Processing method and device for time-sensitive event in complex working environment and storage medium

Abstract

The invention discloses a processing method and device for time-sensitive events in a complex working environment and a storage medium. The method comprises the following steps: determining a corresponding time budget value; carrying out heterogeneous decoupling on the multi-source resource data to obtain a resource capability vector; calculating a time margin; calculating system resistance; determining a resource release probability and a resource newly-added occupation probability, and establishing a probability mapping function for the time disturbance quantity; calculating a prediction correction item through a probability mapping function, wherein the prediction correction item is determined according to the resource release probability, the resource newly-added occupation probability and the corresponding income amount and loss amount; calculating a reachability value of the to-be-processed event according to the time margin, the system resistance and the prediction correction term through a preset reachability function, and comparing the reachability value with a preset threshold set; and determining a target processing strategy from the multiple event processing strategies according to the comparison result.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a method, apparatus and storage medium for processing time-sensitive events in complex operating environments. Background Technology

[0002] Complex operational environments typically contain heterogeneous resources such as multiple execution units, communication links, and buffers, and contain a large number of tasks or events that need to be completed within a specified time. In such environments, task processing involves monitoring resource status, managing task queues, and evaluating time budgets. As the number of events and resource distribution increase, the complexity of task scheduling and event processing also increases.

[0003] In practical applications, event handling typically requires considering task deadlines, current system resource status, and task queue conditions to allocate system resources appropriately and arrange processing order. For time-sensitive events, reasonable time budget calculations and task scheduling decisions are crucial for ensuring timely task completion.

[0004] Task processing in complex operating environments typically involves the acquisition and processing of multi-source data, including status information of execution units, communication links, and buffers, as well as the time requirements and priorities of each task in the task queue. This information can be used to evaluate the system's processing capacity within a given time window and serves as a reference for event scheduling and resource allocation. Summary of the Invention

[0005] To address the aforementioned technical problems, this application provides a method, apparatus, and storage medium for processing time-sensitive events in complex operating environments.

[0006] The technical solution provided in this application is described below: The first aspect of this application provides a method for processing time-sensitive events in complex work environments, the method comprising: The corresponding time budget value is determined based on the deadline of the pre-acquired pending events and the current time. Heterogeneous decoupling is performed on multi-source resource data from execution units, communication links, and buffers to obtain resource capability vectors; Based on the resource capability vector and the current task queue status of the system, calculate the time margin used to characterize the effective processing capacity that the system can provide within the time budget value; Calculate the system resistance used to characterize the adverse impact of the system on the processing of the event to be processed within the time budget value; Based on short-term predictions of future resource changes and event arrival conditions, the probability of resource release and the probability of new resource occupancy are determined, and a probability mapping function for time disturbances is established. The prediction correction term is calculated using the probability mapping function, and the prediction correction term is determined based on the resource release probability, the resource new occupation probability, and the corresponding revenue and loss. The reachability value of the event to be processed is calculated based on the time margin, the system resistance, and the prediction correction term using a preset reachability function, and the reachability value is compared with a preset threshold set. The target event handling strategy is determined from a variety of event handling strategies based on the comparison results.

[0007] Optionally, before establishing the probability mapping function for the time perturbation, the method further includes: The probability of resource release and the probability of new resource occupancy are serialized using a short window to obtain a time series expression representing the probability distribution of future changes; Typical distribution patterns of possible resource change states are extracted from the time series expression; Based on the typical distribution pattern, the time budget value range that may be affected by resource changes is determined, and a set of time scopes containing multiple time scopes is formed. Based on the change factors and resource importance corresponding to different typical distribution patterns, calculate the disturbance intensity weight corresponding to each time range. Based on the time range and the disturbance intensity weight, a prediction disturbance model is constructed; The predicted perturbation model is divided into several perturbation units according to a preset time granularity. Each perturbation unit includes a corresponding time range and perturbation intensity weight. Based on the time range and disturbance intensity weight of each disturbance unit, the corresponding disturbance impact factor is calculated to form a set of quantities used to describe the impact of future resource changes on the time budget value. Optionally, establishing the probability mapping function for the time perturbation includes: Based on the set of influence quantities, a perturbation accumulation sequence is constructed to measure the cumulative effect of perturbation; The perturbation accumulation sequence is subjected to nonlinear normalization to obtain the perturbation intensity scale; Based on the disturbance intensity scale and the time sensitivity parameters corresponding to different resource types, a probability mapping function is established to convert the disturbance influence factor into a time disturbance quantity.

