Information transmission system and method applied to e-commerce platform

Abstract

The invention relates to the technical field of information transmission, in particular to an information transmission system and method applied to an e-commerce platform. The method comprises the following steps: constructing an information transmission topological structure diagram by obtaining log data of an e-commerce platform and combining information transmission line data; based on the order dynamic impact situation data, the warehousing and distribution load conditions are evaluated, and then logistics load abnormity is detected; through analysis of logistics dynamic load damage evolution data, the increase of information transmission data volume is predicted, and the bandwidth bottleneck of a transmission line is evaluated according to the load increase condition. According to the condition of the bandwidth bottleneck of the transmission line, information transmission abnormity is detected, and optimized information transmission link data is generated through link optimization processing; according to the invention, the transmission link is optimized, so that the information transmission is more efficient and stable.

Description

Technical Field

[0001] The present invention relates to the technical field of information transmission, and in particular, to an information transmission system and method applied to an e-commerce platform. Background Art

[0002] In multi-dimensional services such as high-frequency trading and warehousing and logistics collaboration on the platform, higher requirements are put forward for the timeliness, stability, and scalability of information transmission. Especially in scenarios such as promotional holidays, large-scale user concurrent access, and cross-regional distributed warehousing collaborative operations, the information data carried by the platform presents characteristics of high concurrency, large throughput, and low fault tolerance. Traditional information transmission systems generally have problems such as rigid information topology structure, insufficient dynamic link load perception ability, lag in transmission exception handling, and inaccurate information flow scheduling strategies, making it difficult to meet the current e-commerce platform's comprehensive requirements for "millisecond-level response, dynamic link reconstruction, full-link monitoring, and exception prediction". In the prior art, although some systems introduce edge computing and asynchronous communication mechanisms to improve local transmission efficiency, they often do not form a global unified transmission link state recognition and prediction mechanism. However, traditional information transmission has problems of inaccurate analysis of sudden abnormal increases in data volume and inaccurate analysis of data transmission bottlenecks. Summary of the Invention

[0003] Based on this, it is necessary to provide an information transmission system and method applied to an e-commerce platform to solve at least one of the above technical problems.

[0004] To achieve the above object, an information transmission method applied to an e-commerce platform includes the following steps:

[0005] Step S1: Obtain e-commerce platform log data; collect information transmission line data according to the e-commerce platform log data; construct an information transmission topology structure diagram according to the information transmission line data and the e-commerce platform log data;

[0006] Step S2: Predict order dynamic impact trend data according to the e-commerce platform log data; evaluate the abnormal growth condition of the warehousing and distribution load index based on the order dynamic impact trend data; detect the evolution data of physical flow dynamic load damage according to the abnormal growth condition of the warehousing and distribution load index;

[0007] Step S3: Count the growth of information transmission data volume according to the physical flow dynamic load damage evolution data and the order dynamic impact trend data; estimate the growth of the transmission line load of the information transmission topology structure based on the growth of the information transmission data volume; detect the bandwidth bottleneck condition of the transmission line according to the growth of the transmission line load;

[0008] Step S4: Detect the information transmission anomaly status according to the bandwidth bottleneck status of the transmission line; perform information transmission link optimization processing based on the information transmission anomaly status to obtain information transmission link optimization data.

[0009] The present invention obtains the log data of the e-commerce platform and the information transmission line data, and establishes an information transmission topology structure diagram of the platform, so as to clearly display each node and the connection relationship between them. This structure provides important data support for subsequent transmission load analysis, bottleneck prediction, and anomaly detection. Combining the order dynamic impact trend data and the abnormal growth of the warehousing and distribution load index, it is possible to evaluate the potential warehousing and logistics burdens in advance, timely detect potential distribution load anomalies, and quickly respond and take corresponding measures when the load of the logistics system is too high. In addition, by statistically analyzing the growth of information transmission data volume and predicting the load growth of the transmission line, it is possible to identify in advance the transmission lines that are about to reach saturation, evaluate their bandwidth bottleneck status, avoid data congestion and transmission delay, optimize the transmission efficiency of the platform, and through the optimization of the information transmission link, not only improve the transmission speed, but also reduce the waste of bandwidth, so as to ensure that the e-commerce platform can operate continuously, stably and efficiently in a high-concurrency and complex order environment. Therefore, the present invention is an optimization process for traditional information transmission, which solves the problems of inaccurate analysis of abnormal sudden increase in data volume and inaccurate analysis of data transmission bottlenecks existing in traditional information transmission. It improves the accuracy of analyzing abnormal sudden increase in data volume and the accuracy of analyzing data transmission bottlenecks.

[0010] The present invention also provides an information transmission system applied to an e-commerce platform for executing the information transmission method applied to an e-commerce platform as described above. The information transmission system applied to an e-commerce platform includes:

[0011] A topology structure construction module, configured to obtain the log data of the e-commerce platform; collect the information transmission line data according to the log data of the e-commerce platform; construct an information transmission topology structure diagram according to the information transmission line data and the log data of the e-commerce platform;

[0012] A load damage evolution detection module, configured to predict the order dynamic impact trend data according to the log data of the e-commerce platform; evaluate the abnormal growth of the warehousing and distribution load index based on the order dynamic impact trend data; detect the physical flow dynamic load damage evolution data according to the abnormal growth of the warehousing and distribution load index;

[0013] A transmission line bandwidth bottleneck detection module, configured to statistically analyze the growth of information transmission data volume according to the physical flow dynamic load damage evolution data and the order dynamic impact trend data; estimate the transmission line load growth status of the information transmission topology structure based on the growth of information transmission data volume; detect the bandwidth bottleneck status of the transmission line according to the transmission line load growth status;

[0014] An information transmission link optimization module is used to detect abnormal information transmission conditions according to the bandwidth bottleneck conditions of the transmission line; perform information transmission link optimization processing based on the abnormal information transmission conditions to obtain information transmission link optimization data.

[0015] The information transmission system applied to the e-commerce platform of the present invention can implement any information transmission method applied to the e-commerce platform of the present invention, and is used as a medium for coordinating the operations and signal transmissions between various modules to complete the information transmission method applied to the e-commerce platform. The internal modules of the system cooperate with each other, improving the information transmission efficiency, reducing the bandwidth bottleneck and transmission delay, and ensuring the stability of the platform under high concurrency. Brief Description of the Drawings

[0016] Figure 1 It is a schematic diagram of the step flow of an information transmission method applied to an e-commerce platform;

[0017] Figure 2 For Figure 1 it is a schematic diagram of the detailed implementation step flow of step S3 in

[0018] Figure 3 For Figure 1 it is a schematic diagram of the detailed implementation step flow of step S4 in

[0019] The realization, functional features and advantages of the object of the present invention will be further described in conjunction with the embodiments with reference to the drawings. Detailed Embodiment

[0020] The technical method of the present invention will be clearly and completely described below with reference to the drawings. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0021] In addition, the drawings are only schematic diagrams of the present invention and are not necessarily drawn to scale. The same reference numerals in the drawings represent the same or similar parts, and thus their repeated description will be omitted. Some of the block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0022] It should be understood that although terms such as "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are only used to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly, the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0023] To achieve the above object, please refer to Figures 1 to 3 , an information transmission method applied to an e-commerce platform, comprising the following steps:

[0024] Step S1: Obtain e-commerce platform log data; collect information transmission line data according to the e-commerce platform log data; construct an information transmission topology structure diagram according to the information transmission line data and the e-commerce platform log data;

[0025] In the embodiments of the present invention, information is obtained and sorted from the log data of the e-commerce platform, a topology structure diagram of information transmission is constructed, the log data of the platform is obtained, and the key information (such as order generation time, user information, transaction records, system status, etc.) contained in the log file is used to clean and preprocess these data. The log data usually contains a large amount of timestamps, IP addresses, device information, etc. After parsing, relevant information such as various transaction data, platform activity, and system resource consumption is extracted. This data set is the basis for constructing the topology. After obtaining the e-commerce platform log data, the information transmission line data is collected using this data. The information transmission line data includes transmission paths, device status of nodes, network bandwidth, latency, and historical transmission traffic of each transmission node, etc. After obtaining this information, the network topology within the platform is depicted by analyzing the network traffic conditions of each node and the connectivity between nodes. Specifically, using the shortest path algorithm and network flow model in graph theory, combined with the transmission rate and latency between nodes, an information transmission topology diagram of the platform is constructed. Each node in the diagram represents a data transmission terminal, and each edge represents a network link connecting different nodes. This topology structure diagram contains the traffic, bandwidth, and latency information between each node, and is the basis for information transmission evaluation and optimization in subsequent steps. Through this method, the overall structure and flow direction of information transmission can be visualized in the platform, and the performance and bottlenecks of each transmission line can be clarified. This step uses network topology construction tools and algorithms, such as the shortest path algorithm, network flow analysis method, etc. The constructed topology structure diagram.

[0026] Step S2: Predict the order dynamic impact situation data based on the e-commerce platform log data; evaluate the abnormal growth condition of the warehousing and distribution load index based on the order dynamic impact situation data; detect the damage evolution data of the physical flow dynamic load according to the abnormal growth condition of the warehousing and distribution load index;

[0027] In the embodiment of the present invention, the prediction of the order dynamic impact situation is carried out, and the warehousing and distribution load is evaluated based on the prediction result. By analyzing the order time series information in the e-commerce platform log data, the trend analysis of the order volume is carried out. The specific operation is to extract the timestamp of order generation from the log data, construct a time series data set, and use sequence analysis techniques, such as the autoregressive integrated moving average (ARIMA) model, to predict the trend of the order volume. The goal of this step is to identify the order fluctuations that occur in the short term in the future, especially the sharp increase in the order volume. Based on the predicted order dynamic impact situation data, the warehousing and distribution load index is analyzed. The warehousing and distribution load index is a measure of the workload of the warehousing system and the distribution system within a specific period, reflecting the working intensity of the warehouse and the distribution center. By analyzing the historical order data and combining the processing capacity of the warehouse and the resource situation of the distribution, a load index model is constructed. When the order volume fluctuates greatly, the processing capacity of the warehousing system is challenged, and the abnormal growth of the load index is an important problem faced by the warehousing and distribution system. In the case of detecting the abnormal growth of the warehousing and distribution load index, by further analyzing the concentration of the warehousing sorting tasks and the complexity of the distribution path, the damage evolution data of the physical flow dynamic load is detected. This includes monitoring the goods loading situation, stacking clearance, loading pressure, etc. in the logistics process to identify potential damage risks in advance.

