Vibration test data transmission scheduling optimization method based on ant colony algorithm

Abstract

The invention belongs to the technical field of data transmission, and discloses a vibration test data transmission scheduling optimization method based on an ant colony algorithm. The invention aims to solve the problems of low data transmission speed, data loss, poor transmission stability and the like in the prior art. The method comprises the following steps: firstly, constructing a data transmission path set by acquiring topological structure information of a vibration test data transmission network; determining a data transmission task set according to the transmission requirement of the vibration test data; establishing a data transmission scheduling model based on the two, including resource allocation, priority and time limit constraint; carrying out optimization solution on the model by utilizing an ant colony algorithm to obtain an optimal data transmission scheduling scheme which comprises an optimal transmission path and an optimal transmission sequence; and finally, executing data transmission according to the scheme, and dynamically updating a path resource state. According to the invention, the data transmission efficiency, reliability and stability are improved, the data loss risk is reduced, and the network resource utilization rate is optimized.

Description

Technical Field

[0001] This invention relates to the field of data transmission technology, and more specifically, to a method for optimizing the scheduling of vibration test data transmission based on ant colony algorithm. Background Technology

[0002] Vibration test data transmission has significant application value in industrial measurement of micro-vibrations in aerospace, rail transportation, and shipbuilding engineering. Its core objective is to ensure that vibration signals collected by sensors are transmitted in real-time and completely to data centers or cloud servers for analysis and processing through efficient and stable data transmission, thereby providing a reliable basis for equipment condition monitoring, fault diagnosis, and performance optimization. However, existing technologies have many shortcomings in the scheduling and optimization of vibration test data transmission, severely limiting its application in high-precision, complex, and variable scenarios.

[0003] First, the problem of slow data transmission speed is particularly prominent. For example, in aerospace vibration testing, the massive amounts of data collected by sensors need to be transmitted to the data center for real-time analysis within a short period of time. However, existing technologies lack dynamic assessment and optimized scheduling of network resource status, leading to blind selection of transmission paths and frequent entrapment in low-bandwidth or high-latency paths, resulting in low transmission efficiency. Especially during peak network load periods or in complex environments with multiple nodes transmitting concurrently, the transmission speed may even drop to the point where it cannot meet real-time requirements. Second, data loss is a common problem. For example, in the monitoring of the operating status of rail transit vehicles, if critical vibration signal data is lost due to network congestion, latency fluctuations, or path anomalies, it will directly lead to misjudgments or omissions in fault diagnosis. Existing technologies lack effective prevention and remedial measures for this. Third, poor transmission stability is a significant problem. For example, in long-term vibration testing, the network environment may fluctuate frequently due to equipment movement, signal interference, or load changes. Existing technologies fail to monitor and dynamically adjust transmission paths in real time, easily leading to transmission interruptions or sudden drops in efficiency. Furthermore, existing technologies prioritize data transmission tasks too simplistically, relying solely on crude indicators such as data volume or transmission time limits, failing to deeply analyze the content characteristics of vibration test data. For example, in aero-engine testing, vibration signal data containing fault characteristics should be transmitted first, but existing technologies cannot identify their importance, leading to delays or loss of critical data. Additionally, the problem of a single data segmentation method is also significant. For instance, the temporal volatility of data varies significantly in different vibration testing scenarios, but existing technologies typically use a fixed segmentation granularity, failing to adaptively adjust according to data characteristics. This results in poor transmission stability when data fluctuations are severe, and low transmission efficiency when data is stable. Moreover, the problem of low network resource utilization cannot be ignored. For example, in multi-node distributed testing systems, existing technologies lack a global optimization mechanism; some transmission paths may become congested due to overload, while other paths remain idle, resulting in significant resource waste. More importantly, existing technologies lack robust anomaly handling mechanisms. For example, when network packet loss rates surge or latency fluctuations are abnormal, they fail to detect and take timely measures to retransmit data or reselect paths, leading to transmission task failures. This deficiency can have serious consequences, especially in industrial scenarios with high reliability requirements.

[0004] In view of this, the present invention proposes a vibration test data transmission scheduling optimization method based on ant colony algorithm to solve the above problems. Summary of the Invention

[0005] To overcome the above-mentioned deficiencies of the prior art and to achieve the above objectives, the present invention provides the following technical solution: a vibration test data transmission scheduling optimization method based on ant colony algorithm, comprising: acquiring the topology information of the vibration test data transmission network, wherein the topology information includes the connection relationship between the acquisition module, transmission lines, data center and cloud server, as well as the bandwidth, delay and load status of each transmission line; Based on the topology information, a data transmission path set is constructed, which includes complete transmission paths from the acquisition module to the data center and from the data center to the cloud server. Based on the transmission requirements of vibration test data, a set of data transmission tasks is determined. The set of data transmission tasks includes several data transmission tasks, and each data transmission task includes the number of data blocks, priority, and transmission time limit. Based on the set of data transmission paths and the set of data transmission tasks, a data transmission scheduling model is established. The data transmission scheduling model includes resource allocation constraints for transmission paths, priority constraints for data transmission tasks, and transmission time constraints. The data transmission scheduling model is optimized and solved based on the ant colony algorithm to obtain the optimal data transmission scheduling scheme, which includes the optimal transmission path and transmission order for each data transmission task. According to the optimal data transmission scheduling scheme, the vibration test data is transmitted, and the resource status of the data transmission path set is dynamically updated during the transmission process.