[0008] Optionally, the step of calculating the prediction correction term through the probability mapping function, wherein the prediction correction term is determined based on the resource release probability, the resource new occupation probability, and the corresponding revenue and loss amounts, includes: Based on the probability mapping function, each disturbance influence factor is mapped to obtain the corresponding time disturbance sequence. Based on the resource release probability and the resource new occupation probability, calculate the amount of revenue from resource release and the amount of loss caused by resource new occupation, respectively. The time perturbation sequence is weighted and fused with the gain and loss to obtain a time correction sequence for the future prediction window; The time-corrected sequence is subjected to interval compression to obtain a single prediction correction term.

[0009] Optionally, the step of calculating the reachability value of the event to be processed based on the time margin, the system resistance, and the prediction correction term using a preset reachability function includes: The time margin is quantified to obtain a quantified value of the time margin used to characterize the remaining available time corresponding to the event to be processed. The system resistance is normalized to obtain a resistance factor that characterizes the difficulty of system execution. Substitute the prediction correction term into the reachability function, and perform a joint operation on the time margin quantification value and the resistance factor to obtain a reachability value that represents the degree to which the event to be processed can be successfully completed within the target processing window.

[0010] Optionally, determining the target processing strategy from multiple event processing strategies based on the comparison results includes: The preset threshold set includes a first threshold, a second threshold, and a third threshold. The multiple event handling strategies include at least four options, including a normal handling strategy, a preemptive handling strategy, a fast path handling strategy, and a rejection handling strategy. When the reachability value is greater than or equal to the first threshold, the normal processing strategy is selected to perform event processing according to the standard procedure. When the reachability value is between the first threshold and the second threshold, the preemptive processing strategy is selected; When the reachability value is lower than the second threshold but higher than the third threshold, the fast path processing strategy is selected; When the reachability value is lower than the third threshold, the rejection handling strategy is selected.

[0011] Optionally, the calculation used to characterize the system resistance that adversely affects the processing of the event to be processed within the time budget value includes: Obtain the status of the task queue and the estimated execution time for each task; Based on the number of tasks to be executed in the task queue, task priority, and resource usage information, calculate the potential delay contribution to the event to be processed. Calculate the load data of the execution unit, the communication link, and the buffer within a future time budget window; The system resistance is obtained by fusing the potential delay contribution and the load data.

[0012] A second aspect of this application provides an apparatus for processing time-sensitive events in complex working environments, the apparatus comprising: The time budget determination unit is used to determine the corresponding time budget value based on the deadline of the pre-acquired pending events and the current time. The decoupling unit is used to heterogeneously decouple multi-source resource data from the execution unit, communication link and buffer to obtain resource capability vectors; The time margin calculation unit is used to calculate, based on the resource capacity vector and the current task queue status of the system, a time margin that characterizes the effective processing capacity that the system can provide within the time budget value. A system resistance calculation unit is used to calculate the system resistance that characterizes the adverse impact of the system on the processing of the event to be processed within the time budget value. The function building unit is used to determine the probability of resource release and the probability of new resource occupancy based on short-term predictions of future resource changes and event arrival conditions, and to establish a probability mapping function for time disturbances. The function calculation unit is used to calculate the prediction correction term through the probability mapping function. The prediction correction term is determined based on the resource release probability, the resource new occupation probability, and the corresponding revenue and loss. The comparison unit is used to calculate the reachability value of the event to be processed based on the time margin, the system resistance and the prediction correction term using a preset reachability function, and compare the reachability value with a preset threshold set. The strategy determination unit is used to determine the target processing strategy from multiple event processing strategies based on the comparison results.

[0013] A third aspect of this application provides an apparatus for processing time-sensitive events in complex working environments, the apparatus comprising: Processor, memory, input / output units, and bus; The processor is connected to the memory, the input / output unit, and the bus; The memory stores a program, which the processor invokes to execute the first aspect and any one of the optional methods in the first aspect.

