A digital resource scheduling and whole-process tracing system for port piece bulk cargo handling

Abstract

The present application relates to the technical field of cargo scheduling, in particular to a port piece bulk cargo handling digital resource scheduling and whole-process tracing system, which comprises a cargo digital unit, a cargo dynamic interaction unit, a scheduling optimization unit and a tracing verification unit. The present application takes the physical interaction characteristics of cargo and environment as the core driving element, captures the cargo state and environmental disturbance in real time through multi-modal sensor data collection, realizes the full-dimensional digital mapping from static attributes to dynamic behavior, and the physical feature closed-loop driving mechanism not only improves the accuracy of scheduling decision, but also builds a tracing system with chaotic characteristics through irreversible environmental disturbance fingerprint technology, effectively solving the problems of easy tampering and difficult reproduction of traditional tracing system. Secondly, a three-level risk perception scheduling system is innovatively designed, which deeply couples risk assessment and resource scheduling. This layered response strategy not only guarantees operation safety, but also avoids resource waste.

Description

Technical Field

[0001] This invention relates to the field of cargo scheduling technology, specifically to a digital resource scheduling and full-process traceability system for port general cargo handling. Background Technology

[0002] Port cargo handling and scheduling refers to the process of managing, arranging, and scheduling various bulk and miscellaneous goods imported and exported during port operations. This process typically includes loading and unloading, storage, inspection, sorting, and transportation of goods. Cargo handling involves counting, classifying, and recording goods to ensure their safety and accuracy, while scheduling involves rationally arranging the flow and operational sequence of goods within the port based on their characteristics and destinations to improve work efficiency, reduce waiting time, and optimize resource utilization. Effective cargo handling and scheduling can ensure the smooth operation of port operations, improve overall logistics efficiency, and reduce operating costs.

[0003] However, existing technologies have several shortcomings. First, relying on centralized databases or traditional encryption technologies carries the risk of data tampering and excessive computational resource consumption. Second, big data collection often relies on single sensors, failing to comprehensively capture the dynamic interaction characteristics of goods and the environment, resulting in fragmented and easily forged traceability information. Furthermore, traditional scheduling uses static rules or single indicators for decision-making, lacking the ability to perceive and respond to dynamic risks in real time and in a tiered manner. This leads to reliance on manual intervention in high-risk scenarios, resulting in severe response delays, while in low-risk scenarios, resource allocation may be unreasonable, leading to equipment idling and energy waste. These shortcomings significantly affect the efficiency and safety of port operations.

[0004] Based on this, the present invention provides a digital resource scheduling and full-process traceability system for port general cargo handling to solve the aforementioned technical problems. Summary of the Invention

[0005] The purpose of this invention is to provide a digital resource scheduling and full-process traceability system for port general cargo handling. This invention takes the physical interaction characteristics of cargo and environment as the core driving element, and captures cargo status and environmental disturbances in real time through a multimodal sensor network to achieve full-dimensional digital mapping from static attributes to dynamic behavior. This closed-loop driving mechanism of physical characteristics not only improves the accuracy of scheduling decisions, but also constructs a traceability system with chaotic characteristics through irreversible environmental disturbance fingerprint technology, effectively solving the problems of easy tampering and difficulty in reproduction of traditional traceability systems. Secondly, it innovatively designs a three-level risk perception scheduling system, which deeply couples risk assessment with resource scheduling. This hierarchical response strategy not only ensures operational safety, but also avoids resource waste.

[0006] To achieve the above objectives, the present invention provides the following technical solution: This invention provides a digital resource scheduling and full-process traceability system for port general cargo handling, including a cargo digitization unit, a cargo dynamic interaction unit, a scheduling optimization unit, and a traceability verification unit. Specifically: The cargo digitization unit collects big data, measures the mass center coordinates, spatial location, and spectral characteristics of the cargo through multimodal sensor fusion and physical data initialization, integrates this data into structured data, and generates a cargo ID. The cargo dynamic interaction unit captures the dynamic physical characteristics of the cargo during its movement, obtains a physical feature matrix, divides the port operation area, generates multiple three-dimensional grids, maps the structured data of the cargo to the corresponding grids, and then assigns the physical feature matrix to the grid where the cargo is located. The scheduling optimization unit calculates the cargo stacking risk, thermal safety risk, and equipment operation risk respectively, obtains a comprehensive risk index, determines the risk level based on the comprehensive risk index, generates a multi-level scheduling strategy, dynamically adjusts the optimization weights according to the risk level, and outputs the optimal scheduling scheme. The traceability verification unit generates multi-level fingerprints during the cargo scheduling process. After the cargo passes through a key node, the multi-level fingerprints of the cargo are obtained again, and quantum random numbers are introduced for similarity verification to generate a verification report.