[0006] The technical effects and advantages of the vibration test data transmission scheduling optimization method based on ant colony algorithm of the present invention are as follows: This invention improves data transmission efficiency, enabling the rapid transmission of massive amounts of vibration data under stringent real-time requirements. This ensures the timeliness and accuracy of test results, particularly in critical scenarios such as dynamic performance evaluation and fault diagnosis, effectively preventing decision-making delays caused by transmission latency. Secondly, this invention enhances data transmission reliability, minimizing data loss and guaranteeing the complete transmission of key vibration signals, thereby reducing the risk of misjudgments or omissions due to missing data. Furthermore, the invention's improvement in transmission stability is particularly significant, maintaining stable operation even in complex and variable network environments. This avoids transmission interruptions or efficiency degradation caused by network fluctuations, providing solid support for long-term, high-reliability testing tasks. Simultaneously, by optimizing resource allocation, this invention significantly improves network resource utilization efficiency, avoiding resource waste and enabling the entire transmission system to achieve higher performance at a lower cost. Moreover, the invention's robust anomaly handling allows for rapid recovery in the event of network anomalies, greatly improving robustness and fault tolerance, ensuring the continuity and stability of industrial applications. Attached Figure Description

[0007] Figure 1 This is a schematic diagram of a vibration test data transmission scheduling optimization method based on ant colony algorithm according to the present invention. Detailed Implementation

[0008] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0009] This application provides a vibration test data transmission scheduling optimization method based on ant colony algorithm. The execution subject of the method includes, but is not limited to: vibration test equipment, data acquisition gateway, edge computing unit, cloud server, etc., which can be regarded as general computing nodes of this application. The data transmission control system includes, but is not limited to: network controller, distributed data scheduling system, programmable network controller, and at least one of the following:

[0010] This invention provides a vibration test data transmission scheduling optimization method based on ant colony optimization. By acquiring real-time vibration test network topology information and data transmission requirements, a precise set of data transmission paths and tasks is constructed. The data transmission scheduling model is then optimized using the ant colony optimization algorithm, achieving efficient transmission of vibration test data. This method is highly adaptive, capable of optimizing transmission parameters in real-time based on network resource status and transmission task characteristics, significantly improving network utilization and data transmission efficiency while reducing transmission latency and data loss rate.

[0011] Please see Figure 1 In this embodiment of the invention, the detailed implementation steps of a vibration test data transmission scheduling optimization method based on ant colony algorithm include: First, the topology information of the vibration test data transmission network is acquired. This topology information includes the connections between the acquisition modules, transmission lines, data center, and cloud server, as well as the bandwidth, latency, and load status of each transmission line. This data is collected in real time using network detection tools and the network management system. This data provides the basis for formulating transmission scheduling strategies, ensuring the targeting and effectiveness of the scheduling.

[0012] Based on the topology information, a set of data transmission paths is constructed. This set includes complete transmission paths from the acquisition module to the data center and from the data center to the cloud server. The set of transmission paths is extracted from the topology using graph theory algorithms, providing a feasible path space for subsequent scheduling decisions. The construction process considers network connectivity redundancy and alternative solutions to ensure that there are still usable transmission paths in the event of network fluctuations.

[0013] Based on the transmission requirements of vibration test data, a set of data transmission tasks is determined. This set includes several data transmission tasks, each specifying the number of data chunks, priority, and transmission time limit. The task set reflects all data transmission requests that currently need to be processed, providing a basis for resource allocation decisions. The determination of transmission requirements combines the real-time requirements of vibration testing with data integrity requirements, ensuring the timely transmission of critical data.

[0014] A data transmission scheduling model is established based on a set of data transmission paths and a set of data transmission tasks. This model includes resource allocation constraints for transmission paths, priority constraints for data transmission tasks, and transmission time constraints. The scheduling model is a multi-objective optimization problem, aiming to balance network resource utilization, transmission task completion time, and priority requirements, providing optimization objectives and constraints for the ant colony algorithm.

[0015] The ant colony algorithm is used to optimize the data transmission scheduling model and obtain the optimal data transmission scheduling scheme. This scheme includes the optimal transmission path and transmission order for each data transmission task. By simulating ant foraging behavior and combining pheromone mechanisms with heuristic factors, the ant colony algorithm searches for the global optimum in the feasible solution space, effectively avoiding getting trapped in local optima.

[0016] Based on the optimal data transmission scheduling scheme, vibration test data is transmitted, and the resource status of the data transmission path set is dynamically updated during the transmission process. This step applies the optimization results to the actual transmission process, while maintaining the latest information on path resource status through real-time monitoring, providing an accurate basis for subsequent scheduling. The dynamic update mechanism can respond promptly to network fluctuations, maintaining the adaptability and robustness of scheduling.

[0017] The data transmission scheduling model is optimized and solved based on the ant colony algorithm, including: Initialize the parameters of the ant colony algorithm, including the number of ants, pheromone importance factor, heuristic factor importance, initial pheromone concentration, and pheromone evaporation factor. Parameter initialization is fundamental to algorithm execution; reasonable parameter settings directly affect the convergence speed and solution quality. The number of ants is typically set to 20-50, dynamically adjusted according to the problem size. The pheromone importance factor α and heuristic factor importance β control the algorithm's dependence on historical experience and current heuristic information, respectively, and are usually set between 1 and 3. The initial pheromone concentration is set to a small, uniform value to prevent premature convergence. The pheromone evaporation factor ρ controls the pheromone decay rate, typically set between 0.1 and 0.5 to ensure a balance between the algorithm's exploration capability and convergence speed.

[0018] Ant colonies are randomly deployed at each node of the data transmission path set, with each ant representing a scheduling decision for a data transmission task. The initial distribution of ants adopts a random uniform distribution strategy to ensure sufficient exploration of the solution space in the initial stage of the algorithm. Each ant carries complete decision information, including the assigned transmission task, current position, and the path already traversed, providing basic data for path selection and pheromone updates.