[0014] A fourth aspect of this application provides a computer-readable storage medium on which a program is stored, which, when executed on a computer, performs the methods of the first aspect and any one of the first aspects.

[0015] As can be seen from the above technical solutions, this application has the following beneficial effects: The method for handling time-sensitive events in complex operating environments provided in this application comprehensively processes multi-dimensional information such as the time budget value of the event to be processed, system resource capabilities, task queue status, and future resource change trends. This enables a quantitative assessment of the reachability of time-sensitive events and selects an adaptive event handling strategy accordingly. This method can improve the decision-making accuracy of event handling in complex operating environments where resources are dynamically changing and task competition coexist, making the event handling process more consistent with the actual available resources and time requirements of the system.

[0016] By establishing a correlation between time margin, system resistance, and prediction correction terms, this method can more accurately reflect changes in the system's processing capacity within future time windows, thereby improving the ability to judge the feasibility of events. The strategy selection mechanism based on reachability values ​​enables the event processing process to automatically match appropriate processing modes according to different situations, enhancing the system's flexibility and stability in the face of multi-task parallelism and resource fluctuations. This helps improve overall task processing efficiency and reduce the risk of critical event timeouts. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic flowchart of an embodiment of the method for handling time-sensitive events in complex working environments provided in this application; Figure 2 This is a schematic flowchart of a specific embodiment of step S104 in the method for handling time-sensitive events in complex working environments provided in this application; Figure 3 This is a schematic flowchart of a specific embodiment of step S105 in the method for handling time-sensitive events in a complex working environment provided in this application; Figure 4 This is a schematic flowchart of a specific embodiment of step S106 in the method for handling time-sensitive events in complex working environments provided in this application; Figure 5This is a schematic diagram of an embodiment of the device for processing time-sensitive events in a complex working environment provided in this application; Figure 6 This is a schematic diagram of an embodiment of another device for processing time-sensitive events in a complex working environment provided in this application. Detailed Implementation

[0019] The method described in this embodiment of the invention can be executed by a computing device or system with event processing capabilities and resource scheduling functions. The executing entity can be a single server, an edge computing node, an embedded control unit, a distributed platform composed of multiple processing nodes, or an elastic computing resource deployed in a cloud environment. This invention does not limit the hardware type, network architecture, or deployment method of the executing entity; as long as it can execute the steps of the method according to the input event information and system resource status, this invention can be implemented.

[0020] This method can be applied to various complex operational scenarios involving time-sensitive events, such as: In intelligent manufacturing systems, it is used to process real-time scheduling events on high-speed production lines. Used in data centers or cloud platforms to handle sudden, high-priority computing tasks. In edge computing scenarios, it is used to process immediate response requests from sensors or terminal devices; Used in aerospace or traffic control systems to process mission instructions with strict time constraints.

[0021] The above scenarios are only for illustrating the scope of application of the method of the present invention, and are not intended to limit the present invention; any system that needs to judge and schedule events within a limited time can use the method of the present invention.

[0022] To facilitate understanding of the technical solutions of this invention by those skilled in the art, several terms appearing in this application are explained below. This section is for explanation only and should not be construed as limiting the scope of protection of this invention.

[0023] Time-sensitive events refer to information units or processing requests that are triggered in a specific sequence or rhythm during system operation and require a response within a certain period of time.

[0024] Time budget value: refers to the time scale that a system can use to complete related activities around a certain event, inferred from external conditions and the current processing progress.

[0025] Multi-source resource data refers to the collection of status or load information presented by different functional entities, communication paths, or storage areas during system operation.

[0026] Resource capability vector: refers to a set of descriptions abstracted from the internal resource characteristics of the system to characterize the availability of resources, and can be used to represent the overall service potential of resources.

[0027] Time margin: refers to the degree of time flexibility that can still be utilized around a certain activity under the operating conditions of the system, and is used to reflect the time and space that the system can allocate.

[0028] System resistance: refers to a comprehensive description of the factors that may cause process delays when a certain activity is performed within a system, including but not limited to the impact of competitive occupation, environmental load, etc.

