Intelligent monitoring method and system for logistics park

Abstract

The invention relates to the technical field of intelligent monitoring, and provides an intelligent monitoring method and system for a logistics park. The method comprises the steps of obtaining a logistics park plane graph, performing key area identification and arranging an intelligent sensing network; performing real-time monitoring based on the fixed edge node to obtain first monitoring data; obtaining a regional task, evaluating an activity, and calculating a task complexity factor; adjusting the dynamic node based on the complexity factor, and obtaining regional activity data; inputting the activity data into a deep analysis model to obtain a logistics intelligent perception set; and inputting the perception set and the complexity factor into a risk assessment function, and carrying out multi-level early warning matching and feedback. The technical problems of unreasonable resource allocation and untimely risk early warning caused by difficult accurate monitoring of dynamic activity frequency change in a logistics park are solved, and the effects of adjusting dynamic edge nodes in real time through an intelligent sensing network and performing efficient monitoring and accurate early warning in combination with a multi-dimensional task complexity factor are realized.

Description

Technical Field

[0001] The present application relates to the field of intelligent monitoring technology, and specifically to an intelligent monitoring method and system for a logistics park. Background Art

[0002] In the management of modern logistics parks, with the increase in the frequency of logistics activities and the expansion of the park scale, traditional monitoring methods can no longer meet the increasingly complex operational needs of the park. The data of cargo loading and unloading, vehicle entry and exit, environmental monitoring and other aspects in the logistics park change frequently. The monitoring system needs to process a large amount of real-time data and respond quickly to sudden abnormal situations. However, the existing monitoring system usually relies on fixed nodes for monitoring, which cannot flexibly respond to fluctuations in logistics activities, and is prone to waste of resources or monitoring blind spots. In addition, the lack of accurate perception and intelligent analysis of key mission areas leads to low monitoring efficiency and untimely risk warnings, which brings safety hazards and management difficulties to park operations. Summary of the invention

[0003] This application provides a smart monitoring method and system for a logistics park, aiming to solve the technical problems of unreasonable resource allocation and untimely risk warning caused by the difficulty in accurately monitoring the frequency changes of dynamic activities in the logistics park.

[0004] In view of the above problems, the present application provides a smart monitoring method and system for a logistics park.

[0005] The first aspect disclosed in the present application provides a smart monitoring method for a logistics park, the method comprising: obtaining a plan design drawing of the logistics park, identifying key areas based on the plan design drawing, and arranging an intelligent perception network according to the key area identification results; performing real-time monitoring of the key areas based on fixed edge nodes of the intelligent perception network, and obtaining a plurality of first real-time monitoring data; obtaining regional tasks of the key areas, performing regional activity evaluation based on the regional tasks and the first real-time monitoring data, obtaining multi-dimensional indicators of regional tasks, and calculating regional task complexity factors based on the multi-dimensional indicators of regional tasks; dynamically adjusting the edge nodes of the intelligent perception network based on the regional task complexity factors, and obtaining a plurality of regional activity data according to the adjustment results; inputting the plurality of regional activity data into a dynamic perception deep analysis model for multi-layer perception analysis, and obtaining a global logistics intelligent perception set; inputting the vehicle flow index, cargo loading and unloading index, environmental risk index and the regional task complexity factor of the global logistics intelligent perception set into a regional risk assessment function, performing multi-level early warning mechanism matching according to the function calculation results, and performing early warning feedback based on the matching results.

[0006] Another aspect disclosed in the present application provides a smart monitoring system for a logistics park, the system comprising: a key area identification module, the key area identification module is used to obtain a plan design drawing of the logistics park, identify key areas based on the plan design drawing, and arrange an intelligent perception network according to the key area identification results; a real-time monitoring module, the real-time monitoring module is used to perform real-time monitoring of the key areas based on fixed edge nodes of the intelligent perception network, and obtain multiple first real-time monitoring data; a regional activity evaluation module, the regional activity evaluation module is used to obtain regional tasks of the key areas, perform regional activity evaluation based on the regional tasks and the first real-time monitoring data, obtain multi-dimensional indicators of regional tasks, and calculate regional tasks based on the multi-dimensional indicators of regional tasks. task complexity factor; a dynamic edge node adjustment module, the dynamic edge node adjustment module is used to adjust the dynamic edge nodes of the intelligent perception network based on the regional task complexity factor, and obtain multiple regional activity data according to the adjustment result; a multi-layer perception analysis module, the multi-layer perception analysis module is used to input the multiple regional activity data into the dynamic perception deep analysis model for multi-layer perception analysis, and obtain the global logistics intelligent perception set; a multi-level early warning mechanism matching module, the multi-level early warning mechanism matching module is used to input the vehicle flow index, cargo loading and unloading index, environmental risk index and the regional task complexity factor of the global logistics intelligent perception set into the regional risk assessment function, perform multi-level early warning mechanism matching according to the function calculation result, and perform early warning feedback based on the matching result.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] The above-mentioned intelligent monitoring method for a logistics park obtains a plan view of the logistics park, identifies key areas in the park by analyzing the plan view, and arranges an intelligent perception network based on the identification results. Subsequently, these key areas are monitored in real time using fixed edge nodes, the first batch of real-time monitoring data is collected, and then the task information of these key areas is obtained, and regional activity evaluation is performed in combination with the real-time monitoring data to generate multi-dimensional task indicators. Based on these indicators, the complexity factor of each regional task is calculated. According to this complexity factor, the layout of edge nodes in the intelligent perception network is dynamically adjusted to ensure reasonable allocation of resources, and regional activity data is obtained through the adjusted layout. After that, these regional activity data are input into the dynamic perception deep analysis model for multi-level analysis to generate an intelligent perception set of global logistics, including vehicle flow index, cargo loading and unloading index and environmental risk index. Finally, these indexes are input into the regional risk assessment model together with the regional task complexity factor, and a suitable multi-level early warning mechanism is matched based on the calculation results, and early warning feedback is generated according to the matching results to ensure the safe and efficient operation of the logistics park.