[0028] Step S3: Statistically analyze the growth of the information transmission data volume based on the damage evolution data of the physical flow dynamic load and the statistical information of the order dynamic impact situation data; estimate the growth condition of the transmission line load for the information transmission topology based on the growth of the information transmission data volume; detect the bandwidth bottleneck condition of the transmission line according to the growth condition of the transmission line load;

[0029] In the embodiments of the present invention, it includes the statistics of the information transmission data volume and the estimation of the transmission line load. According to the physical flow dynamic load damage evolution data and the order dynamic impact situation data obtained in step S2, the growth of the information transmission data volume is statistically analyzed. Specifically, the growth of the order volume is combined with the processing requirements of the logistics system, and by analyzing various data flows within the platform, the data volume to be transmitted within a certain period in the future is predicted. Using bandwidth monitoring tools and data flow analysis tools, combined with the real-time transmission data flow direction, the magnitude change of the information transmission is evaluated. Based on the growth of the information transmission data volume, the load growth of the information transmission topology structure is predicted. Here, a load prediction algorithm is mainly adopted. By analyzing the historical traffic data of each node in the information transmission path and using techniques such as linear regression and time series prediction, the future data transmission requirements are estimated. This prediction result can help identify the bandwidth bottleneck problems that occur within the platform. Based on the load growth prediction result, the bandwidth bottleneck situation of the transmission line is analyzed. Through the network bandwidth monitoring tool, the bandwidth usage of the information transmission line is monitored in real time to evaluate whether it can meet the future load requirements. The load growth of the information transmission topology structure is estimated. Based on the bandwidth, load, delay and other information of the transmission line, the system predicts the load growth problems encountered by each transmission line. Through the analysis of the existing data flow, the system estimates the load growth trend of each transmission line and makes corresponding marks in the topology diagram to predict the future bandwidth bottleneck. Through the comprehensive evaluation of the transmission line load, it is timely identified which lines will cause bottlenecks due to overload.

[0030] Step S4: Detect the information transmission abnormal situation according to the bandwidth bottleneck situation of the transmission line; perform information transmission link optimization processing based on the information transmission abnormal situation to obtain information transmission link optimization data.

[0031] In the embodiments of the present invention, according to the transmission line load growth situation obtained in step S3, it is detected which transmission lines have bandwidth bottlenecks. When the system detects a bandwidth bottleneck, it will statistically analyze the response delay parameters of the transmission line and analyze the change trend of its response time. If the response delay of a certain transmission line exceeds the set threshold, it means that the line is overloaded, resulting in problems such as data transmission delay and packet loss. The system then estimates the data loss situation in the information transmission based on these delay parameters, and combines the real-time situation of the data transmission to judge whether there is an abnormal information transmission. Once an abnormal information transmission situation is detected, the system will start the information transmission link optimization process. The system analyzes the bottleneck of the current transmission link and identifies which parts of the transmission link need to be expanded or optimized. The optimization methods include strategies such as increasing the transmission bandwidth, adjusting the data transmission protocol, and re-routing. Through the reasonable planning and adjustment of the transmission line, an optimized link data is obtained.

[0032] Preferably, step S1 includes the following steps:

[0033] Step S11: Obtain the log data of the e-commerce platform;

[0034] In the embodiments of the present invention, the log data is obtained from the system architecture of the e-commerce platform. The specific operations include accessing the background log recording system of the platform, which records the running status, user behavior, order data, and transaction information of the platform in real time. The platform generates and stores log data through multiple modules, including but not limited to: user access logs, transaction operation logs, payment information, order status update logs, etc. The method of obtaining these log data is to directly access the database or the log management system interface and use a log scraping tool for batch extraction. During the process of obtaining the log data, the system needs to set corresponding data filtering rules for different modules to ensure the integrity and accuracy of the log data. The data screening is based on multiple dimensions such as the timestamp of the log, the log type, and the log source. The obtained log data can comprehensively reflect the real-time running situation of the platform. The obtained log data will be converted into a standardized format, such as JSON or CSV, for subsequent processing and analysis. These log data contain key data such as transaction requests, response times, access frequencies, and user information, providing the basic data for subsequent information transmission line analysis and topology construction.

[0035] Step S12: Collect information transmission line data according to the log data of the e-commerce platform;

[0036] In the embodiments of the present invention, based on the log data obtained from the platform, the system further collects information transmission line data. The collection of information transmission line data is achieved by analyzing the data flow path within the platform and the involved data transmission protocols. The specific operations include extracting each information transmission link between the user side and the server side through the network request and response data in the log data. For each transmission line, the system records key performance indicators such as its transmission path, bandwidth, latency, and transmission duration. This process is achieved by configuring a data scraping tool in the network traffic monitoring module of the system to track the internal data flow of the platform in real time. Whenever a user requests to access the platform or the order status is updated, the corresponding network transmission path and line information will be captured and recorded in the log. For different transmission protocols used in the platform (such as HTTP / HTTPS, FTP, TCP / IP, etc.), the system extracts different transmission line data according to the different characteristics of the protocols and stores them in the database. Through these data, the system can clearly understand the load situation of each data transmission line, ensuring that the actual situation of the data flow can be accurately reflected when constructing the information transmission topology structure later.

[0037] Step S13: Collect information transmission node data according to the log data of the e-commerce platform;

[0038] In the embodiments of the present invention, based on the e-commerce platform log data, the system also needs to collect information transmission node data. The information transmission node data represents all the transit nodes involved in the data transmission process, such as the user side, load balancer, Web server, database server, etc. To this end, the system identifies all the transmission nodes by analyzing the request / response data in the log. In the specific implementation process, the system extracts the source address and destination address of each request and response from the log data, and infers the relevant nodes according to the network topology. In actual operation, the system identifies which transit nodes the data stream of the user request passes through by analyzing the transmission path of each request. For example, when a user accesses a website, the data request starts from the user device, passes through the load balancer, goes through the Web server, and then reaches the database server. When analyzing each request process, the system records the information of these nodes, such as node ID, node type (such as user side, Web server, database server, etc.), processing capacity of the node (such as CPU, memory, storage, etc.), response time, etc. The data of each transmission node is identified and associated with the transmission line where it is located, and the system records the dependency relationships between the nodes, such as the relationship between the load balancer and the Web server, and the relationship between the Web server and the database server. All these data will be stored in a standardized format (such as JSON) for subsequent analysis and topology construction.

[0039] Step S14: Construct an information transmission topology structure diagram based on the information transmission node data and the information transmission line data.

[0040] In the embodiment of the present invention, based on the log data collected in step S12 and step S13, the system will construct an information transmission topology diagram by integrating the information transmission node and line data. The system constructs a preliminary topology framework based on the node data and line data, and takes all information transmission nodes as nodes in the diagram, and takes the transmission lines as edges between nodes. Each node represents an entity of data transmission, such as a user end, a server end, a load balancer, a router, etc., and each edge represents a transmission line between nodes. The system analyzes each line and node in detail, and marks the transmission characteristics of each line, such as bandwidth, delay, and throughput. The connection between nodes is determined according to the actual data transmission path. For example, when a request is sent from the user end to the Web server, the system will draw an edge in the topology diagram to connect the user end node and the Web server node, and add bandwidth, delay, and other information to this edge. The system also analyzes the characteristics of each line according to the network protocol, such as HTTP requests need to be transferred through multiple servers, and TCP connections involve different routing paths. The system dynamically updates the path of the data flow, and takes into account the load balancing mechanism and the existing data flow adjustment. If the transmission path within the platform changes, such as adding or deleting a node or transmission line, the system can automatically update the topology diagram to ensure that the information in the diagram reflects the actual operation status of the platform in real time. The generated topology diagram can not only help platform administrators understand the information transmission situation within the platform, but also provide support for subsequent data transmission optimization, bandwidth management, troubleshooting and other operations. Through the topology diagram, the system can identify data flow bottlenecks, single point failures and other problems, and propose corresponding optimization measures.

[0041] Preferably, step S14 comprises the following steps:

[0042] Step S141: Counting node information interaction intensity according to information transmission node data and information transmission line data;

[0043] In an embodiment of the present invention, based on the information transmission node data and information transmission line data collected in the previous steps, statistics on the node information interaction intensity are implemented. The information transmission node data includes node ID, node role (such as warehousing service node, order scheduling node, payment confirmation node), node module, number of data packets received and sent by the node, timestamp and other information; the information transmission line data includes data packet transmission path, correspondence between the starting node and the target node, number of transmission packets per unit time and transmission direction information. During the implementation process, a fully connected table of connections between nodes is constructed, and the data packet interaction behavior between node pairs is recorded in an adjacency matrix manner. Subsequently, the number of data exchanges between each node and the node directly connected to it is accumulated and counted within a unit time (1 second), and the following formula is used to calculate Among them, Pi,j (t) represents the number of data packets transmitted between node i and node j within t seconds, and N is the number of neighbor nodes with which this node has direct data interaction. Through the above method, the interaction intensity data of each node can be obtained, with the unit of pkt / s (packets per second). The statistical data structure obtained is a key-value pair structure, where the key is the node ID and the value is the corresponding interaction intensity value. For example, if node A exchanges data of 20, 30, and 60 pkt / s with three nodes respectively, the interaction intensity of node A is 110 pkt / s. This data result will be used as one of the determination conditions in the subsequent step S143 for coupling degree evaluation.

[0044] Step S142: Statistically calculate the node information interaction density based on the information transmission node data and the information transmission line data;

[0045] In the embodiment of the present invention, the connection degree parameter of the node is introduced to construct the calculation process of the information interaction density. The information interaction density is defined as the average number of interaction data packets corresponding to the unit connection number within the unit time, and is used to measure the activity degree of the node in the local network. During the implementation process, first, by analyzing the node connectivity structure in the information transmission line data, a node connection table is constructed to record the number of nodes directly connected to each node. Subsequently, combined with the interaction intensity obtained in step S141, the following calculation method is used to obtain the interaction density: Interaction density = Interaction intensity / Number of connected nodes. If a certain node is connected to 4 nodes and the total interaction intensity is 100 pkt / s, then the interaction density is 25 pkt / s. The interaction density data of all nodes are recorded in the form of a list.