[0029] Resource release probability and resource new occupation probability: These refer to the trends of resource status changes in the system that may be mitigated or aggravated in subsequent periods, respectively, and are used to characterize the uncertain fluctuations in resource status.

[0030] Probability mapping function: refers to a type of transformation relationship used to project the influencing factors corresponding to the trend of resource status changes onto a time scale according to specific rules, so that the relevant changes can be quantified and referenced at the time level.

[0031] Predictive correction term: refers to the quantitative adjustment of the system's time state based on resource change trends, used to characterize the time-level deviation that external uncertainties may bring.

[0032] Reachability function: refers to a model expression that jointly evaluates multiple factors related to event execution based on the system's conditions to generate a comprehensive reference value.

[0033] Reachability value: A numerical result output by the reachability function, used to characterize the degree of probability of an event being completed under specific conditions.

[0034] Event handling strategy: refers to the collective term for various response methods that a system can adopt based on its own status and comprehensive evaluation results, and is used to guide the selection of event handling behavior.

[0035] Please see Figure 1 This application first provides an embodiment of a method for handling time-sensitive events in a complex working environment, the embodiment including: S101. Determine the corresponding time budget value based on the deadline of the pre-acquired pending events and the current time. Upon receiving an event to be processed, the system first reads the event's deadline. For example, when the event carries the constraint "must be completed before T_deadline," the system can calculate it by obtaining the current time T_now: Time budget value = T_deadline - T_now In a real-world system, the deadline for an event may originate from the task scheduler, a business service module, or an external trigger source; different events may have different levels of time constraints. Through this step, the system can clearly define the time range for all subsequent processing.

[0036] S102. Perform heterogeneous decoupling on multi-source resource data from execution units, communication links and buffers to obtain resource capability vectors; During system operation, various types of resource status data are continuously generated. However, this data typically originates from data sources with different structures, such as: the real-time processing speed and parallel capabilities of execution units; the bandwidth, packet loss rate, or round-trip latency of communication links; and the capacity utilization and read / write pressure of the buffer. The structures and update frequencies of these raw data are usually inconsistent, and direct use would result in excessive computational complexity. Therefore, this step uses heterogeneous decoupling to convert the aforementioned data of different formats into a parameter set with a unified expression structure. For example, heterogeneous decoupling might include: sampling and filtering the real-time data stream to eliminate transient outliers; mapping resource statuses of different dimensions to standardized coordinates; and combining indicators such as available computing power, available bandwidth, and available buffer capacity into a vector form. The resulting "resource capability vector" can describe the system's current available resource capabilities using a parameter set with a fixed structure, providing a foundation for subsequent calculations.

[0037] The resource capability vector described in this application is not simply a concatenation of resource indicators, but rather a mapping of resource states from different sources and with different structures in the system to a multi-dimensional space with unified interpretability through an internal regularization method.

[0038] For ease of explanation, the resource capability vector is represented by the following formula: R = [r1, r2, r3, ..., rn] Each dimension ri represents a resource capability component after "normalization". This application provides several examples of these dimensions: Schedulable computing power components (example dimension r1), for example: If the system has 12 execution units, but some of them are in a reserved, frequency-limited, or high-level task state, then r1 may only be reflected as "approximately 0.63 schedulable processing potential". This dimension can be obtained by sampling the instantaneous execution rate, backtracking to historical averages, and smoothing the current load trend.

[0039] Therefore, r1 is a comprehensive "effective processing capacity weight" rather than a physical value.

[0040] Communication link connectivity components (example dimension r2), for example: When the link is lightly loaded, r2≈0.8~0.9; When there is random, sudden congestion, r2 may drop to around 0.4; Furthermore, r2 will take into account factors such as recent packet loss rate fluctuations, protocol backoff behavior, and link contention.

[0041] Buffer and intermediate space performance components (example dimension r3), for example: Even if the remaining cache capacity is large, if read / write contention is frequent and access latency is high, r3 may still be too low; If the cache space is not large, but the flow is stable and the hit rate is high, then r3 may be improved.