[0007] It should be noted that after obtaining structured data, the cargo digitization unit will upload the structured data to the cargo dynamic interaction unit and the scheduling optimization unit. The cargo digitization unit includes a multimodal sensing module, a cargo feature processing module, and a data integration module, wherein: The multimodal sensing module acquires the pressure data, spatial location data, and spectral image of the cargo, and unifies the timestamps of the three types of data. In this embodiment, an 8×8 grid-type pressure sensor array can be arranged in the cargo loading area, with each sensor having an accuracy of ±0.1 kPa, to obtain pressure data; ultra-wideband positioning tags are installed on the cargo, and four base stations are deployed in the port operation area to form a positioning network. The cargo tags communicate with each base station, and spatial location data can be obtained by measuring the transmission time of the signal from the tag to each base station; while spectral images are obtained through a multispectral camera, which performs line scanning on the surface of the cargo at a speed of 1 m / s as the cargo passes by.

[0008] The cargo feature processing module calculates the mass center coordinates and spatial coordinates of the cargo, extracts the average reflectance value of the cargo in key bands from the spectral image, and obtains spectral features. The coordinates of the cargo's center of mass in the horizontal plane are determined by calculating the weighted average of the pressure values ​​of all sensors: the x-coordinate is the sum of the products of the x-coordinates of each sensor and their pressure values ​​divided by the total pressure value; the y-coordinate is calculated similarly; and the vertical direction (z-coordinate) is determined by the contact height between the bottom surface of the cargo and the sensor array. Finally, the coordinates (x, y, z) of the cargo's center of mass are obtained.

[0009] Spatial coordinates can be determined using a time-of-arrival (TOA) positioning algorithm. Based on the signal propagation time and base station coordinates, the three-dimensional spatial coordinates (x, y, z) of the cargo tag can be calculated, i.e.: ; In the formula: ( () represents the coordinates of the base station, and c represents the speed of light. This is the time difference of arrival.

[0010] The data integration module obtains the current timestamp of the goods, generates a multi-digit random number and combines it with the current timestamp to obtain structured data, verifies and stores the structured data, and outputs it to the goods dynamic interaction unit.

[0011] The cargo dynamic interaction unit includes a dynamic physical feature module and a region division module, wherein: The dynamic physical feature module is used to integrate vibration sensors, vision sensors, magnetic field sensors and air pressure sensors to synchronously collect four types of data during the movement of goods, align the four types of data according to time, and obtain the physical feature matrix of the goods. After acquiring vibration sensor data, a fast Fourier transform is performed to convert the time-domain signal to the frequency domain, calculate the spectral energy distribution, and obtain the vibration energy density value. The area division module: determines the port's operating area, divides the operating area into multiple three-dimensional grids, determines the grid position based on the mass center coordinates and spatial position of the cargo, binds the physical feature matrix and spectral features corresponding to the cargo, and calculates the load index of each grid.

[0012] The load index is obtained by multiplying the mass density by the vibration energy, while the mass density is obtained by dividing the mass of the cargo by the grid volume.

[0013] The scheduling optimization unit includes a risk perception module, a risk scheduling module, and an optimal scheduling module, wherein: The risk perception module calculates six types of risk data for the goods and provides risk warnings based on these data. The six types of risk data include the standard deviation of the stacking height, tilt angle, maximum temperature value, hot spot area, acoustic signature characteristics, and the Euclidean distance of the baseline acoustic signature characteristics. It should be noted that a radar can be deployed to acquire three-dimensional point cloud data of the cargo stack, generate a point cloud map of the cargo stack, and then calculate the height standard deviation and tilt angle, which respectively reflect the flatness and stability of the stack. When the height standard deviation exceeds 0.1 meters or the tilt angle exceeds 15 degrees, a stability risk warning is triggered. Simultaneously, infrared thermal imagers are used to monitor the surface temperature distribution of the cargo, recording the highest temperature value and the area of ​​hot spots. A thermal safety risk warning is triggered when the highest temperature exceeds 80℃ or the hot spot area exceeds 2 square meters. In addition, the device's operating voiceprint is collected by an acoustic sensor, and the Euclidean distance between the current voiceprint feature and the reference voiceprint feature is calculated (normalized and divided by 5.0). When the distance exceeds 1.0, a device operation risk warning is triggered.

[0014] The risk scheduling module calculates the comprehensive risk index of the goods and determines the risk level. Based on the risk level, it generates a multi-level scheduling strategy. The optimal scheduling module defines the objective function of the scheduling scheme, formulates the scheduling scheme guide, dynamically adjusts the optimization weights according to the risk level, and feeds them back to the objective function of the scheduling scheme to obtain the optimal scheduling scheme.