[0046] Step S143: Evaluate the transmission node link coverage coupling degree data when the node information interaction density exceeds 97.5 pkt / s and the node information interaction intensity exceeds 100 pkt / s;

[0047] In the embodiment of the present invention, using the node interaction intensity and density data respectively obtained in steps S141 and S142, threshold judgment is performed on all transmission nodes one by one. When the interaction density value of a certain node exceeds 97.5 pkt / s and the interaction intensity value exceeds 100 pkt / s, it enters the link coverage coupling degree calculation process. The technical method used in this step is local subgraph clustering analysis. Taking the node that meets the conditions as the center, all its first-order and second-order connected nodes are extracted, and the corresponding subgraph is constructed. Subsequently, the edges of this subgraph are statistically analyzed to determine whether the same data stream appears on multiple different paths within the unit time. The proportion of the edges where the repeated paths appear is the coupling degree reference value. The higher this value, the stronger the coupling relationship between the node and its link. This evaluation value is expressed as a percentage to obtain the transmission node link coverage coupling degree data.

[0048] Step S144: Identify the real-time flow direction of the transmission data stream based on the data transmission line data;

[0049] In an embodiment of the present invention, based on the transmission line data, the data flow direction of all transmission behaviors of the platform within a specific time period is identified. The sending time, source address, destination address and sequence number of the data packet are extracted from the log data, and the complete path of each logical data flow is identified by recombining the data stream in chronological order. In a scenario with multi-hop relays, the complete path chain of the data stream from the starting node through which relay nodes to the target node is identified in combination with the transmission line data. The actual flow direction of each data stream is organized into a time-series linked list structure, and the data stream direction is stored in the order of data packet transmission. The output format is: {stream number, starting node, target node, path node sequence, start and end timestamps}. This structure is the real-time flow situation of the transmission data stream.

[0050] Step S145: Calculate data transmission edge bandwidth data using the real-time flow situation of the transmission data stream and the data transmission line data;

[0051] In the embodiment of the present invention, each transmission edge passed by each data flow is located based on each data flow path obtained in step S144. For each transmission edge, the total number of data packets passing through the edge per unit time is counted to measure its transmission load. The statistical method is to traverse all data flow path information, count the number of times each edge is traversed, and combine the data packet size field (Byte field) to calculate the total amount of data transmission per unit time. Bandwidth data is recorded in MB / min and named "data transmission edge bandwidth data", where the bandwidth of an edge is defined as the total size of the data flow it carries per unit time to form a structured transmission edge bandwidth data table, including edge number, start node, end node and corresponding bandwidth value.

[0052] Step S146: Analyze the characteristics of the data transmission edge based on the data transmission edge bandwidth data and the real-time flow direction of the transmission data stream;

[0053] In an embodiment of the present invention, the structural features of each transmission edge are analyzed by combining the "data transmission edge bandwidth data" formed in step S145 with the "real-time flow direction of the transmission data stream" in step S144. The analysis process involves the following dimensions: First, whether there is a high-frequency reversal behavior in the transmission direction of the edge, that is, the data flow often goes back and forth on the same edge in a short time; second, whether the edge is at the intersection of high-frequency paths. If the two nodes connected by the edge are both necessary nodes for multiple data flow paths, the edge is considered to be a core transmission edge; third, whether the edge belongs to a bottleneck path, that is, the data segment that carries the largest amount of data in the overall path. After counting these features, three indicators of "round-trip frequency", "path intersection degree" and "edge load ratio" are defined for each edge, respectively, and stored in the data transmission edge feature table to obtain the data transmission edge feature.

[0054] Step S147: Construct an information transmission topology structure diagram based on the data transmission edge features and the coupling degree data of the transmission node link coverage.

[0055] In the embodiment of the present invention, the "coupling degree data of the transmission node link coverage" obtained in step S143 is jointly analyzed with the "data transmission edge feature table" obtained in step S146 to construct an information transmission topology structure diagram. The node connection relationships with a link coverage coupling degree less than 30% are filtered out, and the high-coupling-degree links are retained as the backbone structure. The edges with a path intersection degree higher than the average value are preferentially selected and incorporated into the topological core connection set. Combining the bandwidth data and the path flow information, a weight is assigned to each edge in the topology diagram. The basis for the weight is the proportion of the data flow borne by the edge in all paths. A network diagram is constructed through a graph database graphical processing tool, where the nodes represent information transmission nodes, the edges represent information transmission links, and the weights are presented in the form of line widths to form a complete information transmission topology structure diagram, which is saved as graph structure serialized data.

[0056] Preferably, the prediction of the order dynamic impact situation data in step S2 includes:

[0057] Perform order time series analysis on the e-commerce platform log data to obtain the e-commerce platform order time series data;

[0058] In the embodiment of the present invention, based on the real-time log data generated by the e-commerce platform, all order generation records within 30 consecutive days are obtained. The collected data fields include but are not limited to order numbers, order creation timestamps, order status change timestamps, and order payment status flags. The timestamp field with a time accuracy reaching the second level is used as the main index of the time series, and the creation time points of daily orders are summarized in hourly units to generate an hourly order quantity data series. Subsequently, the ChronoIndex time series data engine is used to perform format normalization processing on the order data, and the processed data is loaded into the data stream middleware module based on Apache Kafka to support high-frequency reading and batch processing distribution. By continuously extracting the fixed-length sequence of order frequencies formed within 24 hours of each day (i.e., the order frequency vector with a length of 24), a continuous time series data group is formed. This time series data uses the "order generation time" field as the horizontal axis and the "total number of orders within a unit hour" as the vertical axis to complete the construction of the order time series data.

[0059] Monitor the change trend of the order volume according to the e-commerce platform order time series data;

[0060] In an embodiment of the present invention, after obtaining the order time series data, a change trend calculation module based on the moving average filtering technology is called. This module uses a sliding window method with a window length of 5 to smooth the order frequency sequence and filter out short-period random fluctuation data. Subsequently, the differential method is used to calculate the order frequency change value within two adjacent time periods. For each time node \(t_i\), calculate the order frequency difference \(\Delta N(t_i)=N(t_i)-N(t_{i - 1})\) between it and the previous moment \(t_{i - 1}\), and import this difference sequence into the trend fluctuation monitoring sub-module. Perform sign judgment and persistence discrimination on the difference sequence. If there are 5 or more consecutive positive difference points, it is determined as an order growth trend; if the number of consecutive negative difference points is greater than or equal to 5, it is recorded as an order decay trend. At the same time, record the start and end times and amplitudes of the growth or decay segments in the trend identifier, output the daily order volume change trend interval segment, and structurally record each segment of the trend in the temporary data table trend_temp_table for subsequent steps to call.

[0061] Construct a business platform order change fluctuation graph based on the order volume change trend;

[0062] In an embodiment of the present invention, combining the order growth and decay interval segment information recorded in the aforementioned trend_temp_table, a two-dimensional coordinate graph is constructed to express the order change fluctuation situation. The horizontal axis of the image uses the time axis with the precision set to the hour level; the vertical axis is set to the order frequency change value. The growth trend interval is marked in red, the decay trend interval is marked in blue, and annotation points are added to the extreme value points of the order frequency. This graph is constructed using the Matplotlib plotting tool, grid auxiliary lines are set to enhance readability, and the daily order fluctuation images are output as PNG format picture files by day, with the file name containing the date index. The order change fluctuation graph not only intuitively shows the order fluctuation trend but also includes the dense order generation time periods and steep fluctuation segments, and the business platform order change fluctuation graph has been generated.

[0063] Calculate the business platform change fluctuation slope parameter according to the business platform order change fluctuation graph;

[0064] In the embodiments of the present invention, based on the data formed by the order change fluctuation graph, the starting frequency and the ending frequency in each trend segment are extracted, combined with the corresponding start and end time points, and the slope formula K = ΔY / ΔX is used for segment-by-segment calculation. Where ΔY represents the order frequency change value, and ΔX represents the length of the corresponding time period (in hours). After the slope of each trend segment is calculated, it is respectively stored in the parameter table slope_table and grouped by day to form a set of multi-dimensional vector data. Further, the maximum slope and the minimum slope values of each day are identified as important reference indicators for judging sudden order fluctuations. To enhance accuracy, an identifier is attached to the slope calculation result of the abnormal segment (such as the slope mutation exceeding three times the previous value) to obtain the business platform change fluctuation slope parameter.

[0065] Identify the sudden sharp increase in business platform orders based on the business platform order change fluctuation graph and the business platform change fluctuation slope parameter;

[0066] In the embodiments of the present invention, the order frequency extreme points in the aforementioned fluctuation graph are jointly analyzed with the slope data in slope_table. For the situation where the order generation frequency exceeds twice the average frequency within two or more consecutive hours and the slope of this segment exceeds three times the daily average value, it is determined as an "order sudden sharp increase" event. This identification operation is performed by associatively comparing two data tables trend_temp_table and slope_table, screening the intervals that meet the sudden increase criteria, extracting the corresponding start and end times, the maximum order frequency, and the maximum frequency growth rate (i.e., the maximum slope), and generating a sudden increase event record table spike_event_table, whose fields include event_id, start_time, end_time, max_slope, and order_peak. If the number of daily sudden increase events is greater than or equal to 2, then the label "high-intensity sudden increase day" is appended.

[0067] Statistical data on the concentrated time periods of order sudden increases based on the business platform order change fluctuation graph and the sudden sharp increase in business platform orders;

[0068] In the embodiment of the present invention, the list of sudden increase events recorded in the spike_event_table is called, and the start and end time periods of all events are summarized by day. If the event interval is less than or equal to 3 hours, they are merged into the same sudden increase concentration segment. Each sudden increase concentration segment records its total duration, maximum order frequency, and maximum frequency growth rate, and is recorded in the order_spike_cluster table. The daily sudden increase concentration time periods are statistically analyzed by time period frequency, and compared with the benchmark order frequency of each time period of the entire platform throughout the day to output the "sudden increase density parameter" value for each hour. The calculation method is the proportion of the sudden increase occurrence frequency in a certain hour to the total sudden increase frequency throughout the day. The data output by this step includes the start time, end time, duration, peak frequency, and occurrence frequency of the sudden increase concentration segment, which are used for subsequent prediction of the dynamic impact situation.

[0069] Predict the data of the order dynamic impact situation based on the data of the order sudden increase concentration time period and the sudden increase situation of the business platform orders.