[0019] Based on the priority of data transmission tasks and the resource status of transmission paths, the transition probability of each ant moving from the current transmission path node to the next transmission path node is calculated. The calculation of the transition probability comprehensively considers pheromone concentration, bandwidth utilization of the transmission path, and latency fluctuation rate, serving as the core basis for the ant's decision-making. The calculation process integrates historical experience (pheromone) with the current state (resource status), balancing global optimization and local optima, ensuring a balance between the algorithm's convergence and exploratory nature.

[0020] Based on the transition probability, the path selection for each ant is determined, and the data transmission efficiency corresponding to each path selection is recorded. Path selection employs a roulette wheel selection strategy; paths with higher probabilities are more likely to be selected, but this uncertainty ensures the algorithm's exploratory capability. Data transmission efficiency is a key indicator for evaluating the quality of path selection, typically considering factors such as transmission time, resource utilization, and task completion rate, providing a basis for pheromone updates.

[0021] After each ant completes a path search, the pheromone concentration of each transmission path in the data transmission path set is updated. This pheromone concentration update, based on the data transmission efficiency of path selection and the pheromone evaporation factor, is the core mechanism for the algorithm's learning and evolution. The update process first calculates the pheromone increment contributed by each ant, then considers the natural pheromone evaporation, ultimately obtaining a new pheromone distribution. Efficient paths receive more pheromone enhancements, while inefficient paths gradually lose pheromone, guiding the algorithm towards the global optimum.

[0022] The algorithm iterates through path selection and pheromone concentration updates until a preset convergence condition is met, outputting the optimal data transmission scheduling scheme. The iterative process is the core of the algorithm's gradual optimization of the solution; through repeated path exploration and pheromone updates, the algorithm continuously improves the quality of the solution. Convergence conditions typically include the maximum number of iterations or the improvement margin of the solution in multiple consecutive iterations falling below a threshold, ensuring that the algorithm obtains a high-quality solution within a reasonable timeframe. The final output optimal data transmission scheduling scheme includes the optimal path selection and execution order for each data transmission task, directly guiding the actual transmission process.

[0023] Methods for calculating transition probabilities include: The system acquires resource status information for the current transmission path nodes, including bandwidth utilization, latency fluctuation, and historical transmission success rate. Resource status information is a direct indicator of transmission path quality and is collected in real-time using network monitoring tools. Bandwidth utilization reflects the path's remaining capacity, latency fluctuation reflects the path's stability, and historical transmission success rate reflects the path's reliability. These three indicators together constitute a comprehensive assessment of path quality, providing fundamental data for calculating the transition probability.

[0024] Based on the resource status information, calculate the heuristic factor from the current transmission path node to the next transmission path node. The formula for calculating the heuristic factor is as follows: ; in, Indicates time From the current transmission path node To the next transmission path node heuristic factor, Indicates the transmission path arrive Available bandwidth, Indicates the transmission path arrive Historical transmission success rate Indicates the transmission path arrive average latency, Indicates the transmission path arrive The delayed volatility.

[0025] Heuristic factors quantify the immediate quality of the path and serve as a crucial basis for ant's decision-making. The formula design reflects four key considerations: the greater the available bandwidth, the better; the higher the historical success rate, the better; the lower the average latency, the better; and the lower the latency fluctuation rate, the better. This comprehensive evaluation ensures the holistic nature of path selection, balancing throughput, reliability, and latency requirements to adapt to the needs of different types of data transmission tasks.

[0026] Obtain the pheromone concentration from the current transmission path node to the next transmission path node. Pheromones concentration reflects the accumulated experience of ants on this path and is a manifestation of swarm intelligence. Pheromones are directly obtained from the path pheromone matrix, which is dynamically updated in each iteration and records the historical decision-making experience of all ants. A high pheromone concentration indicates that the path has performed well in historical decision-making, increasing the probability that this path will be chosen in the current decision.

[0027] Based on the heuristic factor and pheromone concentration, the transition probability is calculated using the following formula: ; in, Indicates time From the current transmission path node To the next transmission path node The transition probability, Represents the set of all reachable transport path nodes. Indicates the pheromone importance factor. Indicates the importance of the heuristic factor.

[0028] The transition probability is the direct basis for ant path selection, integrating historical experience and a comprehensive evaluation of the current state. In the formula, the pheromone concentration term... Representing the influence of historical experience, heuristic factors The product of the two terms represents the impact of the current state, reflecting a balance between experience and real-time state. The denominator ensures that the sum of the transition probabilities of all reachable nodes is 1, forming a standardized probability distribution. and The value of determines the degree to which the algorithm depends on historical experience and the current state; a larger value indicates a greater dependence. The value makes the algorithm more reliant on historical experience, and a larger value... This makes the algorithm pay more attention to the current path state.

[0029] Among them, delayed volatility Calculated in the following manner: for the transmission path arrive The latency values ​​are statistically analyzed using a sliding window to obtain the standard deviation of latency within the sliding window. The ratio of the standard deviation to the average latency within the sliding window is used as the latency volatility. Latency volatility quantifies the stability of the path and is an important indicator for evaluating path quality. The calculation uses a sliding window technique, typically with a window size of 10-30 sampling points, which considers both the statistical characteristics of historical latency and maintains the timeliness of the indicator. The standard deviation reflects the amplitude of latency fluctuations, while the average latency provides a benchmark reference. The ratio of the two forms a dimensionless volatility indicator, facilitating comparisons between different paths. The lower the volatility, the more stable the path, and the more suitable it is for transmitting latency-sensitive data.

[0030] Methods for updating pheromone concentrations include: After each ant completes a path search, the data transmission efficiency corresponding to the path selection is obtained. The formula for calculating the data transmission efficiency is: ; in, Indicates the first The path selection of an ant corresponds to the data transmission efficiency. Indicates the first The amount of data transmitted by a single ant Indicates the first The total time it takes for an ant to complete data transmission. Indicates the first The number of transmission failures in the ant path selection.