[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0011] Figure 1 The figure is a flow chart of a smart monitoring method for a logistics park in one embodiment.

[0012] Figure 2 The present invention is an architecture diagram of a smart monitoring system for a logistics park in one embodiment.

[0013] Explanation of the accompanying drawings: key area identification module 1, real-time monitoring module 2, regional activity assessment module 3, dynamic edge node adjustment module 4, multi-layer perception analysis module 5, multi-level early warning mechanism matching module 6. DETAILED DESCRIPTION

[0014] The embodiments of the present application provide a smart monitoring method and system for a logistics park to solve the technical problems of unreasonable resource allocation and untimely risk warning caused by the difficulty in accurately monitoring the frequency changes of dynamic activities in the logistics park.

[0015] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0016] It should be noted that the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or inherent to these processes, methods, products or devices.

[0017] Embodiment 1, as Figure 1 As shown, the present application provides a smart monitoring method for a logistics park, the method comprising:

[0018] Obtain a plan design drawing of the logistics park, identify key areas based on the plan design drawing, and perform intelligent perception network layout according to the key area identification results.

[0019] In an embodiment of the present application, the system terminal first obtains a complete floor plan of the logistics park, which contains a detailed layout of each functional area in the park, such as cargo loading and unloading areas, storage areas, vehicle access channels, office areas and other auxiliary facilities. Subsequently, by analyzing this floor plan, the functions of different areas in the park are classified and key areas are identified. These areas generally include cargo loading and unloading areas, vehicle import and export areas, storage areas, etc. After that, fixed edge nodes are arranged in each key area of ​​the logistics park. These nodes will cover the basic monitoring needs of the entire park, including cargo loading and unloading, vehicle scheduling and environmental monitoring. The system terminal determines the location of each fixed edge node in combination with the spatial layout and regional monitoring needs in the park, ensures seamless connection of the monitoring coverage, and minimizes monitoring blind spots. According to the functions of different areas, each fixed edge node is configured with a suitable sensor type, such as a camera, temperature and humidity sensor, infrared sensor, etc., to meet the monitoring needs of different areas. In the layout of the intelligent perception network, not only fixed edge nodes need to be arranged, but also dynamic edge nodes. The system terminal analyzes the historical logistics activity data and historical environmental change data of the logistics park, and configures the upper limit of the dynamic edge nodes of the entire logistics park in combination with redundant processing. Then, the dynamic edge nodes are arranged according to the configured upper limit of the dynamic edge nodes. Each dynamic edge node contains all the sensor types required by the logistics park to facilitate the free regulation of the dynamic edge nodes. After completing the arrangement of fixed edge nodes and dynamic edge nodes, the system terminal reasonably arranges the intelligent perception network to ensure real-time monitoring and efficient management of the park.

[0020] Furthermore, the present application provides a dynamic edge node arrangement for an intelligent sensing network, including:

[0021] Acquire historical logistics activity data and historical environmental change data of the logistics park; input the historical logistics activity data and the historical environmental change data into a demand forecasting model to perform minimum coverage demand forecasting and obtain a minimum coverage value of the logistics park; perform redundant processing on the minimum coverage value of the logistics park to obtain a dynamic edge node upper limit; and perform dynamic edge node layout based on the dynamic edge node upper limit.

[0022] Preferably, the system terminal first obtains the logistics activity data and environmental change data of the logistics park in the past period of time from the park's Internet of Things devices. Among them, the logistics activity data includes cargo loading and unloading data, vehicle entry and exit data, cargo circulation data, etc. Environmental change data include temperature, humidity, air quality, etc. in the park. Subsequently, the historical logistics activity data and historical environmental change data obtained are input into the demand forecasting model. The demand forecasting model combines the historical logistics activity data and environmental change data, and uses the learned mapping relationship to predict the monitoring needs of the entire logistics park in the future, and outputs the prediction results as the minimum coverage value of the logistics park. In order to avoid monitoring blind spots or monitoring failures caused by equipment failures, communication delays and other problems during the monitoring process, the system terminal will perform redundant processing based on the minimum coverage value. Specifically, the system terminal sets a certain redundancy coefficient based on the actual needs and importance of the park. This redundancy coefficient is usually 0.2 or 0.3. Then multiply this redundancy coefficient by the minimum coverage value of the logistics park, and add the product to the minimum coverage value of the logistics park to calculate the upper limit of the dynamic edge nodes required for the entire logistics park. This means that the dynamic edge nodes cannot exceed this calculated upper limit in the most dense state to ensure that resources are not wasted excessively. Afterwards, based on the calculated upper limit of the dynamic edge nodes, the system terminal begins to deploy dynamic edge nodes in the logistics park, that is, configure all dynamic edge nodes in the logistics park, and adjust these dynamic edge nodes to a dormant state, so that they can be dynamically scheduled when tasks increase or abnormal situations occur, thereby improving monitoring flexibility.