[0070] In the embodiment of the present invention, the obtained order_spike_cluster table and the emergency event record spike_event_table in step five are subjected to data fusion processing to establish an all-day order impact evolution table order_shock_map. By mapping each sudden increase concentration time period to the all-day timeline and marking its corresponding maximum frequency increase, duration, and occurrence frequency, a multi-dimensional time period impact intensity table is formed. An order impact intensity calculation function is introduced. This function synthesizes three parameters: the duration of the sudden increase segment, the frequency slope, and the maximum frequency, and outputs the order dynamic impact intensity value (unit: orders / h2). The impact intensity value is marked on the timeline to form an impact intensity distribution map in hours, and is output as the order dynamic impact situation data set.

[0071] Preferably, the evaluation of the abnormal growth situation of the warehousing and distribution load index described in step S2 includes:

[0072] Analyze the concentration situation of the warehousing sorting tasks based on the order dynamic impact situation data;

[0073] In the embodiment of the present invention, the obtained order dynamic impact situation data is used as the input data source, and the sorting task distribution extraction module in the structured log analysis platform is called to batch and aggregate the order impact data according to the time stamp. The division granularity is set to a 15-minute time period. The order delivery address, category number, and SKU code fields within each time period are parsed. Based on the above data fields, a sorting function is called to count the task density under each warehousing node, that is, the number of orders that need to be processed per unit time under each warehousing node, to form a warehousing sorting task distribution table. Taking the node number as the row label and the time period number as the column label, a warehousing sorting task concentration situation matrix is generated.

[0074] Estimate the frequency of warehouse area scheduling based on the concentration status of warehouse sorting tasks;

[0075] In an embodiment of the present invention, according to the node records in the warehouse sorting task concentration status matrix where the growth rate of the sorting task volume exceeds 22% in three consecutive time periods, combined with the scheduling response cycle recorded in the historical scheduling table of warehouse scheduling, a frequency comparison function is called to analyze the frequency of warehouse area scheduling. In this analysis, scheduling records with a response cycle lower than 6 minutes are marked, the proportion of these scheduling behaviors in each time period is statistically calculated, and a set of scheduling frequent node identifiers and a data sequence of the frequent scheduling index within the corresponding time period are output.

[0076] Evaluate the instantaneous shortage of warehouse distribution resources based on the frequency of warehouse area scheduling and the concentration status of warehouse sorting tasks;

[0077] In an embodiment of the present invention, an intersection operation is performed on the set of scheduling frequent node identifiers and the above-mentioned set of sorting task concentration nodes, the node numbers that exist in both sets are extracted, and for these nodes, the current inventory data, in-transit replenishment data, and outbound request data are further extracted from the material inventory management system, and a resource gap calculation function is used to calculate the material supply and demand difference per unit time. When the difference is lower than -8 for two consecutive time periods (i.e., instantaneous resource shortage), it is marked that there is an instantaneous shortage of distribution resources at this node. A list of nodes with instantaneous shortage of distribution resources and the corresponding supply and demand difference sequence are output.

[0078] Monitor the congestion status of distribution warehouse sorting based on the instantaneous shortage of warehouse distribution resources;

[0079] In an embodiment of the present invention, among the nodes with marked instantaneous shortage of distribution resources, the AGV (Automated Guided Vehicle) scheduling records and job path tracking records in the internal warehouse WMS system data log are called to analyze the number of intersections and overlaps, waiting duration, and queuing length in each AGV task path. If the average queuing length is greater than 5 vehicles and the path waiting time exceeds 110 seconds, it is determined that there is sorting congestion during this time period, and the congestion index and average waiting time of each node are output as the data of the sorting congestion status.

[0080] Detect a sharp increase in the load of the warehouse distribution center based on the congestion status of distribution warehouse sorting and the instantaneous shortage of warehouse distribution resources;

[0081] In an embodiment of the present invention, the sorting congestion status data and the instantaneous resource shortage data are combined and processed, and the situations where the same node satisfies the above two abnormal states in the same time period are statistically calculated. If this abnormal state lasts for more than two time periods, it is determined that this node enters the state of sharp increase in the load of the warehouse distribution center. Record the task volume, scheduling request frequency, and system response delay data in this sharp increase state to form a record table of the node load sharp increase state.

[0082] Determine the bottleneck critical value state of the warehousing node according to the load surge situation of the warehousing and distribution center and the congestion degree of warehousing, distribution, and sorting.

[0083] In the embodiment of the present invention, based on this table, compare the designed capacity of the node with the current task load, calculate the ratio of the task volume of each load surge node to the upper limit of the warehousing service capacity. If this ratio exceeds 0.85 and the congestion index is greater than 1.2, determine that this node enters the bottleneck critical value state. Construct a bottleneck node identification list based on this determination result, and record the corresponding critical task value, congestion index, and supply-demand difference.

[0084] Evaluate the abnormal growth condition of the warehousing and distribution load index according to the bottleneck critical value state of the warehousing node and the load surge degree of the warehousing and distribution center.

[0085] In the embodiment of the present invention, make the nodes in the above-mentioned bottleneck node identification list correspond one by one with the nodes in the surge state record table, call the exponential change trend function, perform exponential fitting processing on the task volume, scheduling frequency, resource gap, and congestion index of each node, calculate its change rate within the past 30 minutes, and output the load change trend of each node in the format of a five-dimensional vector. If the growth rates of three items in the five-dimensional vector of a certain node exceed 20%, mark it as a node with abnormal growth of the load index, and output the list of nodes with abnormal growth of the warehousing and distribution load index and their corresponding growth index records, which are used as the input data for the logistics load evolution detection in the subsequent steps.

[0086] Preferably, the detection of the damage evolution data of the physical flow dynamic load in step S2 includes:

[0087] Detect the overloading of goods according to the abnormal growth condition of the warehousing and distribution load index.

[0088] In the embodiment of the present invention, use the "abnormal growth condition of the warehousing and distribution load index" obtained from the previous stage evaluation as the input data source. This index is composed of the concentration degree of the warehousing sorting task, the bottleneck critical value state of the warehousing node, the shortage degree of the warehousing and distribution resources, and the load surge frequency. By screening the partition load coefficient (defined as the required scheduling frequency of goods per unit volume per unit time) in this index, select the data section with a value exceeding 20% of the upper limit of the partition average load threshold, and mark it as the over-dense area of the loading stress. On this basis, retrieve the corresponding logistics scheduling record and the cargo volume information of this area, and determine that the part where the cargo volume ratio (the ratio of the total volume of goods to the volume of the warehouse space) in the unit warehouse exceeds the set safe stacking ratio of 1.2 is the "overloading of goods situation", and form a list of overloading time periods and the corresponding identification numbers of loading units.

[0089] Measure the stacking clearance parameter of the goods based on the overloading of goods situation.

[0090] In the embodiment of the present invention, on the basis of obtaining the overloaded unit identifier, the three-dimensional scanned image data in the warehousing outbound monitoring record is processed accordingly. An industrial structured light scanner (accuracy 0.3 mm) is used to obtain the external contour data of each cargo unit, and the minimum gap between adjacent cargo boundaries within the stacked area is extracted. By calculating the average value of all boundary spacings, the average stacking gap value of a single storage position is obtained. The "stacking gap degree parameter" is introduced as a key indicator to characterize the tightness of cargo stacking. This parameter is defined as: the average gap between each cargo divided by the minimum side length of a single piece of cargo. The lower the value, the denser the stacking. When the parameter value is less than 0.05, a statistical table containing the loading unit number and the corresponding stacking gap degree parameter value is output as the state of no effective buffer space.

[0091] Detect the loading extrusion situation of the distribution cargo based on the cargo stacking gap degree parameter and the overloading situation of the cargo loading.

[0092] In the embodiment of the present invention, the loading unit number, the overloading state and the stacking gap degree parameter value obtained in the previous two steps are used as the combined input. According to the superimposed characteristics of the loading density and the gap degree, the judgment logic is set as follows: if the loading density exceeds the safety upper limit of 1.2 and the stacking gap degree parameter is less than 0.05, it is determined as the "loading extrusion situation of the distribution cargo". Here, an auxiliary verification mechanism is introduced, that is, the stress data of the sensors in this area is rechecked. The cumulative stress value of each vertical load path is detected by the pressure sensors embedded at the bottom of the pallet. If the detected value is more than 30% higher than the average value over the years, it is confirmed that physical extrusion actually occurs. After this detection, a "loading extrusion unit identification table" is generated, which includes the corresponding loading unit number, stacking density, gap degree, measured pressure value and extrusion confirmation mark.

[0093] Measure the growth degree of the loading pressure of the distribution cargo according to the loading extrusion situation of the distribution cargo.

[0094] In the embodiment of the present invention, taking the loading unit with confirmed loading extrusion as the object, the change of the loading pressure during the entire distribution path is continuously monitored. Using the dynamic weighing module equipped on the distribution vehicle, the pressure value of the pallet or container is automatically recorded when leaving each station. At the same time, combined with the real-time geographic positioning system, the distribution mileage is recorded for each pressure measurement point. According to the pressure-mileage sequence diagram, the pressure growth rate is calculated, with the unit of kPa / km. By calculating the ratio of the linear fitting slope to the pressure growth rate of the historical normal distribution path, the "pressure growth degree parameter" is formed. When the ratio exceeds 2.0, it is defined as the "distribution path with rapid pressure increase", and a complete table including the numbers of each extrusion unit, the corresponding pressure sequence, the growth degree slope and the abnormal mark is output.

[0095] Measure the lack of buffer space of the distribution cargo based on the growth degree of the loading pressure of the distribution cargo and the cargo stacking gap degree parameter.

[0096] In the embodiments of the present invention, the degree of lack of buffer space needs to comprehensively consider the current stacking density and the pressure growth value per unit time. The system calls the stored database of cargo packaging structures and determines whether the current gap is sufficient to resist the deformation trend caused by the current pressure rate according to the resilience modulus and the deformation threshold of each type of cargo packaging material. If there is a mismatch between the pressure growth rate and the energy absorption rate of the packaging material (the former is more than 20 percentage points higher than the latter), and the gap degree is lower than 80% of the historical average value, the system marks the stacking structure as lacking buffer space and calculates the lack level (for example, divided into three levels: slight, moderate, and severe according to percentages) for impact force propagation analysis.