[0031] Data transmission efficiency is a comprehensive indicator for evaluating path selection quality and directly affects the amount of pheromone updates. The formula design reflects three core considerations: the larger the amount of data transmitted, the better; the shorter the total transmission time, the better; and the fewer the number of transmission failures, the better. The ratio of the amount of data transmitted to the total transmission time reflects the basic transmission rate, while the number of failures is placed in the denominator as a penalty factor, ensuring the important role of transmission reliability in the evaluation. This comprehensive evaluation ensures that the algorithm not only pursues high throughput but also emphasizes transmission stability and reliability, meeting the needs of actual network transmission.

[0032] Based on data transmission efficiency, the pheromone increment is calculated using the following formula: ; in, Indicates the first Only ants on the transmission path arrive The increase in pheromones on the surface This is the normalization constant.

[0033] The pheromone increment determines the ant's contribution to path evaluation and is the direct basis for pheromone updates. The increment is proportional to data transmission efficiency, reflecting the principle that "good paths receive more positive feedback." The normalization constant C controls the absolute magnitude of the pheromone increment, typically set to 1-10, dynamically adjusted according to the problem size to ensure the pheromone value remains within a reasonable range, avoiding excessively large or small values. High-efficiency paths receive more pheromone enhancements, while low-efficiency paths have smaller pheromone increments, guiding subsequent ants to favor more efficient paths.

[0034] The pheromone concentration is updated based on the pheromone increment and pheromone volatile factor. The formula for updating the pheromone concentration is as follows: ; in, Indicates time Transmission path arrive pheromone concentration, Indicates the pheromone volatile factor. This indicates the number of ants.

[0035] Pheromone concentration updates are the core mechanism of algorithm learning and evolution, integrating the decay of historical experience with the addition of new knowledge. In the formula, The item indicates that historical pheromones after evaporation should be considered. The term represents the sum of pheromone contributions from all ants in this iteration. The pheromone evaporation mechanism prevents the algorithm from converging prematurely to a local optimum, ensuring continuous exploration capability; while the addition of pheromones strengthens high-quality paths, promoting the algorithm's convergence towards the global optimum. Balancing these two mechanisms is crucial for the algorithm's success; dynamically adjusting the ρ value can control the algorithm's exploration and convergence tendencies.

[0036] The pheromone evaporation factor ρ is dynamically adjusted based on the load status of the transmission path. When the load rate of the transmission path is higher than a preset load threshold, ρ is increased to accelerate pheromone evaporation and reduce the dependence of path selection on historical information; when the load rate of the transmission path is lower than the preset load threshold, ρ is decreased to enhance the stability of path selection. This dynamic adjustment mechanism enables the algorithm to adapt to changes in network load, enhancing exploratory behavior under high load and enhancing stability under low load. The preset load threshold is typically set at 70-80% and dynamically adjusted according to network characteristics. The adjustment range of the value is typically ±30% of the baseline value to ensure the stability of the algorithm's performance. This mechanism significantly improves the algorithm's adaptability to complex and variable network environments and is a key innovation of this method.

[0037] Methods for determining the priority of data transmission tasks include: The system acquires transmission requirement information for each data transmission task, including the number of data chunks, transmission time limit, and data importance level. This transmission requirement information is the foundational data for determining task priority, directly reflecting the task's characteristics and requirements. The number of data chunks reflects the task's scale, the transmission time limit reflects the task's urgency, and the data importance level reflects the task's business value. This information is specified through system configuration or upper-layer applications, providing comprehensive input for priority calculation.

[0038] Based on the transmission demand information, the priority of each data transmission task is calculated. The priority calculation formula is as follows: ; in, Indicates the first The priority of each data transmission task Indicates the first Number of data blocks per data transmission task This represents the largest number of data blocks across all data transfer tasks. Indicates the first The actual transmission time of each data transmission task Indicates the first The transmission time limit for each data transmission task. Indicates the first The data importance level of each data transmission task These are weighting factors for the number of data blocks, transmission time limit, and data importance level, respectively. .

[0039] The priority calculation formula integrates three dimensions: task size, time urgency, and business importance, forming a comprehensive assessment. (First item) The task size was normalized, reflecting the principle of prioritizing large tasks; the second item This reflects the urgency of the task; the smaller the time leeway, the higher the priority. (Third item) This directly reflects the task's business importance. The three dimensions are weighted and merged using weight coefficients to form the final priority value. The weight coefficients are usually set according to system characteristics and business needs, with a typical configuration of w1=0.2, w2=0.5, and w3=0.3, reflecting the dominant role of time urgency in the priority ranking.

[0040] The data transmission task set is sorted according to priority, and the sorting result serves as the priority constraint for path selection in the ant colony algorithm. The sorting process uses a non-increasing order, with higher-priority tasks listed first and receiving priority scheduling rights. During the execution of the ant colony algorithm, ants carrying high-priority tasks choose paths first, while ants carrying low-priority tasks consider the remaining network resources before making a decision. This mechanism ensures that important and urgent tasks receive priority resource allocation, and through the overall optimization of the ant colony algorithm, it avoids the resource waste that might result from a simple greedy strategy.

[0041] The data importance level Ik is determined as follows: content analysis is performed on the vibration test data to extract its spectral and temporal features; based on these features, it is determined whether the data contains key vibration signals. If key vibration signals are present, the data importance level is increased; otherwise, it is decreased. This content-aware importance assessment mechanism is a significant innovation of this method, ensuring that priority determination is based not only on metadata but also on the actual value of the data content. Feature extraction employs techniques such as Fast Fourier Transform and wavelet analysis. The identification of key vibration signals is based on preset thresholds or machine learning models, enabling accurate identification of data blocks containing abnormal or important information, assigning them higher transmission priority, and ensuring the timely delivery of critical data.