[0023] For the demand forecasting model, the system terminal obtains the sample logistics activity data, sample environmental change data and sample edge node numbers from the sample library, and divides these samples into data to obtain the training set and the test set. Subsequently, a neural network model is designed as the demand forecasting model according to the characteristics of the input data and the complexity of the problem. The weights and biases of the demand forecasting model are initialized with random numbers. After that, the training set is input into the demand forecasting model, and the number of predicted edge nodes is calculated layer by layer through forward propagation, and the mean square error MSE is used to calculate the loss value of the number of predicted edge nodes and the corresponding number of sample edge nodes. Then, the partial derivative of the loss function with respect to each weight is calculated layer by layer from the output layer to the input layer through back propagation as the gradient, and the weight and bias terms of each layer are updated by the gradient descent method, so that the network is gradually optimized and the value of the loss function is reduced until the maximum number of iterations. After the training is completed, the system terminal uses evaluation indicators such as mean square error and determination coefficient to test the prediction performance of the demand forecasting model on the test set. When the test results meet the expected expectations, the system terminal outputs the current demand forecasting model. Otherwise, the learning rate, number of hidden layers, number of neurons, etc. are adjusted to find the optimal neural network structure.

[0024] Based on the fixed edge nodes of the intelligent sensing network, real-time monitoring of the key area is performed to obtain a plurality of first real-time monitoring data.

[0025] In one embodiment, in the intelligent perception network, the system terminal uses the deployed fixed edge nodes to monitor the key areas in the logistics park and collect the logistics activities and environmental change data in the key areas in real time. Through these fixed edge nodes, the activities in these key areas can be continuously monitored to generate the first real-time monitoring data. These data include the entry and exit of vehicles, the progress of cargo loading and unloading, the fluctuation of environmental temperature and humidity, etc., providing basic support for subsequent dynamic adjustment and risk assessment.

[0026] Acquire the regional tasks of the key area, perform regional activity evaluation based on the regional tasks and the first real-time monitoring data, obtain multidimensional indicators of the regional tasks, and calculate the regional task complexity factor based on the multidimensional indicators of the regional tasks; perform dynamic edge node adjustment of the intelligent perception network based on the regional task complexity factor, and obtain multiple regional activity data according to the adjustment result.

[0027] In one embodiment, the system terminal obtains the tasks currently being performed in each key area from the task scheduling platform. The regional tasks include but are not limited to cargo loading and unloading, vehicle scheduling, environmental monitoring, etc. In order to quantitatively evaluate the regional activities of the regional tasks, the system terminal calculates the regional priority, task urgency and task scale of each key area to form a multi-dimensional index of regional tasks. Subsequently, the multi-dimensional index of the regional tasks of the key areas where the abnormalities occur is sequentially input into the pre-built regional complexity function to generate the regional task complexity factors of these key areas where the abnormalities occur. Afterwards, the number of edge nodes required for the abnormal area is calculated according to each regional task complexity factor, and compared with the fixed edge nodes to obtain the number of dynamic edge nodes required for the area. Then, according to the obtained abnormal key areas and the corresponding number of dynamic edge nodes, the nearest dynamic edge nodes are retrieved to monitor the abnormal key areas, obtain multiple second real-time monitoring data, and combine these second real-time monitoring data with multiple first real-time monitoring data to generate multiple regional activity data, laying the foundation for subsequent early warning feedback.

[0028] Furthermore, the present application provides obtaining the regional task of the key area, performing regional activity evaluation based on the regional task and the first real-time monitoring data, obtaining the regional task multidimensional index, and calculating the regional task complexity factor based on the regional task multidimensional index, including:

[0029] According to a preset time step, the multiple cargo loading and unloading data, multiple vehicle flow data and multiple environmental data of the multiple first real-time monitoring data are intercepted at the current time, and the interception results are standardized to obtain multiple cargo loading and unloading data segments, multiple vehicle flow data segments and multiple environmental data segments; based on the multiple cargo loading and unloading data segments, the multiple vehicle flow data segments and the multiple environmental data segments, multi-dimensional frequency calculation is performed in combination with the preset time step to obtain multiple loading and unloading frequencies, multiple vehicle entry and exit frequencies and multiple environmental frequencies; based on the multiple loading and unloading frequencies, the multiple vehicle entry and exit frequencies and the multiple environmental frequencies, multiple regional priorities are obtained and added to the multi-dimensional indicators of the regional tasks.