[0097] Detect the impact force propagation of the distribution cargo by using the degree of pressure growth of the distribution cargo loading to detect the lack of buffer space of the distribution cargo;

[0098] In the embodiments of the present invention, the impact force propagation situation needs to be comprehensively determined by combining the pressure growth curve and the space elastic conduction path. After generating the lack identification of the buffer space, the system collects the microseismic intensity data from the microseismic sensor array arranged in the bottom-up conduction path according to the structural arrangement between the stacking layers, and constructs a shock wave front propagation map by combining the time delay matrix. The impact propagation rate is measured by the time difference of the responses of the sensors on each layer, and the impact force transfer intensity is calculated by combining the interlayer pressure conduction ratio. If the propagation rate is faster than the set threshold (such as more than 2 meters per second), and the impact amplitude in the middle and lower layers is close to 90% of the packaging collapse limit, an impact force propagation strong warning state is output, constituting an impact force propagation parameter set.

[0099] Estimate the impact deformation trend of the distribution cargo according to the impact force propagation of the cargo;

[0100] In the embodiments of the present invention, after obtaining the impact force propagation parameter set, the scheduling system respectively establishes the reference value of the deformation limit according to the dynamic response characteristics of different packaging materials. Taking the shock wave intensity and duration recorded by the sensor as the input, comparing with the maximum energy absorption limit value of the cargo packaging, the deformation risk coefficient is calculated. This process does not require the use of a prediction model, but is obtained by comparing the actually measured time-impact relationship curve with the packaging deformation atlas. If the impact value exceeds 95% of the packaging dynamic compression critical line, it is classified into the high-risk area to generate a set of impact deformation trend values, and the impact coefficients are marked respectively according to the tray position distribution, reflecting the gradient change of the impact deformation trend.

[0101] Detect the evolution data of the dynamic load damage of the logistics based on the impact deformation trend of the distribution cargo and the impact force propagation of the cargo.

[0102] In the embodiments of the present invention, a complete dynamic damage evolution map is constructed by using the obtained impact deformation trend and impact force propagation state. The scheduling system aligns the state at the time of goods warehousing, the loading start time, the pressure rising stage, the impact propagation stage, and the predicted deformation stage on the time axis in a time series combination manner to form a damage evolution chain. This chain identifies the whole process from the start of loading to the generation of high-intensity impact, including the key physical parameters, the duration, and the influence range of each stage, and outputs the dynamic load damage evolution data of the logistics, the content of which includes the damage start time, the propagation path, the risk level, and the list of pallet numbers involved, and is stored in the data platform as an important data entry for logistics task warning and receipt record.

[0103] Preferably, step S3 includes the following steps:

[0104] Step S31: According to the dynamic load damage evolution data of the logistics and the statistical information of the dynamic impact situation data of the order, count the growth of the data transmission volume;

[0105] In the embodiments of the present invention, based on the set of dynamic load damage evolution data of the logistics and the set of dynamic impact situation data of the order obtained in the previous steps, a unified time series data alignment platform is constructed. This platform performs data standardization processing on parameters such as the number of damage warnings recorded at the logistics damage evolution nodes, the change rate of the impact deformation trend, and the impact propagation rate, and at the same time constructs an event-driven information transmission call mapping table in combination with information such as the peak number, duration, and fluctuation slope of the sudden increase in the order volume in the dynamic impact situation data of the order. This mapping table merges the node events corresponding to the timestamps, accumulates the total number of events triggered in each time period, and then calculates the number of information triggers per unit time. By calling the node information scheduling record log in the data link layer, read the number of upstream and downstream data packets of each data node in the corresponding time window, compare with the basic data traffic benchmark value without an event background, calculate the data flow growth multiple after the event is triggered, and use a 5-minute sliding window to calculate the change rate of the total data transmission volume in the continuous window in each event occurrence interval, and obtain the corresponding information transmission data volume growth multiple parameter, which is used for the subsequent growth assessment of the topological structure line load.

[0106] Step S32: Based on the growth of the information transmission data volume, estimate the growth of the transmission line load of the information transmission topological structure;

[0107] In the embodiments of the present invention, by mapping the growth magnification parameter of each node onto the topological structure along the time axis, a two-dimensional matrix combining time and space is constructed. This matrix reflects the information transmission growth of each node in different time periods. Then, in combination with the communication frequency distribution of each node in the platform log data and the weight parameter of historical task distribution, the load growth rate of each transmission line per unit time is calculated using matrix weighted operations. Through this operation, the load growth rates of each line in different time periods are obtained, and these rate values reveal the load change trend of each line in the next period of time. To better understand these change trends, the load growth rates of all lines are then normalized. The normalized data forms a line load growth evaluation map, which shows the load growth of each line. In the map, each line is assigned a slope value representing the short-term information data processing increase trend of the line. This value reflects the load growth expectation of the line in the future. To further analyze the load change of each line, in combination with the change amplitude of the slope in three consecutive time windows, the load growth trend curve of the current line is calculated.

[0108] Step S33: Estimate the data transmission cache overflow situation according to the load growth condition of the transmission line;

[0109] In the embodiments of the present invention, based on the line load growth trend curve generated in step S32, in combination with the cache capacity parameter of each node and the response delay of the data buffer, further analysis is carried out by constructing a distributed cache response table. The specific operation is to calculate the difference between the data write rate and the cache release rate at each level of the node. The integral result of this difference forms a cache backlog function, indicating the change of cache backlog. The cache backlog function measures the load of the cache by accumulating the difference between the inbound data flow rate and the cache processing output rate. Among them, the inbound data flow rate is converted from the line load growth trend curve, reflecting the amount of data transmitted to the node per unit time. The cache processing output rate is limited by the hardware parameters and bandwidth protocol of the node, determining the rate at which data is released from the cache. By calculating the difference between these two rates and integrating, the cache backlog function is obtained. The cache backlog function of each node changes over time, forming a dynamic numerical curve. By performing a first derivative process on the cache backlog function, the growth rate of the cache overflow risk is obtained. This growth rate reflects the acceleration of cache backlog. If the growth rate of the cache backlog function of a certain node exceeds the set threshold within a certain time period, then the node is determined to have a potential cache overflow risk during this time period.

[0110] Step S34: Detect the bandwidth bottleneck situation of the transmission line according to the data transmission cache overflow situation and the load growth situation of the transmission line.

[0111] In the embodiment of the present invention, in the cache overflow risk level table formed in step S33, the cache usage of each transmission node in the information transmission network of the e-commerce platform is clearly recorded. This table is composed of a joint index of the node unique number and the timestamp. Each record item includes the total cache capacity of the node, the current cache occupied space, the cache occupancy growth rate, and the risk level label calculated based on the historical load behavior. The risk level label is jointly set according to the cache occupancy ratio and its growth rate, and is divided into multiple level intervals to indicate the cache pressure of the node in the current time period. The line load growth trend curve generated in step S32 reflects the traffic growth trend of each information transmission line within a certain time range. This trend curve calculates the change sequence of the average transmission rate over time by statistically analyzing time series data such as the number of data packets transmitted on each line and the transmission delay, so as to depict the whole process of the line load migrating from low density to high density. This process is usually supported by the traffic scheduling log, the packet header statistical information, and the node flow control feedback, and has high real-time performance and accuracy. In the joint analysis stage, the above two types of data structures are logically bound to construct a coupling matrix between the node load and the bandwidth carrying capacity. This matrix uses the node number as the row index and the transmission line number as the column index. Each matrix element is used to describe the current relative bandwidth saturation state of a certain transmission node on a certain line. Specifically, the value of the matrix element is jointly determined by three parameters, namely the average data flow rate of the current node on this line (calculated from the historical transmission records), the maximum transmission rate supported by this line (determined by the network configuration parameters), and the current cache occupancy ratio of the node (the real-time value from the cache risk level table). In the construction process of this matrix, the following logical process is adopted for the calculation of each unit: read the data traffic log of the node transmitted on the corresponding line, and calculate its average flow rate per unit time; read the maximum bandwidth value preset in the physical link of the line; finally, extract the cache occupancy percentage of the node in the current time period, and multiply the above three indicators after normalization to obtain the saturation estimation value of the carrying capacity of this node on this line. The closer the saturation estimation value is to 1, the greater the transmission pressure of this node on this line. After completing the construction of the node load bandwidth coupling matrix, perform time series analysis on the values in the matrix. Within the set sliding time window, if a certain matrix element exceeds the preset critical saturation threshold (such as 85%) for three consecutive time periods and the value shows a continuous upward trend, then this line is determined to be in a potential bandwidth bottleneck state. This judgment is based on a systematic monitoring mechanism, and the channels formed by each line and its connected nodes will be continuously tracked to capture the high-risk paths that cause transmission interruption or delay fluctuations. After identifying multiple potential bottleneck lines, further perform connectivity analysis based on the network topology structure.The analysis process is based on the logical diagram formed by each node in the network and the lines it is connected to, and identifies whether there is a physical intersection relationship or a data transmission coupling path between high-load lines. If it is found that three or more lines identified as potential bottlenecks in the same network area are topologically connected channels to each other, and the shared intermediate nodes are also in a state of high cache pressure, then this intersection area is identified as a bandwidth bottleneck resonance point. This bandwidth bottleneck resonance point has typical characteristics such as overlapping and dense transmission paths, concentrated data streams, and high flow control difficulty, and is in an important hub position for data scheduling in the network structure. All identified bottleneck resonance points will be abstracted into "transmission topology bottleneck centers", and a bottleneck identification graph will be constructed to reflect the distribution state of each center in the entire network structure and the path range it affects. This bottleneck identification graph will be superimposed on the information transmission topology graph of the e-commerce platform to form a visual risk map, which will be used as a key reference basis in operations such as information scheduling task rearrangement, bandwidth resource priority adjustment, and line congestion warning push.

[0112] Preferably, step S32 includes the following steps:

[0113] Step S321: Estimate the situation of excessive information transmission flow according to the growth of information transmission data volume;

[0114] In the embodiment of the present invention, it is necessary to monitor the data transmission flow of the e-commerce platform to obtain the transmission data volume in each time period. Through historical transmission data and traffic logs, analyze the trend of traffic increase and decrease in each time period, and calculate the data transmission volume in a period of time (such as per minute). According to the change of the data transmission volume, set a threshold value (for example, 6.7 MB / min) to evaluate whether the data traffic is too large. If the data transmission volume exceeds this threshold value in some time periods, it is initially judged that the information transmission flow has a tendency to be too large, and further analysis and processing are required. Monitor and count the data traffic, use network traffic monitoring tools to collect the traffic data of each node during the transmission process, and perform real-time tracking in combination with the fluctuation trend of historical data. By comparing the difference between the historical maximum transmission flow and the current flow, judge whether there is an unexpected traffic increase, so as to estimate whether the information transmission flow will further increase and identify whether it reaches the standard of excessive flow.