[0042] Resource volatility elasticity components (example dimension r4), for example: When tasks in the system are already showing a distributed completion trend and resources are about to be automatically released, r4 will be increased accordingly; r4 decreases when the system load is increasing and there is a possibility of overload.

[0043] The execution path continuity component (example dimension r5) is used to characterize the continuous availability of the task execution path (including execution units, access queues, data links, etc.) within the current window.

[0044] For example, some systems internally refactor task execution paths in stages (such as context migration, recycling, and switching), which can affect the continuous progress of tasks; r5 can be used to reflect whether "the current stage is easily interrupted".

[0045] If a system's resource capability vector at a certain moment is: R = [0.72, 0.58, 0.66, 0.81, 0.55] Internally, the system does not directly interpret this as "a certain resource is 0.72". Instead, it uses the resource's capability vector as a whole to characterize the system's overall capability status in terms of "availability, interoperability, capacity, and predictability". S103. Based on the resource capability vector and the current task queue status of the system, calculate the time margin used to characterize the effective processing capacity that the system can provide within the time budget value; In this embodiment, time margin is used to characterize the effective processing time that the system can provide for the event within the time budget. This time margin may include the following elements: The resource capacity vector reflects the scale of scalable resources, the load of tasks currently being executed or pending in the task queue, and the typical processing time model required for task execution (e.g., CPU-bound, IO-bound, etc.).

[0046] The system can estimate the amount of processing that can be completed per unit time based on the resource capacity vector, and then deduct the amount used for other tasks based on the congestion of the task queue, thereby obtaining the actual available effective time and space.

[0047] For example: If the current system has 80% idle processing capacity, and an event takes 50ms under optimal resource conditions, then if the time budget is 100ms, the approximate time margin can be estimated as follows: Scheduled time allowed by available processing capacity ≈ Time budget × Available processing ratio However, in real-world systems, this calculation requires combining task queue arrangement strategies and resource reservation strategies for more accurate estimation.

[0048] S104. Calculate the system resistance used to characterize the adverse impact of the system on the processing of the event to be processed within the time budget value; In this embodiment, system resistance is used to characterize the combined factors that adversely affect the execution of events.

[0049] System resistance here may include, but is not limited to: potential delays caused by high-priority tasks jumping in the task queue, load trends of execution units (such as the inertia of high load lasting for a period of time), and bottlenecks that may occur when communication links or buffers are limited.

[0050] In practice, potential waiting times can be estimated based on the occupancy times of adjacent tasks in the queue; fluctuations in processing efficiency can be estimated based on the current temperature, frequency limiting status, or historical statistics of the execution unit; and the potential blocking probability can be estimated based on the throughput of communication links and buffers. By combining these negative factors, a parameter characterizing future latency risk, namely system resistance, can be obtained.

[0051] See Figure 2 In an optional embodiment, the system resistance can be determined as follows: S1041. Obtain the status of the task queue and the estimated execution time for each task; In this embodiment, the task queue status may include the number of tasks currently awaiting execution in the queue, the task distribution within the system, the dependencies between tasks, and the estimated execution time and associated resource requirements for each task. By obtaining the above information, the system's pending workload structure at the current point in time can be determined.

[0052] S1042. Calculate the potential delay contribution to the pending event based on the number of tasks to be executed, task priority, and resource usage information in the task queue state. In this step, the potential queuing time for each task can be estimated based on factors such as task priority configuration, the degree of competition among tasks on the execution path, and the level of resource consumption, thereby obtaining the potential latency caused by each task for the events to be processed. The potential latency contribution can be determined through methods such as accumulation, weighted accumulation, or trend estimation.

[0053] S1043. Calculate the load data of the execution unit, the communication link, and the buffer within a future time budget window; Load data can include changes in the processing density of execution units, trends in the data exchange rate of communication links, and changes in buffer occupancy. This data is uniformly mapped to a future time budget window to reflect the potential pressure on various resources in the system over a future period.

[0054] S1044. The potential delay contribution and the load data are fused to obtain the system resistance.

[0055] Specifically, the resistance value can be calculated by combining the cumulative amount of potential delay contribution and the concentration of future load data to form a resistance value that characterizes the degree of adverse impact of unprocessed events within the system. System resistance can be expressed as a quantitative index to reflect the overall hindering characteristics of the system in the current and foreseeable timeframe.