[0030] Preferably, the system terminal sets a time step according to the type of task and monitoring requirements, for example, every 5 minutes or 10 minutes is a time period, and regularly extracts data segments from the real-time data stream for analysis. Subsequently, according to the set time step, data information within the corresponding time period is intercepted from multiple cargo loading and unloading data, vehicle flow data and environmental data of multiple first real-time monitoring data, and multiple groups of data segments are generated, each group of data segments corresponds to a key area, and includes cargo loading and unloading data segments, vehicle flow data segments and environmental data segments. Then, these intercepted data segments are standardized to eliminate deviations or outliers in the data, so that different types of data can be analyzed on a unified scale. For example, the data is mapped to a range of 0-1 by the maximum-minimum method. Afterwards, based on the cargo loading and unloading data segments within each time step, the total amount of cargo loaded and unloaded in the cargo loading and unloading data segments is calculated by ratio with the time step, and the cargo loading and unloading frequency of each key area is obtained. The loading and unloading frequency can reflect the amount of loading and unloading tasks completed per unit time. According to the vehicle flow data fragments within each time step, the vehicle entry and exit frequency is obtained by using the ratio calculation of the number of vehicles entering and exiting in the vehicle flow data fragments and the time step. The vehicle entry and exit frequency represents the number of vehicles entering and exiting per unit time. According to the environmental data fragments within each time step, the absolute temperature change and the absolute humidity change in the environmental data fragments are added and calculated, and then the sum result is calculated by ratio calculation with the time step to obtain the environmental frequency. Then, the system terminal sums the calculated loading and unloading frequency, vehicle entry and exit frequency, and environmental frequency of each key area, and uses the sum result as the regional priority of the key area. Finally, the calculated multiple regional priorities are added to the multidimensional indicators of the regional tasks as part of the multidimensional indicators of the tasks, which further optimizes resource allocation and provides a basis for subsequent monitoring and node scheduling.

[0031] Furthermore, the present application provides obtaining the regional task of the key area, performing regional activity evaluation based on the regional task and the first real-time monitoring data, obtaining the regional task multidimensional index, and calculating the regional task complexity factor based on the regional task multidimensional index, including:

[0032] Perform threshold verification on the multiple loading and unloading frequencies, the multiple vehicle entry and exit frequencies, and the multiple environmental frequencies, and obtain the abnormal activity area from the key area according to the verification results; extract the regional tasks of the abnormal activity area, obtain the basic data of the regional tasks, calculate the task urgency and task scale based on the basic data of the regional tasks, and add them to the multidimensional indicators of the regional tasks; input the multidimensional indicators of the regional tasks into the regional complexity function to generate the regional task complexity factor.

[0033] Optionally, the system terminal sets a threshold for the loading and unloading frequency, vehicle entry and exit frequency, and environmental frequency respectively based on historical data and expert advice to identify whether the activity is abnormal. Subsequently, multiple loading and unloading frequencies, multiple vehicle entry and exit frequencies, and multiple environmental frequencies are compared with the corresponding thresholds. If the loading and unloading frequency is higher than the loading and unloading frequency threshold, it means that the loading and unloading activities in the key area are abnormally active and may require more monitoring resources. At this time, the key area is marked as abnormal. If the vehicle entry and exit frequency is higher than the vehicle entry and exit frequency threshold, it means that the vehicle flow in the area is too frequent, which may lead to risks such as traffic congestion or poor scheduling. At this time, the key area is marked as abnormal. If the environmental frequency is higher than the environmental frequency threshold, it means that the environmental conditions fluctuate violently, which may affect the warehousing or logistics process. At this time, the key area is marked as abnormal. Afterwards, the system terminal extracts these key areas with abnormal marks from the key areas to generate abnormal activity areas. Once the abnormal activity areas are determined, the system terminal extracts the tasks currently being executed in these abnormal activity areas and obtains the basic data of the regional tasks in these abnormal activity areas. These basic data of regional tasks include task type, task objectives, task time requirements, etc., such as scheduled completion time, task deadline, quantity of goods, etc. Then, these regional task basic data are input into the task urgency calculation formula and task scale calculation formula respectively, and the task urgency and task scale of these abnormal activity areas are obtained, and added to the regional task multidimensional index to fully reflect the complexity and urgency of the tasks in the abnormal activity areas. Then, the system terminal inputs the regional task multidimensional index into the pre-built regional complexity function to calculate the task complexity factor of these abnormal activity areas. This complexity factor is used for subsequent dynamic edge node adjustment to ensure the reasonable allocation of monitoring resources according to the complexity of the task.

[0034] In combination with the node load capacity, the number of dynamic edge nodes is obtained according to the regional task complexity factor and the fixed edge nodes; a node scheduling instruction is generated based on the abnormal activity area and the number of dynamic edge nodes, and the dynamic edge node movement is executed based on the node scheduling instruction; real-time monitoring of the abnormal activity area is performed based on the dynamic edge nodes, and second real-time monitoring data is obtained, which is combined with the multiple first real-time monitoring data to generate multiple regional activity data.