[0115] Step S322: Detect the insufficient processing capacity of the transmission network when the excessive information transmission flow exceeds 6.7 MB / min;

[0116] In the embodiment of the present invention, when it is confirmed in step S321 that the information transmission traffic exceeds the set threshold (6.7 MB / min), it is necessary to further evaluate whether the processing capacity of the network transmission device can meet the current traffic demand. The processing capacity of the transmission network is usually limited by the hardware performance of the network device, the processing algorithm, and the bandwidth. In this step, the real-time load condition of the network device will be checked, the bottleneck points in the network will be analyzed, and the performance bottlenecks existing in the transmission path will be detected. In specific operations, the real-time processing load information of each transmission node (such as switches, routers, etc.) is obtained. According to the CPU usage rate, memory occupancy rate of the device, and the bandwidth occupancy of the network port, it is analyzed whether there is a situation of insufficient processing capacity. If it is found that the usage rate of the device processing capacity exceeds the set threshold (for example, 78.8%), it is determined that the transmission network has a situation of insufficient processing capacity, resulting in its inability to effectively process the current information transmission traffic.

[0117] Step S323: Predict the local congestion degree of the transmission network when the insufficient processing capacity condition of the transmission network is greater than 78.8%;

[0118] In the embodiment of the present invention, once it is detected that the processing capacity of the transmission network is greater than the set threshold (for example, 78.8%), next, it is necessary to predict the local congestion situation that occurs by analyzing the processing capacity of each node in the network. At this time, the analysis target is to identify which network nodes or links have excessive data traffic, resulting in local congestion. By comprehensively considering the traffic, processing capacity, and bandwidth utilization rate of each network node, and combining the network topology structure, it is evaluated which transmission paths have the greatest congestion risk. During the implementation process, real-time data traffic information is obtained from the network traffic monitoring system, and according to the real-time load condition of the device, the matching situation between the current bandwidth usage rate and the processing capacity of each line is calculated. Using the historical data trend and combining with the current load data, the congestion area that will occur in the future period of time is speculated, and the congested area in the network is marked.

[0119] Step S324: Detect the abnormal situation of data flow balance when the local congestion degree of the transmission network is greater than or equal to 1.5 due to the excessive information transmission traffic;

[0120] In the embodiments of the present invention, when it is predicted in step S323 that the local congestion degree is greater than or equal to 1.5, it is necessary to further evaluate whether there is an imbalance in the data stream. An abnormal data stream balance usually means that the data traffic on some paths is too high, while other paths are not fully utilized. To detect the balance of the data stream, it is necessary to collect the actual transmission data of each line, calculate the data traffic distribution on different lines, and analyze whether there is a phenomenon that some lines are overloaded while other lines are underloaded. During the operation, the load of each line is calculated based on the real-time traffic data, and the imbalance on the transmission path is detected through the data stream balance evaluation algorithm. By comparing the traffic ratio of each line with the expected traffic distribution in the network design, it is judged whether there is an obvious traffic imbalance phenomenon. Once a data traffic imbalance is found, it will further exacerbate the local congestion and affect the overall network performance.

[0121] Step S325: Estimate the overload situation of the node processing capacity according to the data stream balance abnormality and the local congestion degree of the transmission network;

[0122] In the embodiments of the present invention, in the case of confirming the existence of an abnormal data stream balance, combined with the local congestion degree of the transmission network, it is further evaluated whether the processing capacity of each node in the network is overloaded. The imbalance of the data stream will cause some nodes to bear too high a load, which in turn leads to the overload of the processing capacity of these nodes. By analyzing the load situation of each node in the transmission path in detail, the nodes with overloaded processing capacity are identified. During the implementation, each node in the network is monitored in real time, combined with the processing capacity indicators of each node (such as CPU occupancy rate, memory usage rate, etc.), and the data traffic distribution, to evaluate whether the processing capacity of each node has exceeded its maximum carrying capacity. When the load of a certain node exceeds its set threshold, it is judged that the node has an overloaded processing capacity, and the performance degradation trend of the overall network is estimated accordingly.

[0123] Step S326: Estimate the load growth situation of the transmission line based on the overload situation of the node processing capacity and the data stream balance abnormality.

[0124] In the embodiment of the present invention, after confirming the overload of the processing capacity of the node and the abnormality of data flow balance in the previous step, it is necessary to estimate the load growth of the transmission line based on this information. Since node overload and traffic imbalance will directly affect the network load condition, it is necessary to calculate the load growth trend of each transmission line in the future based on these factors. By collecting the historical traffic data and current load of each node, and combining with the traffic prediction algorithm, the load growth of each transmission line is simulated. According to factors such as the bandwidth utilization rate of each line, the processing capacity of the node, and the balance of data traffic, the changes in future data traffic are predicted, and then the change trend of line load is estimated. This prediction will help network administrators identify potential bottlenecks and overloaded lines in advance, so as to adjust the scheduling strategy in time and avoid network overload.

[0125] Particularly importantly, step S33 includes the following steps:

[0126] Step S331: Analyze the attenuation of network transmission performance according to the load growth of the transmission line;

[0127] In the embodiment of the present invention, the load data of the transmission line is collected, and this data is obtained from the transmission flow statistics of each level of node. The transmission flow is usually collected by a traffic monitoring tool in the network management system. This tool is based on the interface statistical information of network devices and records the data traffic of the transmission line in real time through network probes and monitoring protocols. These traffic data will be preprocessed to eliminate unnecessary noise signals to ensure data accuracy. Then, by analyzing these load data, a curve of the line load growth trend is obtained, which reflects the change of line traffic per unit time. In specific operations, trend analysis is performed based on historical load data, and the regression analysis method is used to predict the load growth rate in the future for a period of time. At this time, by monitoring the data traffic change rate of each time period and combining with the calculation of the load growth rate, the dynamic growth of the current line load is obtained. If it is found that the load growth exceeds the normal range and there is a significant deviation from the historical trend, it indicates that the network resources are insufficient, resulting in performance attenuation. In this case, the network parameters of this line, such as bandwidth occupancy rate, delay, and packet loss rate, will be further monitored to verify whether it is affected by the load growth.

[0128] Step S332: Detect the aggravation of network jitter based on the attenuation of network transmission performance;

[0129] In the embodiments of the present invention, detecting the exacerbation of transmission network jitter mainly relies on the changes in network latency and packet loss rate. According to the transmission performance degradation data obtained in step S331, analyze the changing trends of line latency and packet loss rate. If it is found that the latency fluctuation of the line increases and the packet loss rate rises significantly, it is inferred that the network jitter has exacerbated. Network jitter is usually caused by factors such as insufficient bandwidth, network contention, or abnormal interference. Obtain the latency data of each line from the network monitoring system, and detect the standard deviation and maximum value of the latency through time series analysis. If the latency fluctuation increases significantly and exceeds the preset normal fluctuation range, it is considered that the network jitter has exacerbated. At this time, it is necessary to further analyze the impact of this jitter on the transmission quality and check the influencing factors on the network, such as routing policies and network topology changes.

[0130] Step S333: Estimate the decrease in transmission throughput based on the exacerbation of transmission network jitter and the degradation of network transmission performance;

[0131] In the embodiments of the present invention, based on the exacerbation of network jitter in step S332 and combined with the transmission performance degradation obtained in step S331, estimate the decrease in transmission throughput. The decrease in transmission throughput is usually jointly affected by multiple factors such as insufficient bandwidth, network contention, and increased latency. By analyzing the historical data of network jitter and performance degradation and using the sliding window method, calculate the changing trend of throughput in different time periods. By constructing a throughput prediction model based on historical data, consider factors such as network load, bandwidth occupancy, and jitter. If the prediction result indicates that the transmission throughput will decrease significantly in a certain future time period, adjust network resources in advance or take optimization measures according to this situation.

[0132] Step S334: Predict the degree of exacerbation of network transmission competition based on the decrease in transmission throughput and the exacerbation of transmission network jitter;

[0133] In the embodiments of the present invention, step S334 predicts the degree of exacerbation of network transmission competition based on the decrease in transmission throughput and the exacerbation of network jitter. Network transmission competition refers to the phenomenon that multiple nodes or tasks simultaneously compete for network bandwidth, which will lead to insufficient bandwidth resources and a decrease in throughput. In this step, analyze the network bandwidth occupancy of each node according to the trend of the decrease in transmission throughput, and combine the indicators of the exacerbation of network jitter to determine whether there is a risk of bandwidth competition. Based on the estimated decrease in transmission throughput in step S333, identify those nodes facing insufficient bandwidth, and predict the overall network bandwidth competition trend by calculating the bandwidth occupancy rate of each node and the competition situation of the nodes. If the gap in bandwidth occupancy rates between nodes increases, it is determined that the transmission competition has exacerbated, thus affecting the overall transmission efficiency.

[0134] Step S335: Estimate the data transmission cache overflow situation based on the degree of intensifying network transmission competition and the decrease in transmission throughput.

[0135] In the embodiments of the present invention, based on the degree of intensifying network transmission competition and the decrease in transmission throughput, estimate the data transmission cache overflow situation. Data cache overflow means that the cache capacity of some nodes in the network is insufficient to handle excessive data traffic, resulting in packet loss or retransmission. According to the prediction result of intensifying transmission competition obtained in step S334, combined with the trend of decreasing transmission throughput, determine which nodes' caches are at risk of overflow. Calculate the cache capacity and current cache usage of each node, and combined with the trend of decreasing transmission throughput of the node, estimate whether its cache processing capacity can meet the requirements. If it is found that the cache capacity of some nodes does not match the growth of data traffic and there is a trend of overflow, it is necessary to estimate the probability of overflow, especially in high-load situations. In this process, utilize the distribution of data traffic, the cache capacity of nodes, and the changes in transmission throughput.