[0042] The methods for dynamically updating the resource status of a data transmission path set include: During data transmission, the resource status of each transmission path in the data transmission path set is monitored in real time. Resource status includes bandwidth utilization, latency, and packet loss rate. Resource status monitoring is the foundation for dynamic adjustments and is obtained in real time through network probing tools or system APIs. Monitoring employs a combination of periodic sampling and event triggering, ensuring data timeliness while avoiding the additional burden of overly frequent monitoring. Bandwidth utilization reflects the remaining capacity of the path, latency reflects the time cost of transmission, and packet loss rate reflects the reliability of transmission; together, these three factors constitute a comprehensive assessment of path quality.

[0043] Based on the resource status, the dynamic load factor of each transmission path is calculated. The formula for calculating the dynamic load factor is as follows: ; in, Indicates time Transmission path arrive The dynamic load factor Indicates the transmission path arrive Used bandwidth Indicates the transmission path arrive Total bandwidth, Indicates the transmission path arrive The current delay, This represents the maximum delay across all transmission paths. Indicates the transmission path arrive The packet loss rate is given by λ and μ, which are the weighting factors for latency and packet loss rate, respectively.

[0044] The dynamic load factor comprehensively considers bandwidth utilization, latency level, and packet loss, serving as a holistic indicator for assessing path load status. In the formula, the bandwidth utilization term reflects the proportion of resources used, the latency normalization term reflects the relative latency level, and the packet loss rate... This directly reflects transmission reliability. The three indicators are weighted and fused together using weighting factors to form a unified load assessment. Weighting factors , Typical settings for the dynamic load factor are λ=0.5 and μ=0.3, reflecting the dominant role of bandwidth occupancy in load assessment while also fully considering the impact of latency and packet loss. The dynamic load factor ranges from [0, 1], with larger values ​​indicating heavier path loads and making it less suitable for allocating new transmission tasks.

[0045] When the dynamic load factor exceeds a preset load threshold, a path reselection mechanism is triggered. This mechanism includes: pausing data transmission tasks on the current transmission path; re-invoking the ant colony algorithm to recalculate the transition probability and select a new transmission path based on the latest resource state information; and reallocating the paused data transmission tasks to the new transmission path and resuming data transmission. The path reselection mechanism is a proactive strategy to address network congestion, ensuring that transmission tasks can adjust their paths in a timely manner, avoiding efficiency degradation caused by continuing transmission on congested paths. The preset load threshold is typically set to 0.8-0.9; reselection is triggered when the load factor exceeds this value. The reselection process first safely pauses the current transmission, saving the transmission state and completed portions. Then, based on the latest network state, the ant colony algorithm is re-executed to select a new optimal path. Finally, transmission is resumed and execution continues. This dynamic adjustment mechanism significantly improves adaptability and robustness in complex and changing network environments, and is a key feature of this method.

[0046] Methods for determining preset convergence conditions include: After each iteration, the global transmission efficiency of the optimal data transmission scheduling scheme for the current iteration is calculated. The formula for calculating the global transmission efficiency is: ; in, Indicates time Global transmission efficiency, Indicates the total number of data transmission tasks. Indicates the first The amount of data in each data transmission task Indicates the first The transmission time of each data transmission task Indicates the first The number of transmission failures for each data transmission task. This is a penalty factor for transmission failure.

[0047] Global transmission efficiency is a comprehensive indicator for evaluating the overall quality of the current scheduling scheme and directly affects the algorithm's convergence determination. The formula design considers overall data throughput, transmission time, and reliability, and is an integration of the transmission efficiency of individual tasks. For each task, its efficiency is the ratio of data volume to time and failure penalty; the global efficiency is the sum of the efficiencies of all tasks. The value is typically set between 1 and 5, and is dynamically adjusted according to reliability requirements. A larger value indicates a more severe penalty for failure. Higher global transmission efficiency indicates a better scheduling scheme and that it is closer to the global optimum.

[0048] The algorithm compares the difference in global transmission efficiency between two consecutive iterations. If the difference is less than a preset efficiency threshold, the algorithm is considered to have met the preset convergence condition; alternatively, if the number of iterations reaches a preset maximum number of iterations, the algorithm is considered to have met the preset convergence condition. The convergence condition is the basis for determining when the algorithm stops iterating, and it includes two dimensions: the magnitude of efficiency improvement and the maximum number of iterations. The efficiency difference judgment reflects the algorithm's adaptive stopping strategy; when the improvement from consecutive iterations is small, the algorithm can be considered close to the optimal solution. The maximum number of iterations is a guarantee mechanism to ensure that the algorithm completes within a finite time. The preset efficiency threshold is usually set to 0.1-1% of the previous efficiency, and the preset maximum number of iterations is usually 100-500, dynamically adjusted according to the problem size and time requirements. This dual judgment mechanism ensures the quality of the solution while avoiding unnecessary computational waste.

[0049] The preset efficiency threshold is dynamically adjusted based on the transmission time limit of the vibration test data. When the transmission time limit is short, the preset efficiency threshold is decreased to improve the convergence speed; when the transmission time limit is long, the preset efficiency threshold is increased to improve the optimization degree of the scheduling scheme. This dynamic adjustment mechanism allows the algorithm to flexibly adjust the optimization strategy according to actual business needs, prioritizing convergence speed in emergency situations and pursuing higher quality solutions when time is ample. The adjustment strategy typically uses a piecewise function or an exponential function, determining the threshold adjustment coefficient based on the ratio of the time limit to the reference time. For example, when the time limit is less than 50% of the reference time, the threshold is reduced to half of the reference value; when the time limit is more than 200% of the reference time, the threshold is increased to twice the reference value. This mechanism enables the algorithm to exhibit good adaptability under different time pressures and is an important optimization for the practical application of this method.