[0035] Optionally, the system terminal multiplies the regional task complexity factor of the abnormal activity area by the regional coverage requirement (coverage area) corresponding to the area, and then calculates the ratio with the load capacity of a single node to obtain the total number of edge nodes required for the abnormal activity area. Subsequently, the total number of edge nodes is used to perform a difference calculation with the fixed edge nodes of the abnormal activity area to obtain the required number of dynamic edge nodes. After obtaining the number of dynamic edge nodes, the system terminal combines the abnormal activity area with the corresponding number of dynamic edge nodes to generate a node scheduling instruction, which indicates the destination to which the dynamic edge node is to move and the number of times it needs to move. After that, the node scheduling instructions are arranged in descending order according to the regional task complexity factor of the abnormal activity area, that is, the area with the larger regional task complexity factor is prioritized. For the abnormal activity area with the largest regional task complexity factor, the system terminal evaluates the moving cost of each dynamic edge node based on the Dijkstra algorithm, considering the cost of each step. Then, according to the number of dynamic edge nodes in the node scheduling instruction, multiple dynamic edge nodes with the smallest distance are extracted from all dynamic edge nodes, and moved to the destination indicated in the node scheduling instruction. When the dynamic edge node arrives at the corresponding abnormal activity area, it will immediately start to perform monitoring tasks, including data collection and sensor working mode switching. For example, monitoring cargo loading and unloading, vehicle entry and exit frequency, or environmental parameters. These dynamic edge nodes will collect the second real-time monitoring data of the abnormal activity area and add this data to the first real-time monitoring data of the corresponding area to form multiple regional activity data. This data will serve as the basis for the next step of monitoring, scheduling, and decision-making.

[0036] Furthermore, the present application provides a regional complex function, including:

[0037] The regional complex functions are as follows:

[0038]

[0039] Optionally, the regional complexity function is used to calculate the complexity factor C of the regional task task , which combines the urgency, scale and priority of the task, and reflects the complexity of task execution. This complexity factor provides a basis for subsequent dynamic node adjustment. The regional complexity function is as follows: Among them, C task is the regional task complexity factor. For task urgency. standard T is the standard completion time, which is the time expected to complete the task. actual is the actual execution time of the task, that is, the time currently consumed by the task. priorityl It is the regional priority, reflecting the importance of the task. is the task size. storage The larger the storage area, the larger the task scale, and the more monitoring resources and nodes are required. cargo The more cargo there is, the more complex the loading and unloading operations are, and more dynamic edge nodes are needed to cover the monitoring needs. vehicle The number of vehicles is proportional to the transportation complexity of the logistics task. More vehicles require more complex scheduling and monitoring.

[0040] The multiple regional activity data are input into a dynamic perception deep analysis model to perform multi-layer perception analysis to obtain a global logistics intelligent perception set.

[0041] In one embodiment, after obtaining multiple regional activity data, the system terminal inputs these regional activity data into a pre-built dynamic perception deep analysis model. The dynamic perception deep analysis model maps and calculates the vehicle flow data, cargo loading and unloading data and environmental data in the regional activity data through the internal vehicle flow data calculation layer, cargo loading and unloading data calculation layer and environmental data calculation layer to generate a global logistics intelligent perception set. This global logistics intelligent perception set includes a vehicle flow index, a cargo loading and unloading index and an environmental risk index, which are used for subsequent regional risk assessments.

[0042] For the dynamic perception deep analysis model, the system terminal first extracts sample vehicle flow data, sample cargo loading and unloading data, sample environmental data, and sample global logistics intelligent perception set from the sample library, and divides these sample data into training set and test set. Subsequently, the dynamic perception deep analysis model structure is constructed based on a multi-layer neural network, including an input layer, a hidden layer, a vehicle flow data calculation layer, a cargo loading and unloading data calculation layer, an environmental data calculation layer, an output layer, etc. After that, a random number is used to initialize the constructed dynamic perception deep analysis model, and the weights and biases of each layer are set. Then, the dynamic perception deep analysis model is trained based on the training set in the same way as the aforementioned training demand prediction model, until the maximum number of iterations, and the dynamic perception deep analysis model is evaluated using the test set. When the evaluation passes, the current dynamic perception deep analysis model is output.

[0043] The vehicle flow index, cargo loading and unloading index, environmental risk index of the global logistics intelligent perception set and the regional task complexity factor are input into the regional risk assessment function, and a multi-level early warning mechanism is matched according to the function calculation results, and early warning feedback is performed based on the matching results.