[0015] The risk scheduling module calculates the comprehensive risk index of the goods and determines the risk level. Based on the risk level, it generates a multi-level scheduling strategy. The specific process is as follows: The material of the cargo is determined based on its spectral characteristics, vibration sensor data, and magnetic field sensor data. Obtain historical accident data for the corresponding material, and define the weights of cargo stacking risk, thermal safety risk, and equipment operation risk based on the proportion of risk types within the historical accident data; The risks of cargo stacking, thermal safety, and equipment operation are normalized and summed with corresponding risk weights to obtain a comprehensive risk index. Then, the risk level is determined, and a multi-level scheduling strategy is generated based on the risk level.

[0016] The specific rules for determining the risk level are as follows: Risk level thresholds are set as M and N, where M > N > 0, M is 0.6 and N is 0.3. If the comprehensive risk index is ≤ N, it is considered low risk, and the standard scheduling procedure is adopted. The standard scheduling procedure can be formulated and slightly adjusted according to the specific scheduling situation. If N < comprehensive risk index ≤ M, it is considered medium risk. If advanced scheduling procedures are adopted, it is necessary to replan the route, set a safety distance, and develop a backup plan. If the comprehensive risk index is greater than M, it is considered high risk and enters the emergency dispatch process. In this case, immediate deployment is required to isolate the risk area, temporarily suspend the dispatch process, and conduct an inspection.

[0017] The objective function of the scheduling scheme is: Maximize overall benefit Z = w1 × efficiency + w2 × safety - w3 × energy consumption Efficiency: expressed as the reciprocal of the task completion time; Safety: Represented by subtracting the comprehensive risk index from 1; Energy consumption: The ratio of total power to rated power of equipment; Weights are dynamically adjusted based on risk level: High risk: Safety weight w2=0.7, efficiency weight w1=0.2, energy consumption weight w3=0.1; Medium risk: Efficiency weight w1=0.5, safety weight w2=0.3, energy consumption weight w3=0.2; Low risk: Efficiency weight w1=0.7, safety weight w2=0.1, energy consumption weight w3=0.2.

[0018] The traceability verification unit includes a multi-level fingerprint generation module, a data acquisition module, and a verification data module, wherein: The multi-level fingerprint generation module generates a first-level fingerprint based on the spectral characteristics of the cargo, calculates the statistics of the cargo's vibration energy entropy, thermal gradient variance, and acoustic resonance skewness, and combines these three statistics to form a second-level fingerprint; and constructs a Markov chain by combining the time series of the three statistics as a third-level fingerprint. The data acquisition module: sets key nodes in the cargo scheduling process, and acquires the multi-level fingerprint of the cargo again after the cargo passes through the key nodes; The verification data module receives multi-level fingerprints corresponding to the multi-level fingerprint generation module and the data acquisition module, introduces quantum random numbers as perturbation factors, performs similarity calculations, and generates a verification report.

[0019] The multi-level fingerprint generation module generates a first-level fingerprint based on the spectral characteristics of the cargo, calculates the statistics of the cargo's vibration energy entropy, thermal gradient variance, and acoustic resonance skewness, and combines these three statistics to form a second-level fingerprint; then, it constructs a Markov chain by combining the time series of the three statistics as the third-level fingerprint. The specific process is as follows: Primary fingerprint: Multiple key band reflectance vectors are collected from the spectral characteristics of the goods and concatenated to obtain a string. The string is then processed using a hash algorithm to generate a multi-digit hexadecimal string, which is then truncated to serve as the primary fingerprint. A hash algorithm adds a specific identifier and length information to the end of a string to make its length meet the algorithm's requirements. The padded data is then divided into blocks of fixed size, and the data is gradually obfuscated through multiple rounds of non-linear operations, ultimately outputting a hash value of fixed length.

[0020] Secondary fingerprint: The Shannon entropy of energy distribution, the variance of cargo temperature data, and the skewness of acoustic resonance amplitude are calculated. These three statistics are combined into a three-dimensional vector, which is regarded as the secondary fingerprint. It should be noted that the formulas for calculating these three statistics are as follows: Shannon entropy of energy distribution: ; In the formula: Let n be the energy of the i-th frequency band, and n be the total number of frequency bands. Variance of cargo temperature data: ; In the formula: The temperature gradient value of the i-th pixel The mean value of all pixels is N, where N is the total number of pixels. Skewness of acoustic resonance amplitude: ; In the formula: Let be the resonance amplitude at the i-th time point. The mean value across all time points. The standard deviation is denoted as .

[0021] Level 3 fingerprint: Discretize the time series data of three statistics, divide the continuous values ​​into five levels of states, count the transition frequencies between these states, construct a transition frequency matrix and convert it into a probability matrix, expand the three probability matrices row by row and concatenate them to obtain the level 3 fingerprint.