[0136] Particularly importantly, step S34 includes the following steps:

[0137] Step S341: Estimate the data transmission loss probability based on the data transmission cache overflow situation;

[0138] In the embodiments of the present invention, through the evaluation of the cache overflow situation in the foregoing steps, collect various indicators related to the data transmission cache. These indicators include but are not limited to cache usage rate, frequency of cache overflow events, duration of cache overflow, and the amount of data at the time of overflow. By analyzing the data of these overflow events, combined with the growth of the load on the transmission line, determine whether the cache can no longer process the incoming data stream in a timely manner. In specific implementation, analyze the historical records of cache overflow to determine the time, frequency, and related data traffic at the time of overflow. On this basis, utilize traffic estimation technology to compare the change trend of data traffic with the response ability of the cache. For each packet transmission process, evaluate whether it is affected by cache overflow, and then calculate the probability of loss during the transmission process. The loss probability is usually obtained by analyzing the ratio of the number of packets that failed to be transmitted successfully at the time of cache overflow to the total number of packets, so as to accurately estimate the data loss probability that will occur in a certain period in the future.

[0139] Step S342: Detect the data traffic retransmission situation based on the data transmission loss probability;

[0140] In the embodiments of the present invention, the estimated loss probability in step S341 is used to further detect the duplicate transmission situation in the data traffic. Duplicate transmission usually occurs after data loss, when the source node re - sends the packets that have not been successfully transmitted. To detect duplicate transmission situations, it is necessary to collect the packet sequence number, transmission timestamp, and the status of the data packet (whether it is successfully received) for each data transmission through a network - layer monitoring tool. By comparing different transmission paths of the same data packet in the network, duplicate transmission events are identified. Specifically, when implementing, a data - packet tracking mechanism is deployed at each node, and tracking is carried out by recording the unique identifier of each data packet (such as the packet sequence number or hash value). Then, these identifiers are compared with the transmission logs to determine which data packets have been transmitted repeatedly. If it is found that the duplicate transmission situation is relatively frequent, it means that the lost data packets cannot be recovered or re - transmitted in time, resulting in a waste of bandwidth resources. Therefore, it is necessary to perform additional transmission monitoring on the data packets with a high loss probability during network transmission to reduce the occurrence of duplicate transmission situations.

[0141] Step S343: Estimate the growth of the occupied transmission bandwidth resources according to the duplicate transmission situation of the data traffic and the estimated data loss probability;

[0142] In the embodiments of the present invention, the goal of step S343 is to estimate the growth of the occupied bandwidth resources due to the duplicate transmission of the data traffic and the data loss probability. A high probability of duplicate transmission and loss will directly increase the amount of bandwidth used in the network because each unsuccessfully transmitted data packet will be re - sent. By collecting the historical records of duplicate transmission, including the number of data packets transmitted repeatedly each time and the required time, combined with the data of the loss probability, the growth of bandwidth occupancy in a future period of time is predicted. The bandwidth usage of each node is obtained through the monitoring system, and the additional bandwidth consumed during each data re - transmission is recorded. Then, these data are combined with the loss probability, and the growth trend of the bandwidth resources is obtained through incremental calculation. If it is found that the bandwidth occupancy increases significantly during a certain period of time, measures need to be taken, such as adjusting the traffic control strategy or optimizing the data transmission path, to reduce the bandwidth occupancy caused by duplicate transmission.

[0143] Step S344: Estimate the saturation of the processing capacity of the transmission line based on the growth of the transmission - line load;

[0144] In the embodiments of the present invention, according to the load growth condition of the transmission line, the saturation condition of the processing capacity of the transmission line is estimated. The saturation of the processing capacity of the transmission line generally means that the line cannot handle more data traffic under high load, resulting in a significant decrease in data transmission efficiency. In this step, it is necessary to analyze the load growth of each line, especially the sharp rise section in the load growth curve. By comparing historical data with the current load status, it is predicted whether the processing capacity of the line is approaching the saturation point. By real-time monitoring indicators such as the bandwidth occupancy rate, latency, and packet loss rate of the transmission line, combined with the load growth trend, a capacity estimation tool is used to predict the processing capacity of the line. If the load continues to grow and the bandwidth and processing capacity of the line cannot be expanded in time, the line will likely reach the saturation state of the processing capacity at a certain future time, resulting in an increase in data transmission latency or an exacerbation of packet loss. At this time, measures must be taken in advance, such as expanding the bandwidth, optimizing the network topology, or performing traffic scheduling, to avoid line saturation.

[0145] Step S345: Calculate the attenuation of the data stream transmission efficiency based on the saturation condition of the transmission line processing capacity and the load growth condition of the transmission line;

[0146] In the embodiments of the present invention, based on the saturation condition of the transmission line processing capacity and the load growth condition, the attenuation of the data stream transmission efficiency is calculated. The attenuation of the transmission efficiency is usually accompanied by the saturation of the bandwidth, an increase in the packet transmission latency on the line, and intensified network competition, resulting in a decrease in throughput. By analyzing the relationship between the load growth and the line processing capacity, the change trend of the data transmission efficiency is inferred. Through real-time monitoring and historical data analysis, combined with key performance indicators such as the bandwidth utilization rate, packet loss rate, and network latency, the transmission efficiency of each node is calculated. If the load growth of the line approaches the saturation state, the transmission efficiency will begin to decline, and phenomena such as packet queuing and network jitter will occur. By comparing the transmission efficiency in different time periods, the quantitative data of the efficiency attenuation is obtained.

[0147] Step S346: Detect the bandwidth bottleneck condition of the transmission line according to the attenuation of the data stream transmission efficiency and the growth of the transmission bandwidth resource occupancy.

[0148] In the embodiments of the present invention, according to the attenuation of the data stream transmission efficiency and the growth of the transmission bandwidth resource occupancy, it is detected whether there is a bandwidth bottleneck in the transmission line. The bandwidth bottleneck usually manifests as the data transmission rate being unable to keep up with the transmission demand under high load, resulting in an increase in transmission latency or data loss. By comparing the attenuation curve of the data stream transmission efficiency with the bandwidth occupancy curve, it is detected whether there is a bottleneck phenomenon. The bandwidth occupancy data of each transmission line is obtained from the network monitoring system and compared with the transmission efficiency curve of the data stream. If it is found that there is a significant correlation between the bandwidth occupancy rate and the attenuation of the transmission efficiency of a certain line, and the transmission latency increases, it can be determined that there is a bandwidth bottleneck in this line.

[0149] Preferably, step S4 includes the following steps:

[0150] Step S41: Statistically analyze the transmission line response delay parameters according to the bandwidth bottleneck condition of the transmission line;

[0151] In the embodiment of the present invention, by collecting the real-time monitoring data of the transmission line, the response delay parameters of each transmission line are obtained. The response delay refers to the time required for data to travel from the sending end to the receiving end. In the case of a bandwidth bottleneck, the transmission delay usually increases. Therefore, it is necessary to statistically analyze the response delay of each line according to the bandwidth bottleneck condition during the information transmission process. Specifically, during operation, according to the bandwidth bottleneck condition of the transmission line, the network state of each transmission line is analyzed in real time through a network performance monitoring tool (such as the NetFlow or SNMP protocol). Record the time interval from the sending to the receiving of the data packet to obtain the response time of each data packet. Through multiple samplings and accumulations, calculate the average response delay of each line, and then obtain the response delay value of the line under the bandwidth bottleneck condition. In addition, it is also necessary to combine the network topology to identify the delay change trend of each bottleneck location.

[0152] Step S42: Estimate the data loss condition of the information transmission based on the transmission line response delay parameters and the bandwidth bottleneck condition of the transmission line;

[0153] In the embodiment of the present invention, based on the transmission line response delay parameters statistically analyzed in the previous step and the bandwidth bottleneck condition in the current network, the data loss situation in the information transmission is further evaluated. The increase in the bandwidth bottleneck and response delay usually leads to data packet loss because the network cannot process a large number of transmission requests in time. Use a network monitoring system (such as Wireshark or a dedicated network analysis tool) to track the data packets in the network and record the status of the data packets sent and received on each transmission line. By comparing the predetermined transmission success rate with the actual number of received data packets, initially judge the data loss situation. If the packet loss rate significantly increases in the bandwidth bottleneck section, it indicates that the bandwidth carrying capacity of this line has exceeded its maximum load, resulting in the failure of the data packet to be successfully transmitted. Combining the response delay parameters can further predict future data loss situations. The increase in delay is often accompanied by buffer overflow and traffic overload, thus affecting the smooth transmission of data packets. Based on the change trend of the delay parameters, evaluate the data loss probability in the network through a simulation or prediction algorithm (such as a queue model or queuing theory).

[0154] Step S43: Detect the information transmission anomaly condition according to the information data transmission loss condition and the transmission line response delay parameters;

[0155] In an embodiment of the present invention, after detecting a data loss situation in step S42, in combination with the transmission line response delay parameter, it is further analyzed whether there is an abnormal information transmission situation. Abnormal information transmission usually refers to problems such as too high a packet loss rate, too long a delay, or an unstable transmission path, resulting in the inability to transmit information normally. During the operation process, it is necessary to summarize and analyze the traffic data and delay information from each transmission node. Combining the packet loss situation and its response delay value of each line, a real-time monitoring system is used to comprehensively evaluate the network health status. If the packet loss rate of a certain line is significantly higher than the normal range and the delay exceeds the expected value, it is determined that there is an abnormal transmission situation on this line. Further, analyze the bottleneck position of the line, combine the change law of the transmission period, determine the time and location of the abnormality, and then identify the transmission link in the network that is most likely to have problems. Through the dynamic tracking of abnormal situations and the analysis of historical data, it is possible to monitor the abnormal information transmission in the network in real time, and timely discover and handle the bottlenecks and unstable factors in the network.

[0156] Step S44: Based on the abnormal information transmission situation, perform information transmission link optimization processing to obtain information transmission link optimization data.

[0157] In an embodiment of the present invention, after confirming the abnormal information transmission, finally, it is necessary to perform information transmission link optimization processing based on the existing abnormal data. The goal of this optimization process is to reduce or eliminate abnormal phenomena and improve transmission efficiency and stability by adjusting the network structure, adjusting the transmission path, or reallocating traffic. Analyze the abnormal transmission link in detail to identify bottleneck, overloaded, or unbalanced areas. Improve the link performance by means such as traffic redirection, adjusting bandwidth allocation, optimizing data routing, or adding redundant lines. For example, if there is a bottleneck in some transmission lines, increase the bandwidth of this line, or redirect the data traffic to a relatively idle line through a load balancing strategy, so as to reduce the load on the bottleneck part. During the link optimization process, it is necessary to re-plan and adjust the transmission path to ensure load balancing of each node in the network. Based on the real-time monitoring and analysis results, use a network traffic scheduling tool to dynamically adjust the traffic allocation, so that the data can be transmitted on the optimal path. Through comprehensive analysis of the load status, transmission efficiency, and response time of each line, a set of optimized information transmission link optimization data is obtained and provided to the information scheduling system as a decision-making basis.