[0044] In one embodiment, after obtaining the global logistics intelligent perception set, the system terminal combines the vehicle flow index, cargo loading and unloading index, environmental risk index and regional task complexity factor in the global logistics intelligent perception set into the pre-built regional risk assessment function to calculate the regional risk coefficient of each key area. Subsequently, the calculated regional risk coefficient of each key area is matched with the multi-level early warning mechanism, that is, it is determined whether the regional risk coefficient falls within the corresponding threshold. Among them, the multi-level early warning mechanism includes primary response early warning, enhanced response early warning and deep intervention early warning. The primary response early warning is a reminder notification early warning, indicating that the task complexity of the area is low, the vehicle flow is stable, the cargo loading and unloading volume is moderate, and the environmental risk is controllable. This reminder is intended to maintain continuous monitoring of the regional conditions. The enhanced response early warning is a preventive intervention early warning, indicating that there are certain potential risks in the area, such as increased task complexity, increased vehicle flow, significantly increased cargo loading and unloading volume or deterioration of environmental conditions. At this time, preventive intervention suggestions are sent to managers to prompt them to take intervention measures in advance, such as increasing the monitoring frequency or dispatching more resources. This early warning is intended to prevent potential problems from escalating. Deep intervention warning is an active intervention warning, indicating that the regional situation has deteriorated significantly, such as extremely high task complexity, excessive concentration of vehicles, high pressure on cargo loading and unloading, or a sharp increase in environmental risks. At this time, managers will be notified immediately to take active intervention measures. For example, adding dynamic edge nodes, reallocating logistics resources, or reducing logistics activities in the area. This warning is designed to respond quickly to high-risk situations and prevent more serious consequences.

[0045] Furthermore, the present application provides a regional risk assessment function, including:

[0046] The regional risk assessment function is as follows:

[0047] Preferably, the regional risk assessment function is used to assess the risk level of each area in the logistics park. By combining the indexes of multiple dimensions such as vehicle flow, cargo loading and unloading, and environmental changes, the risk coefficient R of the area is calculated through the function, providing a basis for the subsequent early warning mechanism. The regional risk assessment function is as follows: Among them, R is the regional risk coefficient, which indicates the overall risk level of the region after assessment. The higher the risk coefficient, the greater the potential risk in the region, and more monitoring resources and response measures are needed. vehicle It is the vehicle flow index, which reflects the flow frequency and intensity of vehicles in the area.cargo It is the cargo handling index, which reflects the frequency, speed and volume of cargo handling in the region. env It is an environmental risk index used to quantify the potential impact of the environment on logistics activities. impact C is a time dynamic factor, reflecting the time sensitivity of risk. task is the regional task complexity factor, which reflects the complexity of the regional task.

[0048] In summary, the embodiments of the present application have at least the following technical effects:

[0049] The embodiment of the present application obtains the floor plan of the park and identifies the key areas, and arranges the intelligent perception network based on the identification results, including fixed edge nodes and dynamic edge nodes. Real-time monitoring is then carried out through fixed nodes, and regional activities are evaluated in combination with regional tasks and the first real-time monitoring data, and multi-dimensional indicators are generated and the complexity factor of regional tasks is calculated. Subsequently, the layout and scheduling of dynamic nodes are adjusted according to the complexity factor, and then the regional activity data is obtained and multi-level analysis is performed, and finally a global logistics intelligent perception set is formed. By inputting the vehicle flow index, cargo loading and unloading index, environmental risk index and complexity factor in the global logistics intelligent perception set into the regional risk assessment function, the risk coefficient is calculated and the multi-level early warning mechanism is matched, and feedback is given based on the matching results. The layout of dynamic nodes also predicts the minimum coverage demand based on historical data and sets the node upper limit. Complex tasks are determined by the frequency calculation of multiple real-time data. Through threshold verification and multi-dimensional indicators of regional tasks, nodes are dynamically adjusted to respond to abnormal activity areas. These technical effects jointly solve the technical problems of unreasonable resource allocation and untimely risk warning caused by the difficulty in accurately monitoring the frequency changes of dynamic activities in logistics parks. They achieve the effect of real-time adjustment of dynamic edge nodes through intelligent perception networks, and combine multi-dimensional task complexity factors to carry out efficient monitoring and accurate warning.

[0050] Embodiment 2 is based on the same inventive concept as the intelligent monitoring method of a logistics park in the above embodiment. Figure 2As shown, the present application provides a smart monitoring system for a logistics park, and the system includes: a key area identification module 1: the key area identification module 1 is used to obtain a plan design drawing of the logistics park, identify key areas based on the plan design drawing, and arrange an intelligent perception network according to the key area identification results; a real-time monitoring module 2: the real-time monitoring module 2 is used to perform real-time monitoring of the key areas based on fixed edge nodes of the intelligent perception network, and obtain multiple first real-time monitoring data; a regional activity evaluation module 3: the regional activity evaluation module 3 is used to obtain regional tasks of the key areas, perform regional activity evaluation based on the regional tasks and the first real-time monitoring data, obtain multi-dimensional indicators of regional tasks, and calculate the complexity of regional tasks based on the multi-dimensional indicators of regional tasks. degree factor; dynamic edge node adjustment module 4: the dynamic edge node adjustment module 4 is used to adjust the dynamic edge nodes of the intelligent perception network based on the regional task complexity factor, and obtain multiple regional activity data according to the adjustment result; multi-layer perception analysis module 5: the multi-layer perception analysis module 5 is used to input the multiple regional activity data into the dynamic perception depth analysis model for multi-layer perception analysis to obtain the global logistics intelligent perception set; multi-level early warning mechanism matching module 6: the multi-level early warning mechanism matching module 6 is used to input the vehicle flow index, cargo loading and unloading index, environmental risk index and the regional task complexity factor of the global logistics intelligent perception set into the regional risk assessment function, perform multi-level early warning mechanism matching according to the function calculation result, and perform early warning feedback based on the matching result.