[0050] The methods for determining the number of data blocks in a data transmission task include: The total data volume and characteristics of the vibration test data are obtained, including the spectral distribution and temporal volatility. These factors form the basis for determining the data partitioning strategy and directly influence the choice of partitioning granularity. The total data volume is obtained directly from the file size or the number of data records, while the data characteristics are extracted through signal processing algorithms. The spectral distribution reflects the frequency composition of the data, and the temporal volatility reflects the data's variation characteristics; both together determine the data's complexity and redundancy, providing a basis for partitioning decisions.

[0051] Based on the data characteristics, the granularity of the data blocks is determined. The method for determining the granularity of the data blocks is as follows: wavelet decomposition is performed on the vibration test data to obtain the sub-band energy of different frequency bands; based on the sub-band energy, the time-domain volatility of the data is calculated. The formula for calculating the time-domain volatility is: ; in, Indicates the time-domain volatility of the data. This indicates the number of frequency bands in the wavelet decomposition. Indicates the first Subband energy of each frequency band Indicates the first The standard deviation of signals in each frequency band.

[0052] Temporal volatility is a comprehensive indicator for assessing the drastic nature of data changes and directly influences the choice of block granularity. The calculation process first involves multi-level wavelet decomposition of the original data, typically using Daubechies or Haar wavelet bases, with a decomposition depth of 3-5 levels to obtain sub-bands in different frequency bands. Then, the energy and standard deviation of each sub-band are calculated. Energy reflects the importance of the frequency band, and standard deviation reflects the degree of volatility within the band. The weighted product of these two values ​​forms the sub-band contribution, and the sum of all sub-band contributions is the temporal volatility. A larger volatility value indicates more drastic data changes and lower internal correlation, making a smaller block granularity more suitable. Conversely, a smaller volatility value indicates more stable data with higher internal correlation, allowing for a larger block granularity to improve transmission efficiency.

[0053] When data volatility exceeds a preset volatility threshold, the granularity of data blocks is reduced to improve data transmission stability; conversely, when volatility falls below the preset threshold, the granularity of data blocks is increased to improve data transmission efficiency. This adaptive block-splitting strategy dynamically adjusts the block-splitting scheme based on data characteristics, ensuring both transmission stability and optimized efficiency. The preset volatility threshold is typically set based on historical data statistics or expert experience and serves as a key reference point for judging data characteristics. The adjustment of block granularity follows a piecewise or exponential function; for example, when volatility exceeds the threshold by 50%, the granularity is reduced to half the baseline value; when volatility falls below the threshold by 50%, the granularity is increased to twice the baseline value. This fine-grained control ensures a precise match between the block-splitting strategy and data characteristics, improving overall transmission performance.

[0054] The number of data blocks is calculated based on the granularity of the data blocks and the total data volume. The number of data blocks is a direct result of task decomposition and determines the granularity and scale of parallel transmission. The calculation formula is the total data volume divided by the block granularity and rounded up, ensuring that all data is included in the blocks. Determining the appropriate number of blocks is crucial for optimizing transmission efficiency, avoiding both increased management overhead from too many small blocks and decreased parallelism from too few large blocks. In practical applications, the calculation results are checked for reasonableness to ensure the number of blocks is within a manageable range (typically 10-1000), and fine-tuning is performed as necessary to balance management overhead and parallel efficiency.

[0055] Anomaly detection and recovery mechanisms include: During data transmission, the packet loss rate and latency fluctuation rate of each transmission path in the data transmission path set are monitored in real time. Packet loss rate and latency fluctuation rate are key indicators for evaluating transmission quality and directly reflect the health status of the path. Monitoring adopts a combination of periodic sampling and threshold triggering, which ensures the continuity of monitoring while avoiding the additional burden of over-monitoring. The packet loss rate is calculated by comparing the number of packets sent and received, and the latency fluctuation rate is calculated by the ratio of the standard deviation to the average value of multiple consecutive latency measurements. These two indicators can sensitively capture abnormal changes in network status and provide a reliable basis for anomaly detection.

[0056] When the packet loss rate of a transmission path exceeds a preset packet loss threshold or the latency fluctuation rate exceeds a preset fluctuation threshold, the transmission path is deemed to have encountered an anomaly. Anomaly detection is a prerequisite for triggering the recovery mechanism, and the use of dual-indicator judgment ensures the accuracy and comprehensiveness of detection. The preset packet loss threshold is typically set at 1-5%, and the preset fluctuation threshold is typically set at 0.2-0.5%, dynamically adjusted according to network characteristics and service requirements. This dual-indicator judgment strategy can capture different types of network anomalies; packet loss rate is sensitive to connection reliability issues, while latency fluctuation rate is sensitive to network congestion and routing instability. This comprehensive judgment improves the accuracy and coverage of anomaly detection.

[0057] For data transmission tasks on an abnormal transmission path, a data retransmission mechanism is triggered. This mechanism includes: determining retransmission priority based on the data transmission task's priority; for high-priority data transmission tasks, immediately re-invoking the ant colony algorithm to select a new transmission path and execute data retransmission; for low-priority data transmission tasks, adding them to the pending task queue and reallocating transmission paths in the next scheduling cycle. The data retransmission mechanism is a proactive response strategy to transmission anomalies, implementing differentiated processing based on task importance. Retransmission priority is typically inherited from the original task priority, but can be appropriately adjusted based on the severity of the anomaly and the proportion of tasks already completed. High-priority tasks receive immediate processing, ensuring timely delivery of critical data; low-priority tasks enter a queue to wait, avoiding system performance degradation caused by resource contention. This hierarchical processing strategy ensures both the continuity of critical business operations and overall stability, and is the core design of the anomaly recovery mechanism.