[0158] The present invention also provides an information transmission system applied to an e-commerce platform for executing the information transmission method applied to the e-commerce platform as described above. The information transmission system applied to the e-commerce platform includes:

[0159] A topology construction module, which is used to obtain the e-commerce platform log data; collect the information transmission line data according to the e-commerce platform log data; construct an information transmission topology diagram according to the information transmission line data and the e-commerce platform log data;

[0160] A payload damage evolution detection module, which is used to predict the order dynamic impact situation data according to the e-commerce platform log data; evaluate the abnormal growth condition of the warehousing and distribution load index based on the order dynamic impact situation data; detect the physical flow dynamic payload damage evolution data according to the abnormal growth condition of the warehousing and distribution load index;

[0161] A transmission line bandwidth bottleneck detection module, which is used to count the growth of the information transmission data volume according to the physical flow dynamic payload damage evolution data and the order dynamic impact situation data; estimate the growth of the transmission line load for the information transmission topology based on the growth of the information transmission data volume; detect the transmission line bandwidth bottleneck condition according to the growth of the transmission line load;

[0162] An information transmission link optimization module, which is used to detect the information transmission abnormal condition according to the transmission line bandwidth bottleneck condition; perform information transmission link optimization processing based on the information transmission abnormal condition to obtain information transmission link optimization data.

[0163] The above are only specific embodiments of the present invention, enabling those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but rather to the widest scope consistent with the principles and novel features invented herein.

Claims

1. An information transmission method applied to an e-commerce platform, characterized in that, Including the following steps: Step S1: Obtain the e-commerce platform log data; collect the information transmission line data according to the e-commerce platform log data; Construct an information transmission topology structure diagram based on the information transmission line data and the e-commerce platform log data; Step S2: Predict the order dynamic impact situation data according to the e-commerce platform log data; evaluate the abnormal growth condition of the warehousing and distribution load index based on the order dynamic impact situation data; detect the evolution data of the physical flow dynamic load damage according to the abnormal growth condition of the warehousing and distribution load index; Step S3: Statistically analyze the growth of the information transmission data volume according to the physical flow dynamic load damage evolution data and the order dynamic impact situation data; estimate the growth condition of the transmission line load for the information transmission topology structure based on the growth of the information transmission data volume; Detect the bandwidth bottleneck condition of the transmission line according to the growth condition of the transmission line load; Step S4: Detect the information transmission abnormal condition according to the bandwidth bottleneck condition of the transmission line; Perform information transmission link optimization processing based on the information transmission abnormal condition to obtain information transmission link optimization data.

2. The information transmission method applied to an e-commerce platform according to claim 1, wherein, Step S1 includes the following steps: Step S11: Obtain the e-commerce platform log data; Step S12: Collect the information transmission line data according to the e-commerce platform log data; Step S13: Collect the information transmission node data according to the e-commerce platform log data; Step S14: Construct an information transmission topology structure diagram based on the information transmission node data and the information transmission line data.

3. The information transmission method applied to an e-commerce platform according to claim 2, wherein Step S14 includes the following steps: Step S141: Statistically analyze the node information interaction intensity according to the information transmission node data and the information transmission line data; Step S142: Statistically analyze the node information interaction density according to the information transmission node data and the information transmission line data; Step S143: Evaluate the transmission node link coverage coupling degree data when the node information interaction density exceeds 97.5 pkt / s and the node information interaction intensity exceeds 100 pkt / s; Step S144: Identify the real-time flow direction of the transmission data stream based on the data transmission line data; Step S145: Calculate the data transmission edge bandwidth data using the real-time flow direction of the transmission data stream and the data transmission line data; Step S146: Analyze the data transmission edge characteristics based on the data transmission edge bandwidth data and the real-time flow direction of the transmission data stream; Step S147: Construct an information transmission topology structure diagram based on the data transmission edge characteristics and the transmission node link coverage coupling degree data.

4. The information transmission method applied to an e-commerce platform according to claim 1, wherein The prediction of the order dynamic impact situation data in Step S2 includes: Perform order time series analysis according to the e-commerce platform log data to obtain the business platform order time series data; Monitor the change trend of the order volume according to the business platform order time series data; Construct a business platform order change fluctuation diagram based on the order volume change trend; Calculate the business platform change fluctuation slope parameter according to the business platform order change fluctuation diagram; Identify the sudden sharp increase situation of the business platform order according to the business platform order change fluctuation diagram and the business platform change fluctuation slope parameter; Statistically analyze the data of the concentrated time period of the order sudden increase according to the business platform order change fluctuation diagram and the sudden sharp increase situation of the business platform order; Predict the dynamic impact situation data of orders based on the data during the concentrated time period of sudden order increase and the sudden increase of orders on the business platform.

5. The information transmission method applied to an e-commerce platform according to claim 1, characterized in that The evaluation of the abnormal growth situation of the warehousing and distribution load index in step S2 includes: Analyze the concentration status of warehousing sorting tasks based on the dynamic impact situation data of orders; Estimate the frequency of warehousing area scheduling based on the concentration status of warehousing sorting tasks; Evaluate the instantaneous shortage of warehousing and distribution resources according to the frequency of warehousing area scheduling and the concentration status of warehousing sorting tasks; Monitor the congestion status of distribution warehousing sorting according to the instantaneous shortage of warehousing and distribution resources; Detect the sudden increase in the load of the warehousing and distribution center based on the congestion status of distribution warehousing sorting and the instantaneous shortage of warehousing and distribution resources; Determine the critical value state of the bottleneck of the warehousing node according to the sudden increase in the load of the warehousing and distribution center and the congestion degree of distribution warehousing sorting; Evaluate the abnormal growth situation of the warehousing and distribution load index according to the critical value state of the bottleneck of the warehousing node and the sudden increase in the load of the warehousing and distribution center.

6. The information transmission method applied to an e-commerce platform according to claim 1, characterized in that, The detection of the evolution data of the damage of the physical flow dynamic load in step S2 includes: Detect the overloading of goods loading according to the abnormal growth situation of the warehousing and distribution load index; Measure the stacking clearance parameter of goods based on the overloading of goods loading; Detect the squeezing of distribution goods loading based on the stacking clearance parameter of goods and the overloading of goods loading; Measure the growth degree of the loading pressure of distribution goods according to the squeezing of distribution goods loading; Measure the lack of buffer space for distribution goods based on the growth degree of the loading pressure of distribution goods and the stacking clearance parameter of goods; Detect the propagation of the impact force of goods by using the growth degree of the loading pressure of distribution goods for the lack of buffer space for distribution goods; Estimate the impact deformation trend of distribution goods according to the propagation of the impact force of goods; Detect the evolution data of the damage of the physical flow dynamic load based on the impact deformation trend of distribution goods and the propagation of the impact force of goods.

7. The information transmission method applied to an e-commerce platform according to claim 1, characterized in that Step S3 includes the following steps: Step S31: Statistically analyze the growth of the data volume of information transmission according to the evolution data of the damage of the physical flow dynamic load and the dynamic impact situation data of orders; Step S32: Estimate the growth situation of the load of the transmission line for the information transmission topology based on the growth of the data volume of information transmission; Step S33: Estimate the overflow situation of the data transmission cache according to the growth situation of the load of the transmission line; Step S34: Detect the bandwidth bottleneck situation of the transmission line according to the overflow situation of the data transmission cache and the growth situation of the load of the transmission line.

8. The information transmission method applied to an e-commerce platform according to claim 7, characterized in that Step S32 includes the following steps: Step S321: Estimate the situation of excessive information transmission flow according to the growth of the data volume of information transmission; Step S322: Detect the insufficient processing capacity of the transmission network when the excessive information transmission flow exceeds 6.7MB / min; Step S323: Predict the local congestion degree of the transmission network when the insufficient processing capacity of the transmission network is greater than 78.8%; Step S324: Detect the abnormal situation of data flow balance when the local congestion degree of the transmission network is greater than or equal to 1.5 by using the excessive information transmission flow; Step S325: Estimate the overload situation of the node processing capacity according to the abnormal situation of data flow balance and the local congestion degree of the transmission network. Step S326: Estimate the load growth situation of the transmission line based on the overload situation of node processing capabilities and the abnormal situation of data flow balance.

9. The information transmission method applied to an e-commerce platform according to claim 1, characterized in that, Step S4 includes the following steps: Step S41: Statistically calculate the transmission line response delay parameter according to the bandwidth bottleneck situation of the transmission line; Step S42: Estimate the information transmission data loss situation based on the transmission line response delay parameter and the bandwidth bottleneck situation of the transmission line; Step S43: Detect the information transmission abnormal situation according to the information data transmission loss situation and the transmission line response delay parameter; Step S44: Perform information transmission link optimization processing based on the information transmission abnormal situation to obtain information transmission link optimization data.

10. An information transmission system applied to an e-commerce platform, characterized in that, For implementing the information transmission method applied to an e-commerce platform as described in claim 1, the information transmission system applied to an e-commerce platform includes: A topology structure construction module, configured to obtain e-commerce platform log data; collect information transmission line data according to the e-commerce platform log data; construct an information transmission topology structure diagram according to the information transmission line data and the e-commerce platform log data; A load damage evolution detection module, configured to predict order dynamic impact trend data according to the e-commerce platform log data; evaluate the abnormal growth situation of the warehousing and distribution load index based on the order dynamic impact trend data; detect the physical flow dynamic load damage evolution data according to the abnormal growth situation of the warehousing and distribution load index; A transmission line bandwidth bottleneck detection module, configured to statistically calculate the growth of information transmission data volume according to the physical flow dynamic load damage evolution data and the order dynamic impact trend data; estimate the load growth situation of the transmission line for the information transmission topology structure based on the growth of information transmission data volume; detect the bandwidth bottleneck situation of the transmission line according to the load growth situation of the transmission line; An information transmission link optimization module, configured to detect the information transmission abnormal situation according to the bandwidth bottleneck situation of the transmission line; perform information transmission link optimization processing based on the information transmission abnormal situation to obtain information transmission link optimization data.

Images