[0022] The verification data module receives multi-level fingerprints from the multi-level fingerprint generation module and the data acquisition module, introduces quantum random numbers as perturbation factors, performs similarity calculations, and generates a verification report. The specific process is as follows: First, the multi-level fingerprints output by the multi-level fingerprint generation module are used as historical multi-level fingerprints, and the multi-level fingerprints output by the data acquisition module are used as real-time multi-level fingerprints. The Hamming distance of the first-level fingerprint and the cosine similarity of the second-level fingerprint are verified sequentially. Then, the probability matrix similarity between the historical multi-level fingerprint and the real-time multi-level fingerprint is calculated to obtain the similarity value. Based on the similarity value, the degree of anomaly is determined and a verification report is generated.

[0023] The similarity of the probability matrices between historical multi-level fingerprints and real-time multi-level fingerprints is calculated using the following formula: ; In the formula: F is the square root of the sum of the squares of all elements of the matrix. This is the transpose of the real-time matrix.

[0024] Compared with the prior art, the beneficial effects of the present invention are: This invention uses the physical interaction characteristics between goods and the environment as the core driving element, and captures the status of goods and environmental disturbances in real time through a multimodal sensor network to achieve a full-dimensional digital mapping from static attributes to dynamic behavior. This closed-loop driving mechanism of physical characteristics not only improves the accuracy of scheduling decisions, but also constructs a traceability system with chaotic characteristics through irreversible environmental disturbance fingerprint technology, effectively solving the problems of easy tampering and difficulty in reproduction in traditional traceability systems. Secondly, it innovatively designs a three-level risk perception scheduling system, which deeply couples risk assessment with resource scheduling. This hierarchical response strategy ensures operational safety and avoids resource waste. Attached Figure Description

[0025] Figure 1 This is a system diagram of a digital resource scheduling and full-process traceability system for port general cargo handling according to the present invention; Figure 2 This is a flowchart of the scheduling optimization unit in a digital resource scheduling and full-process traceability system for port general cargo handling according to the present invention; Figure 3 This is a diagram illustrating the overall system architecture of a digital resource scheduling and end-to-end traceability system for port general cargo handling according to the present invention. Figure 4 This is a traceability verification flowchart for a digital resource scheduling and full-process traceability system for port general cargo handling according to the present invention. Detailed Implementation

[0026] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0027] Example: like Figures 1-4 As shown, this embodiment provides a digital resource scheduling and full-process traceability system for port general cargo handling, including a cargo digitization unit, a cargo dynamic interaction unit, a scheduling optimization unit, and a traceability verification unit, wherein: like Figure 3 As shown, the cargo digitization unit: collects big data, and through multimodal sensor fusion and physical data initialization, measures the cargo's center of mass coordinates, spatial location, and spectral characteristics, integrates them into structured data, and generates a cargo ID; The cargo dynamic interaction unit captures the dynamic physical characteristics of the cargo during its movement, obtains the physical characteristic matrix of the cargo, divides the port operation area, generates multiple three-dimensional grids, maps the structured data of the cargo to the corresponding grids, and then assigns the physical characteristic matrix to the grid where the cargo is located. The scheduling optimization unit calculates cargo stacking risk, thermal safety risk, and equipment operation risk respectively to obtain a comprehensive risk index. Based on the comprehensive risk index, it determines the risk level, generates a multi-level scheduling strategy, dynamically adjusts the optimization weight according to the risk level, and outputs the optimal scheduling scheme. The traceability verification unit generates multi-level fingerprints during the cargo scheduling process. After the cargo passes through a key node, it acquires the multi-level fingerprints of the cargo again, introduces quantum random numbers for similarity verification, and generates a verification report.

[0028] It should be noted that after obtaining structured data, the cargo digitization unit will upload the structured data to the cargo dynamic interaction unit and the scheduling optimization unit. The cargo digitization unit includes a multimodal sensing module, a cargo feature processing module, and a data integration module, wherein: The multimodal sensing module acquires the pressure data, spatial location data, and spectral image of the cargo, and unifies the timestamps of the three types of data. In this embodiment, an 8×8 grid-type pressure sensor array can be arranged in the cargo loading area, with each sensor having an accuracy of ±0.1 kPa, to obtain pressure data; ultra-wideband positioning tags are installed on the cargo, and four base stations are deployed in the port operation area to form a positioning network. The cargo tags communicate with each base station, and spatial location data can be obtained by measuring the transmission time of the signal from the tag to each base station; while spectral images are obtained through a multispectral camera, which performs line scanning on the surface of the cargo at a speed of 1 m / s as the cargo passes by.

[0029] The cargo feature processing module calculates the mass center coordinates and spatial coordinates of the cargo, extracts the average reflectance value of the cargo in key bands from the spectral image, and obtains spectral features. The coordinates of the cargo's center of mass in the horizontal plane are determined by calculating the weighted average of the pressure values ​​of all sensors: the x-coordinate is the sum of the products of the x-coordinates of each sensor and their pressure values ​​divided by the total pressure value; the y-coordinate is calculated similarly; and the vertical direction (z-coordinate) is determined by the contact height between the bottom surface of the cargo and the sensor array. Finally, the coordinates (x, y, z) of the cargo's center of mass are obtained.