[0051] Furthermore, the key area identification module 1 is also used to perform the following method:

[0052] Acquire historical logistics activity data and historical environmental change data of the logistics park; input the historical logistics activity data and the historical environmental change data into a demand forecasting model to perform minimum coverage demand forecasting and obtain a minimum coverage value of the logistics park; perform redundant processing on the minimum coverage value of the logistics park to obtain a dynamic edge node upper limit; and perform dynamic edge node layout based on the dynamic edge node upper limit.

[0053] Furthermore, the regional activity assessment module 3 is also used to execute the following method:

[0054] According to a preset time step, the multiple cargo loading and unloading data, multiple vehicle flow data and multiple environmental data of the multiple first real-time monitoring data are intercepted at the current time, and the interception results are standardized to obtain multiple cargo loading and unloading data segments, multiple vehicle flow data segments and multiple environmental data segments; based on the multiple cargo loading and unloading data segments, the multiple vehicle flow data segments and the multiple environmental data segments, multi-dimensional frequency calculation is performed in combination with the preset time step to obtain multiple loading and unloading frequencies, multiple vehicle entry and exit frequencies and multiple environmental frequencies; based on the multiple loading and unloading frequencies, the multiple vehicle entry and exit frequencies and the multiple environmental frequencies, multiple regional priorities are obtained and added to the multi-dimensional indicators of the regional tasks.

[0055] Furthermore, the regional activity assessment module 3 is also used to execute the following method:

[0056] Perform threshold verification on the multiple loading and unloading frequencies, the multiple vehicle entry and exit frequencies, and the multiple environmental frequencies, and obtain an abnormal activity area from the key area based on the verification results; extract the regional tasks of the abnormal activity area, obtain basic data of the regional tasks, calculate the task urgency and task scale based on the basic data of the regional tasks, and add them to the multidimensional indicators of the regional tasks; input the multidimensional indicators of the regional tasks into the regional complexity function to generate a regional task complexity factor; combine the node load capacity, and obtain the number of dynamic edge nodes according to the regional task complexity factor and the fixed edge nodes; generate a node scheduling instruction based on the abnormal activity area and the number of dynamic edge nodes, and execute dynamic edge node movement based on the node scheduling instruction; perform real-time monitoring of the abnormal activity area based on the dynamic edge nodes, obtain second real-time monitoring data, and combine it with the multiple first real-time monitoring data to generate multiple regional activity data.

[0057] Furthermore, the regional activity assessment module 3 is also used to execute the following method:

[0058] The regional complex functions are as follows:

[0059] Among them, C task is the regional task complexity factor, For task urgency, T standard is the standard completion time, T actual is the actual execution time of the task, P priorityl is the regional priority, is the task size, A storage is the storage area, Q cargo is the quantity of goods, N vehicle is the number of vehicles.

[0060] Furthermore, the regional activity assessment module 3 is also used to execute the following method:

[0061] The normal activity area is obtained for the regional activity frequency that passes the threshold verification; the edge node composition of the normal activity area is judged, and if there is a dynamic edge node, the dynamic edge node sleep operation is performed.

[0062] Furthermore, the multi-level early warning mechanism matching module 6 is also used to execute the following method:

[0063] The regional risk assessment function is as follows: Among them, R is the regional risk coefficient, I vehicle is the vehicle mobility index, I cargo is the cargo handling index, I env is the environmental risk index, t impact is the time dynamic factor, C task is the regional task complexity factor.

[0064] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order and continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0065] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0066] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A smart monitoring method for a logistics park, characterized in that: The method comprises: Obtaining a plan design of the logistics park, identifying key areas based on the plan design, and performing intelligent perception network layout according to the key area identification results; Based on the fixed edge nodes of the intelligent sensing network, real-time monitoring of the key area is performed to obtain a plurality of first real-time monitoring data; Acquire the regional tasks of the key area, perform regional activity evaluation based on the regional tasks and the first real-time monitoring data, acquire multidimensional indicators of the regional tasks, and calculate the complexity factor of the regional tasks based on the multidimensional indicators of the regional tasks; Dynamically adjusting the edge nodes of the intelligent perception network based on the regional task complexity factor, and obtaining multiple regional activity data according to the adjustment result; Inputting the plurality of regional activity data into a dynamic perception depth analysis model to perform multi-layer perception analysis to obtain a global logistics intelligent perception set; The vehicle flow index, cargo loading and unloading index, environmental risk index of the global logistics intelligent perception set and the regional task complexity factor are input into the regional risk assessment function, and a multi-level early warning mechanism is matched according to the function calculation results, and early warning feedback is performed based on the matching results.