[0058] The scheduling order of the pending tasks queue is dynamically adjusted based on the priority and time limit of data transmission tasks. When the remaining time limit of a data transmission task is lower than a preset time limit threshold, its scheduling order in the pending tasks queue is increased. This dynamic adjustment mechanism ensures the flexibility and adaptability of queue management and avoids the starvation problem that may be caused by static priorities. The adjustment process comprehensively considers the original priority and time urgency of the tasks. When the remaining time limit is close to the preset threshold (usually 10-20% of the original time limit), the task's position in the queue is automatically promoted, ensuring that it can get a processing opportunity before the deadline. This "temporary priority promotion" mechanism effectively balances importance and urgency, improves the responsiveness to time-sensitive tasks, and is an important supplement to the anomaly recovery mechanism.

[0059] This invention achieves efficient scheduling and reliable transmission of vibration test data through ant colony optimization and a multi-dimensional monitoring and adjustment mechanism. The adaptive optimization feature of this invention can adjust transmission parameters in real time according to network resource status and transmission task characteristics, significantly improving network utilization and data transmission efficiency while reducing transmission latency and data loss rate.

[0060] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0061] It should be noted that all formulas in this manual are calculated by removing dimensions and taking their numerical values. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0062] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. A vibration test data transmission scheduling optimization method based on ant colony algorithm, characterized in that, include: The topology information of the vibration test data transmission network is obtained, including the connection relationship between the acquisition module, transmission lines, data center and cloud server, as well as the bandwidth, delay and load status of each transmission line; Based on the topology information, a data transmission path set is constructed, which includes complete transmission paths from the acquisition module to the data center and from the data center to the cloud server. Based on the transmission requirements of vibration test data, a set of data transmission tasks is determined. The set of data transmission tasks includes several data transmission tasks, and each data transmission task includes the number of data blocks, priority, and transmission time limit. Based on the set of data transmission paths and the set of data transmission tasks, a data transmission scheduling model is established. The data transmission scheduling model includes resource allocation constraints for transmission paths, priority constraints for data transmission tasks, and transmission time constraints. The data transmission scheduling model is optimized and solved based on the ant colony algorithm to obtain the optimal data transmission scheduling scheme, which includes the optimal transmission path and transmission order for each data transmission task. According to the optimal data transmission scheduling scheme, the vibration test data is transmitted, and the resource status of the data transmission path set is dynamically updated during the transmission process.

2. The vibration test data transmission scheduling optimization method based on ant colony algorithm according to claim 1, characterized in that, The optimization and solution of the data transmission scheduling model based on the ant colony algorithm includes: Initialize the parameters of the ant colony algorithm, including the number of ants, pheromone importance factor, heuristic factor importance, initial pheromone concentration, and pheromone volatility factor; Ant colonies are randomly deployed on each transmission path node of the data transmission path set, and each ant in the ant colony represents a scheduling decision for a data transmission task. Based on the priority of the data transmission task and the resource status of the transmission path, the transfer probability of each ant moving from the current transmission path node to the next transmission path node is calculated. The calculation of the transfer probability takes into account the pheromone concentration, the bandwidth utilization of the transmission path, and the latency fluctuation rate. Based on the transition probability, determine the path selection for each ant and record the data transmission efficiency corresponding to the path selection for each ant. After each ant completes a path search, the pheromone concentration of each transmission path in the data transmission path set is updated. The update of the pheromone concentration is based on the data transmission efficiency of the path selection and the pheromone evaporation factor. The path selection and pheromone concentration update process is executed iteratively until the preset convergence condition is met, and the optimal data transmission scheduling scheme is output.

3. The vibration test data transmission scheduling optimization method based on ant colony algorithm according to claim 2, characterized in that, The method for calculating the transition probability includes: Obtain the resource status information of the current transmission path node, including bandwidth utilization, latency fluctuation rate, and historical transmission success rate; Based on the resource status information, calculate the heuristic factor from the current transmission path node to the next transmission path node. The formula for calculating the heuristic factor is as follows: ； in, Indicates time From the current transmission path node To the next transmission path node heuristic factor, Indicates the transmission path arrive Available bandwidth, Indicates the transmission path arrive Historical transmission success rate Indicates the transmission path arrive average latency, Indicates the transmission path arrive The delayed volatility; Obtain the pheromone concentration from the current transmission path node to the next transmission path node. ; Based on the heuristic factor and the pheromone concentration, the transition probability is calculated using the following formula: ； in, Indicates time From the current transmission path node To the next transmission path node The transition probability, Represents the set of all reachable transport path nodes. Indicates the pheromone importance factor. Indicates the importance of heuristic factors; Wherein, the delay volatility Calculated in the following way: For transmission path arrive The delay value is statistically analyzed using a sliding window to obtain the delay standard deviation within the sliding window, and the ratio of the delay standard deviation to the average delay within the sliding window is taken as the delay volatility.

4. The vibration test data transmission scheduling optimization method based on ant colony algorithm according to claim 2, characterized in that, The method for updating the pheromone concentration includes: After each ant completes a path search, the data transmission efficiency corresponding to the path selection is obtained. The formula for calculating the data transmission efficiency is: ； in, Indicates the first The path selection of an ant corresponds to the data transmission efficiency. Indicates the first The amount of data transmitted by a single ant Indicates the first The total time it takes for an ant to complete data transmission. Indicates the first The number of transmission failures in ant path selection; Based on the data transmission efficiency, the pheromone increment is calculated, and the formula for calculating the pheromone increment is as follows: ； in, Indicates the first Only ants on the transmission path arrive The increase in pheromones on the surface This is the normalization constant; The pheromone concentration is updated based on the pheromone increment and pheromone evaporation factor, and the update formula for the pheromone concentration is as follows: ； in, Indicates time Transmission path arrive pheromone concentration, Indicates the pheromone volatile factor. Indicates the number of ants; Among them, the pheromone volatile factor The load is dynamically adjusted based on the load status of the transmission path. When the load rate of the transmission path exceeds a preset load threshold, the load is increased. To accelerate pheromone evaporation and reduce the reliance of path selection on historical information; when the load rate of the transmission path is lower than a preset load threshold, reduce... To enhance the stability of path selection.