[0030] Spatial coordinates can be determined using a time-of-arrival (TOA) positioning algorithm. Based on the signal propagation time and base station coordinates, the three-dimensional spatial coordinates (x, y, z) of the cargo tag can be calculated, i.e.: ; In the formula: ( () represents the coordinates of the base station, and c represents the speed of light. This is the time difference of arrival.

[0031] The data integration module obtains the current timestamp of the goods, generates a multi-digit random number and combines it with the current timestamp to obtain structured data, verifies and stores the structured data, and outputs it to the goods dynamic interaction unit.

[0032] The cargo dynamic interaction unit includes a dynamic physical feature module and a region division module, wherein: The dynamic physical feature module is used to integrate vibration sensors, vision sensors, magnetic field sensors and air pressure sensors to synchronously collect four types of data during the movement of goods, align the four types of data according to time, and obtain the physical feature matrix of the goods. After acquiring vibration sensor data, a fast Fourier transform is performed to convert the time-domain signal to the frequency domain, calculate the spectral energy distribution, and obtain the vibration energy density value. The area division module: determines the port's operating area, divides the operating area into multiple three-dimensional grids, determines the grid position based on the mass center coordinates and spatial position of the cargo, binds the physical feature matrix and spectral features corresponding to the cargo, and calculates the load index of each grid.

[0033] The load index is obtained by multiplying the mass density by the vibration energy, while the mass density is obtained by dividing the mass of the cargo by the grid volume.

[0034] like Figure 2 As shown, the scheduling optimization unit includes a risk perception module, a risk scheduling module, and an optimal scheduling module, wherein: The risk perception module calculates six types of risk data for the goods and provides risk warnings based on these data. The six types of risk data include the standard deviation of the stacking height, tilt angle, maximum temperature value, hot spot area, acoustic signature characteristics, and the Euclidean distance of the baseline acoustic signature characteristics. It should be noted that a radar can be deployed to acquire three-dimensional point cloud data of the cargo stack, generate a point cloud map of the cargo stack, and then calculate the height standard deviation and tilt angle, which respectively reflect the flatness and stability of the stack. When the height standard deviation exceeds 0.1 meters or the tilt angle exceeds 15 degrees, a stability risk warning is triggered. Simultaneously, infrared thermal imagers are used to monitor the surface temperature distribution of the cargo, recording the highest temperature value and the area of ​​hot spots. A thermal safety risk warning is triggered when the highest temperature exceeds 80℃ or the hot spot area exceeds 2 square meters. In addition, the device's operating voiceprint is collected by an acoustic sensor, and the Euclidean distance between the current voiceprint feature and the reference voiceprint feature is calculated (normalized and divided by 5.0). When the distance exceeds 1.0, a device operation risk warning is triggered.

[0035] The risk scheduling module calculates the comprehensive risk index of the goods and determines the risk level. Based on the risk level, it generates a multi-level scheduling strategy. The optimal scheduling module defines the objective function of the scheduling scheme, formulates the scheduling scheme guide, dynamically adjusts the optimization weights according to the risk level, and feeds them back to the objective function of the scheduling scheme to obtain the optimal scheduling scheme.

[0036] The risk scheduling module calculates the comprehensive risk index of the goods and determines the risk level. Based on the risk level, it generates a multi-level scheduling strategy. The specific process is as follows: The material of the cargo is determined based on its spectral characteristics, vibration sensor data, and magnetic field sensor data. Obtain historical accident data for the corresponding material, and define the weights of cargo stacking risk, thermal safety risk, and equipment operation risk based on the proportion of risk types within the historical accident data; The risks of cargo stacking, thermal safety, and equipment operation are normalized and summed with corresponding risk weights to obtain a comprehensive risk index. Then, the risk level is determined, and a multi-level scheduling strategy is generated based on the risk level.

[0037] The specific rules for determining the risk level are as follows: Risk level thresholds are set as M and N, where M > N > 0. In this embodiment, M is 0.6 and N is 0.3. If the comprehensive risk index is ≤ N, it is considered low risk, and the standard scheduling procedure is adopted. The standard scheduling procedure can be formulated and slightly adjusted according to the specific scheduling situation. If N < comprehensive risk index ≤ M, it is considered medium risk. If advanced scheduling procedures are adopted, it is necessary to replan the route, set a safety distance, and develop a backup plan. If the comprehensive risk index is greater than M, it is considered high risk and enters the emergency dispatch process. In this case, immediate deployment is required to isolate the risk area, temporarily suspend the dispatch process, and conduct an inspection.