2. The method according to claim 1, characterized in that Performing dynamic edge node deployment of an intelligent sensing network, the method includes: Obtaining historical logistics activity data and historical environmental change data of the logistics park; Inputting the historical logistics activity data and the historical environmental change data into a demand forecasting model to perform minimum coverage demand forecasting, and obtaining a minimum coverage value for the logistics park; Redundant processing is performed on the minimum coverage value of the logistics park to obtain an upper limit of dynamic edge nodes; Dynamic edge nodes are arranged based on the upper limit of dynamic edge nodes.

3. The method according to claim 1, characterized in that Acquiring the regional tasks of the key areas, performing regional activity evaluation based on the regional tasks and the first real-time monitoring data, acquiring multidimensional indicators of the regional tasks, and calculating the regional task complexity factor based on the multidimensional indicators of the regional tasks, the method comprising: According to a preset time step, a plurality of cargo loading and unloading data, a plurality of vehicle flow data, and a plurality of environmental data of the plurality of first real-time monitoring data are intercepted at the current time, and the interception results are standardized to obtain a plurality of cargo loading and unloading data segments, a plurality of vehicle flow data segments, and a plurality of environmental data segments; Based on the multiple cargo loading and unloading data segments, the multiple vehicle flow data segments, and the multiple environmental data segments, a multi-dimensional frequency calculation is performed in combination with a preset time step to obtain multiple loading and unloading frequencies, multiple vehicle entry and exit frequencies, and multiple environmental frequencies; Based on the multiple loading and unloading frequencies, the multiple vehicle entry and exit frequencies, and the multiple environmental frequencies, multiple regional priorities are obtained and added to the regional task multidimensional index.

4. The method according to claim 3, characterized in that Acquiring the regional tasks of the key areas, performing regional activity evaluation based on the regional tasks and the first real-time monitoring data, acquiring multidimensional indicators of the regional tasks, and calculating the regional task complexity factor based on the multidimensional indicators of the regional tasks, the method comprising: Performing threshold verification on the multiple loading and unloading frequencies, the multiple vehicle entry and exit frequencies, and the multiple environmental frequencies, and obtaining an abnormal activity area from a key area according to the verification result; Extracting regional tasks of the abnormal activity area, obtaining basic data of regional tasks, calculating task urgency and task scale based on the basic data of regional tasks, and adding them to multi-dimensional indicators of regional tasks; Inputting the regional task multidimensional index into the regional complexity function to generate a regional task complexity factor; In combination with the node load capacity, the number of dynamic edge nodes is obtained according to the regional task complexity factor and the fixed edge nodes; generating a node scheduling instruction based on the abnormal activity area and the number of dynamic edge nodes, and executing dynamic edge node movement based on the node scheduling instruction; Based on the dynamic edge node, real-time monitoring of the abnormal activity area is performed to obtain second real-time monitoring data, which is combined with the multiple first real-time monitoring data to generate multiple regional activity data.

5. The method according to claim 4, characterized in that The regional complex functions are as follows: Among them, C task is the regional task complexity factor, is the task urgency, T standard is the standard completion time, T actual is the actual execution time of the task, P priorityl is the regional priority, is the task size, A storage is the storage area, Q cargo is the quantity of goods, N vehicle is the number of vehicles.

6. The method according to claim 4, characterized in that Methods include: For the activity frequency of the area that has passed the threshold verification, obtain the normal activity area; The edge node composition of the normal activity area is judged, and if there are dynamic edge nodes, a dynamic edge node sleep operation is performed.

7. The method according to claim 1, characterized in that The regional risk assessment function is as follows: Among them, R is the regional risk coefficient, I vehicle is the vehicle mobility index, I cargo is the cargo handling index, I env is the environmental risk index, t impact is the time dynamic factor, C task is the regional task complexity factor.

8. An intelligent monitoring system for a logistics park, characterized in that: The steps for implementing the intelligent monitoring method of a logistics park as described in any one of claims 1 to 7 include: Key area identification module: obtains the plane design drawing of the logistics park, identifies the key areas based on the plane design drawing, and performs intelligent perception network layout according to the key area identification results; Real-time monitoring module: performs real-time monitoring of the key area based on the fixed edge nodes of the intelligent sensing network, and obtains a plurality of first real-time monitoring data; Regional activity evaluation module: obtaining the regional tasks of the key areas, performing regional activity evaluation based on the regional tasks and the first real-time monitoring data, obtaining multi-dimensional indicators of the regional tasks, and calculating the complexity factor of the regional tasks based on the multi-dimensional indicators of the regional tasks; Dynamic edge node adjustment module: performs dynamic edge node adjustment of the intelligent perception network based on the regional task complexity factor, and obtains multiple regional activity data according to the adjustment result; Multi-layer perception analysis module: input the multiple regional activity data into the dynamic perception depth analysis model to perform multi-layer perception analysis and obtain the global logistics intelligent perception set; Multi-level early warning mechanism matching module: the vehicle flow index, cargo loading and unloading index, environmental risk index and the regional task complexity factor of the global logistics intelligent perception set are input into the regional risk assessment function, a multi-level early warning mechanism is matched according to the function calculation results, and early warning feedback is provided based on the matching results.

Images