5. The vibration test data transmission scheduling optimization method based on ant colony algorithm according to claim 2, characterized in that, The method for determining the priority of the data transmission task includes: Obtain the transmission requirement information for each data transmission task, including the number of data blocks, transmission time limit, and data importance level; Based on the transmission requirement information, the priority of each data transmission task is calculated, and the priority calculation formula is as follows: ； in, Indicates the first The priority of each data transmission task Indicates the first Number of data blocks per data transmission task This represents the largest number of data blocks across all data transfer tasks. Indicates the first The actual transmission time of each data transmission task Indicates the first The transmission time limit for each data transmission task. Indicates the first The data importance level of each data transmission task These are weighting factors for the number of data blocks, transmission time limit, and data importance level, respectively. ; The data transmission task set is sorted according to the priority, and the sorting result is used as the priority constraint for path selection in the ant colony algorithm. Among them, the data importance level Determined in the following ways: Content analysis is performed on vibration test data to extract spectral and temporal characteristics. Based on the spectral and temporal characteristics, it is determined whether the data contains key vibration signals. If key vibration signals are contained, the importance level of the data is increased; if key vibration signals are not contained, the importance level of the data is decreased.

6. The vibration test data transmission scheduling optimization method based on ant colony algorithm according to claim 2, characterized in that, The method for dynamically updating the resource status of the data transmission path set includes: During data transmission, the resource status of each transmission path in the data transmission path set is monitored in real time. The resource status includes bandwidth utilization, latency, and packet loss rate. Based on the resource status, the dynamic load factor of each transmission path is calculated. The formula for calculating the dynamic load factor is as follows: ； in, Indicates time Transmission path arrive The dynamic load factor, Indicates the transmission path arrive The bandwidth already used Indicates the transmission path arrive Total bandwidth, Indicates the transmission path arrive The current delay, This represents the maximum delay across all transmission paths. Indicates the transmission path arrive packet loss rate, , These are the weighting factors for latency and packet loss rate, respectively. When the dynamic load factor is higher than a preset load threshold, a path reselection mechanism is triggered, which includes: Pause the data transmission task on the current transmission path; The ant colony algorithm is invoked again, and based on the latest resource status information, the migration probability is recalculated and a new transmission path is selected. The paused data transmission task is reassigned to a new transmission path, and data transmission resumes.

7. The vibration test data transmission scheduling optimization method based on ant colony algorithm according to claim 2, characterized in that, The method for determining the preset convergence condition includes: After each iteration, the global transmission efficiency of the optimal data transmission scheduling scheme for the current iteration is calculated. The formula for calculating the global transmission efficiency is as follows: ； in, Indicates time Global transmission efficiency, Indicates the total number of data transmission tasks. Indicates the first The amount of data in each data transmission task Indicates the first The transmission time of each data transmission task Indicates the first The number of transmission failures for each data transmission task. This is a penalty factor for transmission failure; Compare the difference in global transmission efficiency between two consecutive iterations. If the difference is less than a preset efficiency threshold, it is determined that the preset convergence condition is met. Alternatively, when the number of iterations reaches the preset maximum number of iterations, it is determined that the preset convergence condition is met; The preset efficiency threshold is dynamically adjusted based on the transmission time of the vibration test data. When the transmission time is short, the preset efficiency threshold is reduced to improve the convergence speed; when the transmission time is long, the preset efficiency threshold is increased to improve the optimization of the scheduling scheme.

8. The vibration test data transmission scheduling optimization method based on ant colony algorithm according to claim 2, characterized in that, The method for determining the number of data blocks in the data transmission task includes: Acquire the total amount of vibration test data and its characteristics, including the spectral distribution and temporal fluctuations of the data. Based on the data characteristics, the granularity of the data blocks is determined. The method for determining the granularity of the data blocks is as follows: The vibration test data were decomposed using wavelet decomposition to obtain the sub-band energy of different frequency bands; Based on the subband energy, the time-domain volatility of the data is calculated, and the formula for calculating the time-domain volatility is as follows: ； in, Indicates the time-domain volatility of the data. This indicates the number of frequency bands in the wavelet decomposition. Indicates the first Subband energy of each frequency band Indicates the first The standard deviation of signals in each frequency band; When the time-domain volatility is higher than a preset volatility threshold, the granularity of data blocks is reduced to improve the stability of data transmission; when the time-domain volatility is lower than the preset volatility threshold, the granularity of data blocks is increased to improve the efficiency of data transmission. The number of data blocks is calculated based on the granularity of the data blocks and the total amount of data.

9. The vibration test data transmission scheduling optimization method based on ant colony algorithm according to claim 2, characterized in that, The method also includes an anomaly detection and recovery mechanism during data transmission, the anomaly detection and recovery mechanism including: During data transmission, the packet loss rate and latency fluctuation rate of each transmission path in the data transmission path set are monitored in real time. When the packet loss rate of a certain transmission path is higher than a preset packet loss threshold or the latency fluctuation rate is higher than a preset fluctuation threshold, the transmission path is determined to be abnormal. For data transmission tasks on abnormal transmission paths, a data retransmission mechanism is triggered, which includes: Determine the retransmission priority based on the priority of the data transmission task; For high-priority data transmission tasks, the ant colony algorithm is immediately invoked again to select a new transmission path and perform data retransmission. For low-priority data transmission tasks, add them to the pending task queue and reallocate the transmission path in the next scheduling cycle; The scheduling order of the task queue is dynamically adjusted according to the priority and transmission time limit of the data transmission task. When the remaining transmission time limit of a data transmission task is lower than the preset time limit threshold, its scheduling order in the task queue is increased.

Images