[0038] The objective function of the scheduling scheme is: Maximize overall benefit Z = w1 × efficiency + w2 × safety - w3 × energy consumption Efficiency: expressed as the reciprocal of the task completion time; Safety: Represented by subtracting the comprehensive risk index from 1; Energy consumption: The ratio of total power to rated power of equipment; Weights are dynamically adjusted based on risk level: High risk: Safety weight w2=0.7, efficiency weight w1=0.2, energy consumption weight w3=0.1; Medium risk: Efficiency weight w1=0.5, safety weight w2=0.3, energy consumption weight w3=0.2; Low risk: Efficiency weight w1=0.7, safety weight w2=0.1, energy consumption weight w3=0.2.

[0039] As a concrete example, suppose a port handles two types of cargo simultaneously: steel cargo, which is easily stacked and unstable with low thermal risk, and timber cargo, which is easily stacked and has high thermal risk but stable stacking. The sensor data are shown in Table 1. Table 1: Sensor Data Sheet ; These risk indicators were processed as shown in Table 2: Table 2: Risk Indicator Processing Table ; Among them, when the division of each risk indicator by the risk threshold is greater than 1, the value is directly taken as 1; By statistically analyzing the proportion of each risk type in historical accidents and using it as the risk weight, the stack stability risk is 0.5, the thermal safety risk is 0.3, and the equipment operation risk is 0.2. The comprehensive risk index for steel products is: 0.5×1 + 0.3×0.75 + 0.2×0.9 = 0.905; The comprehensive risk index for timber goods is: 0.5×0.8+0.3×1+0.2×0.7=0.84; In summary, all of the above are high-risk, namely: safety weight w2=0.7, efficiency weight w1=0.2, and energy consumption weight w3=0.1.

[0040] The traceability verification unit includes a multi-level fingerprint generation module, a data acquisition module, and a verification data module, wherein: The multi-level fingerprint generation module generates a first-level fingerprint based on the spectral characteristics of the cargo, calculates the statistics of the cargo's vibration energy entropy, thermal gradient variance, and acoustic resonance skewness, and combines these three statistics to form a second-level fingerprint; and constructs a Markov chain by combining the time series of the three statistics as a third-level fingerprint. The data acquisition module: sets key nodes in the cargo scheduling process, and acquires the multi-level fingerprint of the cargo again after the cargo passes through the key nodes; The verification data module receives multi-level fingerprints corresponding to the multi-level fingerprint generation module and the data acquisition module, introduces quantum random numbers as perturbation factors, performs similarity calculations, and generates a verification report.

[0041] The multi-level fingerprint generation module generates a first-level fingerprint based on the spectral characteristics of the cargo, calculates the statistics of the cargo's vibration energy entropy, thermal gradient variance, and acoustic resonance skewness, and combines these three statistics to form a second-level fingerprint; then, it constructs a Markov chain by combining the time series of the three statistics as the third-level fingerprint. The specific process is as follows: Primary fingerprint: Multiple key band reflectance vectors are collected from the spectral characteristics of the goods and concatenated to obtain a string. The string is then processed using a hash algorithm to generate a multi-digit hexadecimal string, which is then truncated to serve as the primary fingerprint. A hash algorithm adds a specific identifier and length information to the end of a string to make its length meet the algorithm's requirements. The padded data is then divided into blocks of fixed size, and the data is gradually obfuscated through multiple rounds of non-linear operations, ultimately outputting a hash value of fixed length.

[0042] Secondary fingerprint: The Shannon entropy of energy distribution, the variance of cargo temperature data, and the skewness of acoustic resonance amplitude are calculated. These three statistics are combined into a three-dimensional vector, which is regarded as the secondary fingerprint. It should be noted that the formulas for calculating these three statistics are as follows: Shannon entropy of energy distribution: ; In the formula: Let n be the energy of the i-th frequency band, and n be the total number of frequency bands. Variance of cargo temperature data: ; In the formula: The temperature gradient value of the i-th pixel The mean value of all pixels is N, where N is the total number of pixels. Skewness of acoustic resonance amplitude: ; In the formula: Let be the resonance amplitude at the i-th time point. The mean value across all time points. The standard deviation is denoted as .

[0043] Level 3 fingerprint: Discretize the time series data of three statistics, divide the continuous values ​​into five levels of states, count the transition frequencies between these states, construct a transition frequency matrix and convert it into a probability matrix, expand the three probability matrices row by row and concatenate them to obtain the level 3 fingerprint.

[0044] The verification data module receives multi-level fingerprints from the multi-level fingerprint generation module and the data acquisition module, introduces quantum random numbers as perturbation factors, performs similarity calculations, and generates a verification report. The specific process is as follows: The significance of introducing quantum random numbers as a perturbation factor lies in the fact that they are generated based on the principles of quantum mechanics, possess true randomness, and cannot be predicted or copied. This enhances the security of the verification system, preventing attackers from forging fingerprints using fixed patterns.

