X3 = (1.15, 27, 43), X4 = (1.2, 24, 41), and X5 = (1.1, 25, 40).

[0133] According to the collaborative control strategy, the system calculates the adjustment amount ΔX that needs to be performed for each container n , so that the entire container group tends to the target state X tre =(1.1,25,41). Set the adjustment step coefficient η = 0.6, and the system uses the following iterative relationship for adjustment:

[0134] ΔX n =η(X tre -X n )

[0135] For container 1, the adjustment ΔX1 is calculated as follows:

[0136] ΔX1=0.6×((1.1,25,41)-(1.1,28,40))

[0137] The calculation results are:

[0138] ΔX1=0.6×(0,-3,1)=(0,-1.8,0.6)

[0139] Therefore, the status of container 1 is updated to:

[0140] X 1,n+1 =X1+ΔX1=(1.1,28,40)+(0,-1.8,0.6)=(1.1,26.2,40.6)

[0141] Similarly, the adjustment ΔX2 for container 2 is calculated as:

[0142] ΔX2=0.6×((1.1,25,41)-(1.0,26,42))

[0143] The calculation results are:

[0144] ΔX2=0.6×(0.1,-1,-1)=(0.06,-0.6,-0.6)

[0145] Therefore, the status of container 2 is updated to:

[0146] X 2,n+1 =X2+ΔX2=(1.0,26,42)+(0.06,-0.6,-0.6)=(1.06,25.4,41.4)

[0147] Continue to perform similar calculations on other containers to obtain their respective state updates. After each adjustment, each container feeds back the new state vector to the dynamic balance network, and performs the next optimization adjustment based on the feedback results.

[0148] If the system detects that the center of gravity position, temperature, humidity and other state parameters have approached or reached the target state Xtre =(1.1,25,41), and the deviations of all parameters are within the set allowable range (e.g. within 0.5), then stop adjusting.

[0149] In this method, the selection of the adjustment step coefficient η is very important. Its value is between 0 and 1 (0<η≤1), which determines the magnitude of each adjustment. In this example, the value of η=0.6 is selected to speed up the convergence speed while ensuring the stability of the adjustment and avoid unnecessary energy consumption caused by too many iterations.

[0150] The above shows and describes the basic principles, main features and advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for air container control, characterized in that The following steps are involved: S1. Intelligent state perception and control: S1.

1. Integrate a multi-dimensional sensor network, including temperature, humidity, pressure, vibration, and position sensors, to sense the state of the air container in real time; the real-time sensing includes monitoring the internal and external environment of the container, external force impact or vibration; S1.2, the sensor data is transmitted in real time to the control unit ICU integrated in the container, and the control unit ICU triggers signal instructions including adjusting the temperature or issuing an alarm; S2. Dynamic load and balancing management: S2.

1. During loading and transportation, by monitoring the center of gravity and load distribution of the container, real-time sensor data is calculated and instructions are issued to control mechanical hydraulic equipment to adjust the cargo placement; S2.

2. During transportation, if the posture or load distribution of the container is detected to be abnormal, the position of the internal cargo or the counterweight device inside the container is adjusted to restore the balance state; S3. Container group collaborative control: S3.

1. Multiple smart containers on the same flight are interconnected through wireless communication technology to form a dynamic balancing network; loads and environmental conditions between containers are balanced through data exchange in the dynamic balancing network; S3.2, sharing data among containers, including temperature, pressure, and location, and based on the shared data, the container group jointly adjusts their respective environmental control parameters; The dynamic load and balance management includes a data acquisition system, a dynamic balance algorithm and an automatic control system for mechanical hydraulic equipment; the data acquisition system is composed of weight sensors, inertial measurement units (IMUs), laser rangefinders and ultrasonic sensors installed on the bottom, walls and top of the air container; The dynamic balance algorithm includes design and autonomous control decision-making and instruction issuance, wherein a dynamic balance control model is established using a reinforcement learning algorithm of adaptive learning, and the future center of gravity change trend is predicted and the cargo adjustment strategy is optimized through a combination of historical data and real-time data; the optimization path of the model is to achieve the minimum value of the nonlinear objective function: Among them, J is the optimization objective function of dynamic balance control, which represents the comprehensive index of the overall performance of the system in the time interval from 0 to T; is the rate of change of the center of gravity position C(t) relative to time t, reflecting the moving speed and direction of the center of gravity, expressed as the modulus of the vector; P(t) is the balance factor of the load distribution, indicating the uniformity of the positions of the cargoes in the container, and its value is calculated by the current load state; E(t) is the energy consumption function of the hydraulic equipment during cargo adjustment, indicating the energy consumed in performing each adjustment operation; α and β are adjustment parameters, representing the weight coefficients of load balance and energy consumption in the optimization objectives respectively; when it is detected that the center of gravity position deviates from the preset safety range or the load distribution is unbalanced, an adjustment instruction is generated according to the algorithm, and the instruction includes the direction, distance, speed and operation details of the required hydraulic equipment.