[0045] First, the multi-level fingerprints output by the multi-level fingerprint generation module are used as historical multi-level fingerprints, and the multi-level fingerprints output by the data acquisition module are used as real-time multi-level fingerprints. The Hamming distance of the first-level fingerprint and the cosine similarity C of the second-level fingerprint are verified sequentially. Then, the probability matrix similarity between the historical multi-level fingerprint and the real-time multi-level fingerprint is calculated to obtain the similarity value. Based on the similarity value, the degree of anomaly is determined and a verification report is generated.

[0046] Verify the Hamming distance of the primary fingerprint: perform hash alignment, convert to binary mode, compare bit by bit, and count the number of different bits, which is the number of difference bits. If the number of difference bits is ≥4, it is considered an anomaly. If the cosine similarity is ≤0.9, it is considered an anomaly. The similarity of the probability matrices between historical multi-level fingerprints and real-time multi-level fingerprints is calculated using the following formula: ; In the formula: F is the square root of the sum of the squares of all elements of the matrix. This is the transpose of the real-time matrix.

[0047] Minor anomalies: 0.75≤S<0.85, increase monitoring frequency.

[0048] Moderate abnormality: 0.6≤S<0.75, triggering manual inspection.

[0049] Serious abnormality: S < 0.6, stop operation and re-inspect.

[0050] The priority order for three-level fingerprints is as follows: the first-level fingerprint has the highest priority, followed by the third-level fingerprint, and the second-level fingerprint serves as a supplement. That is, if the second-level fingerprint record is abnormal, it can be corrected by the third-level fingerprint.

[0051] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0052] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A digital resource scheduling and full-process traceability system for port general cargo handling, characterized in that, It includes a cargo digitization unit, a cargo dynamic interaction unit, a scheduling optimization unit, and a traceability verification unit, among which: The cargo digitization unit: collects big data, and through multimodal sensor fusion and physical data initialization, measures the cargo's center of mass coordinates, spatial location, and spectral characteristics, integrates them into structured data, and generates a cargo ID; The cargo dynamic interaction unit captures the dynamic physical characteristics of the cargo during its movement, obtains the physical characteristic matrix of the cargo, divides the port operation area, generates multiple three-dimensional grids, maps the structured data of the cargo to the corresponding grids, and then assigns the physical characteristic matrix to the grid where the cargo is located. The scheduling optimization unit calculates cargo stacking risk, thermal safety risk, and equipment operation risk respectively to obtain a comprehensive risk index. Based on the comprehensive risk index, it determines the risk level, generates a multi-level scheduling strategy, dynamically adjusts the optimization weight according to the risk level, and outputs the optimal scheduling scheme. The traceability verification unit generates multi-level fingerprints during the cargo scheduling process. After the cargo passes through a key node, it acquires the multi-level fingerprints of the cargo again, introduces quantum random numbers for similarity verification, and generates a verification report.

2. The port general cargo handling digital resource scheduling and full-process traceability system according to claim 1, characterized in that, The cargo digitization unit includes a multimodal sensing module, a cargo feature processing module, and a data integration module, wherein: The multimodal sensing module acquires the pressure data, spatial location data, and spectral image of the cargo, and unifies the timestamps of the three types of data. The cargo feature processing module calculates the mass center coordinates and spatial coordinates of the cargo, extracts the average reflectance value of the cargo in key bands from the spectral image, and obtains spectral features. The data integration module obtains the current timestamp of the goods, generates a multi-digit random number and combines it with the current timestamp to obtain structured data, verifies and stores the structured data, and outputs it to the goods dynamic interaction unit.

3. The port general cargo handling digital resource scheduling and full-process traceability system according to claim 2, characterized in that, The cargo dynamic interaction unit includes a dynamic physical feature module and a region division module, wherein: The dynamic physical feature module is used to integrate vibration sensors, vision sensors, magnetic field sensors and air pressure sensors to synchronously collect four types of data during the movement of goods, align the four types of data according to time, and obtain the physical feature matrix of the goods. The area division module: determines the port operation area, divides the operation area to obtain multiple three-dimensional grids, determines the grid position based on the mass center coordinates and spatial position of the cargo, binds the physical feature matrix and spectral features corresponding to the cargo, and calculates the load index of each grid.

4. The port general cargo handling digital resource scheduling and full-process traceability system according to claim 1, characterized in that, The scheduling optimization unit includes a risk perception module, a risk scheduling module, and an optimal scheduling module, wherein: The risk perception module calculates six types of risk data for the goods and provides risk warnings based on these data. The six types of risk data include the standard deviation of the stacking height, tilt angle, maximum temperature value, hot spot area, acoustic signature characteristics, and the Euclidean distance of the baseline acoustic signature characteristics. The risk scheduling module calculates the comprehensive risk index of the goods and determines the risk level. Based on the risk level, it generates a multi-level scheduling strategy. The optimal scheduling module defines the objective function of the scheduling scheme, formulates the scheduling scheme guide, dynamically adjusts the optimization weights according to the risk level, and feeds them back to the objective function of the scheduling scheme to obtain the optimal scheduling scheme.