2. The method for air container control according to claim 1, characterized in that The data acquisition system monitors the weight distribution, center of gravity position, cargo height change and space occupancy of each area in the container in real time; including the use of a data fusion algorithm: assuming that the weight distribution function of any point (x, y, z) in the container is W(x, y, z, t), where x, y, z are three-dimensional spatial coordinates, and t is time, indicating the time point of the current acquisition, by solving the multidimensional integral: Where C(t) is the coordinate vector of the center of gravity of the container at time t, expressed as C(t) = (C x (t),C y (t),C z (t)), Ω represents the three-dimensional spatial volume of the container, and W(x, y, z, t) is the weight density function at the three-dimensional coordinate point (x, y, z) at time t.

3. The method for air container control according to claim 2, characterized in that The automatic control system of the mechanical hydraulic equipment includes mechanical hydraulic equipment integration and intelligent control and real-time feedback closed loop; Hydraulic push rods, adjustable support rods and roller platforms are configured at the bottom and inside of the container to move and redistribute the cargo according to the instructions issued by the control unit ICU. A closed-loop control system is formed by integrating the mechanical hydraulic equipment with the sensor network and ICU in the container. After the equipment executes the instructions, the sensor feeds back the adjusted center of gravity position and load status, and the adjustment effect is verified by an iterative formula: C n+1 =C n +ΔC n =C n +λ(C target -C n ) Among them, C n+1 is the center of gravity position after the n+1th adjustment, indicating the new center of gravity coordinates after adjustment; C n is the center of gravity position before the nth adjustment, indicating the currently known center of gravity coordinates; ΔC n is the change in the center of gravity during the nth adjustment, indicating the change in the center of gravity position during this adjustment; λ is the step coefficient, indicating the control parameter of the adjustment amplitude, and its value is between 0<λ≤1; C target It is the preset target center of gravity position, which indicates the coordinates of the center of gravity when the cargo inside the container is in a balanced state. If the target balance state is still not reached after adjustment, a second or multiple iterative adjustments are performed according to the feedback data until a balance state is reached.

4. The method for air container control according to claim 1, characterized in that The dynamic balancing network construction includes the construction of a multi-container wireless communication network, the coordinated processing of data exchange and dynamic balancing algorithms, and the coordinated balancing of multiple containers and environmental optimization; A wireless communication module is integrated in each air container, and the wireless communication module is configured with a unique identity identifier ID; self-organizing network technology is used to enable the communication module of each container to discover and connect to other smart containers on the same flight, forming a multi-hop dynamic balance network; During the network formation process, each container announces its current status through a broadcast protocol, including the center of gravity position, temperature, humidity and load information, and selects a communication path based on link quality, signal strength and network topology.

5. The method for air container control according to claim 4, characterized in that The collaborative processing method of data exchange and dynamic balancing algorithm includes: the data sharing process uses a distributed ledger based on blockchain technology, and each container runs a collaborative dynamic balancing algorithm and calculates an adjustment strategy based on the shared data and status.

6. The method for air container control according to claim 4, characterized in that The multi-container coordinated balancing and environmental optimization method comprises: When a container detects a center of gravity deviation or abnormal environmental conditions, an adjustment plan is calculated based on the collaborative control strategy and notified to other containers through the wireless network; after each container performs the adjustment action, new status data is collected in real time and fed back to the dynamic balance network, forming an adaptive learning closed-loop system; Through data exchange and adjustment, the entire container group tends to a stable equilibrium state; suppose that after each adjustment, the state vector of the container is updated to X n+1 =X n +ΔX n , where X n is the state vector before the nth adjustment, representing the state set of all containers before adjustment, ΔX n is the adjustment amount, indicating the state change during the adjustment process, and satisfies the following iterative relationship: ΔX n =η(X tre -X n ) Among them, η is the adjustment step coefficient, which controls the amplitude of each adjustment and its value range is between 0<η≤1; X tre is the target state vector, which represents the state set when the container group reaches the equilibrium state. The adjustment process is iterated until the target state is reached.

Images