5. The port general cargo handling digital resource scheduling and full-process traceability system according to claim 4, characterized in that, The risk scheduling module calculates the comprehensive risk index of the goods and determines the risk level. Based on the risk level, it generates a multi-level scheduling strategy. The specific process is as follows: The material of the cargo is determined based on its spectral characteristics, vibration sensor data, and magnetic field sensor data. Obtain historical accident data for the corresponding material, and define the weights of cargo stacking risk, thermal safety risk, and equipment operation risk based on the proportion of risk types within the historical accident data; The risks of cargo stacking, thermal safety, and equipment operation are normalized and summed with corresponding risk weights to obtain a comprehensive risk index. Then, the risk level is determined, and a multi-level scheduling strategy is generated based on the risk level.

6. The port general cargo handling digital resource scheduling and full-process traceability system according to claim 5, characterized in that, The specific rules for determining the risk level are as follows: Risk level thresholds are set as M and N, where M > N > 0; If the comprehensive risk index is ≤ N, it is considered low risk, and the standard scheduling procedure is adopted. If N < comprehensive risk index ≤ M, it is considered medium risk and advanced scheduling process is adopted; If the comprehensive risk index is greater than M, it is considered high risk and an emergency dispatch process is initiated.

7. The port general cargo handling digital resource scheduling and full-process traceability system according to claim 1, characterized in that, The traceability verification unit includes a multi-level fingerprint generation module, a data acquisition module, and a verification data module, wherein: The multi-level fingerprint generation module generates a first-level fingerprint based on the spectral characteristics of the cargo, calculates the statistics of the cargo's vibration energy entropy, thermal gradient variance, and acoustic resonance skewness, and combines these three statistics to form a second-level fingerprint; and constructs a Markov chain by combining the time series of the three statistics as a third-level fingerprint. The data acquisition module: sets key nodes in the cargo scheduling process, and acquires the multi-level fingerprint of the cargo again after the cargo passes through the key nodes; The verification data module receives multi-level fingerprints corresponding to the multi-level fingerprint generation module and the data acquisition module, introduces quantum random numbers as perturbation factors, performs similarity calculations, and generates a verification report.

8. The port general cargo handling digital resource scheduling and full-process traceability system according to claim 7, characterized in that, The multi-level fingerprint generation module generates a first-level fingerprint based on the spectral characteristics of the cargo, calculates the statistics of the cargo's vibration energy entropy, thermal gradient variance, and acoustic resonance skewness, and combines these three statistics to form a second-level fingerprint; then, it constructs a Markov chain by combining the time series of the three statistics as the third-level fingerprint. The specific process is as follows: Primary fingerprint: Multiple key band reflectance vectors are collected from the spectral characteristics of the goods and concatenated to obtain a string. The string is then processed using a hash algorithm to generate a multi-digit hexadecimal string, which is then truncated to serve as the primary fingerprint. Secondary fingerprint: The Shannon entropy of energy distribution, the variance of cargo temperature data, and the skewness of acoustic resonance amplitude are calculated. These three statistics are combined into a three-dimensional vector, which is regarded as the secondary fingerprint. Level 3 fingerprint: Discretize the time series data of three statistics, divide the continuous values ​​into five levels of states, count the transition frequencies between these states, construct a transition frequency matrix and convert it into a probability matrix, expand the three probability matrices row by row and concatenate them to obtain the level 3 fingerprint.

9. The port general cargo handling digital resource scheduling and full-process traceability system according to claim 7, characterized in that, The verification data module receives multi-level fingerprints from the multi-level fingerprint generation module and the data acquisition module, introduces quantum random numbers as perturbation factors, performs similarity calculations, and generates a verification report. The specific process is as follows: First, the multi-level fingerprints output by the multi-level fingerprint generation module are used as historical multi-level fingerprints, and the multi-level fingerprints output by the data acquisition module are used as real-time multi-level fingerprints. The Hamming distance of the first-level fingerprint and the cosine similarity of the second-level fingerprint are verified sequentially. Then, the probability matrix similarity between the historical multi-level fingerprint and the real-time multi-level fingerprint is calculated to obtain the similarity value. Based on the similarity value, the degree of anomaly is determined and a verification report is generated.

10. The port general cargo handling digital resource scheduling and full-process traceability system according to claim 9, characterized in that, The similarity of the probability matrices between historical multi-level fingerprints and real-time multi-level fingerprints is calculated using the following formula: ； In the formula: F is the square root of the sum of the squares of all elements of the matrix. This is the transpose of the real-time matrix.

Images