[0056] S105. Based on short-term predictions of future resource changes and event arrival conditions, determine the probability of resource release and the probability of new resource occupancy, and establish a probability mapping function for time disturbances. In complex operating environments, resource status is not static but changes dynamically over time. Therefore, in this step, the system will predict resource changes in the near future based on historical and recent sampled data.

[0057] The predictions include the potential release of resources (e.g., existing tasks are about to complete) and the potential addition of resources to the system's schedule (e.g., new tasks arrive or internal system scheduling is triggered). Based on these trends, the system determines the probability of resource release and the probability of new resource occupancy to describe the direction and likelihood of resource status fluctuations within a future window.

[0058] Based on this, a "probability mapping function" is established to convert the probability of resource changes into a "time disturbance" calculation relationship, so that the system can reflect the impact of resource fluctuations on a time scale.

[0059] See Figure 3In an optional embodiment, before establishing the probability mapping function to characterize the amount of time disturbance, the method may further include the following steps to structurally characterize future resource change trends, so that subsequent extrapolation processes can better reflect the impact of short-term disturbances on time budget values: S1051. Perform short-window time series processing on the resource release probability and the resource new occupation probability to obtain a time series expression representing the probability distribution of future changes; In this step, the resource release probability and the probability of new resource occupancy within the short-term prediction interval can be expanded along a time axis and organized into sequence data with continuous or weakly continuous characteristics. The generated time series expression can be regarded as a probability distribution trajectory reflecting the future fluctuation trend of resource status, enabling subsequent steps to extract the fluctuation characteristics in a structured manner.

[0060] S1052. Extract typical distribution patterns of possible resource change states from the time series expression; When analyzing the above time series expressions, common, recurring, or representative resource change patterns can be extracted through pattern aggregation, window analysis, or other forms of trend induction. These typical distribution patterns can reflect different rhythms that resource changes may exhibit, such as periodic release, intermittent occupancy, and short-peak surges, but the specific pattern structure may vary depending on the actual system characteristics.

[0061] S1053. Based on the typical distribution pattern, determine the time budget value range that may be affected by resource changes, and form a time range set containing multiple time ranges. After identifying typical distribution patterns, the system can infer the range of time intervals that may affect the time budget value based on the time-period variation characteristics involved in the pattern. This generates multiple time ranges, which are then categorized or combined according to their potential impact on event handling, forming a set of time ranges encompassing multiple scales and locations.

[0062] S1054. Calculate the disturbance intensity weight corresponding to each time range based on the change factors and resource importance corresponding to the different typical distribution patterns. In this step, different disturbance intensity weights can be assigned to each time range based on the characteristics of typical distribution patterns (such as rate of change, magnitude of change, and uncertainty of change) and the importance of the affected resources in the overall processing path. These weights are used to reflect the actual impact of different resource change events on the time budget.

[0063] S1055. Based on the time range and the disturbance intensity weight, construct a prediction disturbance model; By combining the temporal scope with the corresponding disturbance intensity weights, a predictive disturbance model can be established to describe the structure of disturbance impacts over future time periods. This model can be presented in a hierarchical or segmented manner, allowing the impact of disturbances at different time periods to be distinguished and accumulated, thereby supporting subsequent more granular analysis.

[0064] S1056. The predicted disturbance model is divided into several disturbance units according to a preset time granularity. Each disturbance unit includes a corresponding time range and disturbance intensity weight. To facilitate computation and subsequent mapping, the predicted perturbation model can be decomposed into multiple independent perturbation units. Each perturbation unit can correspond to a refined time slice and include the possible range of influence and related perturbation intensity weights that may be manifested in that time slice, so that the structure of future perturbation effects can be expressed in a more dispersed form.

[0065] S1057. Based on the time range of each disturbance unit and the disturbance intensity weight, calculate the corresponding disturbance impact factor to form a set of quantities used to describe the impact of future resource changes on the time budget value. In this step, based on the influence range and weight information of each disturbance unit, its potential time disturbance effect is quantified to obtain the corresponding disturbance influence factor. These factors, when aggregated, form a quantitative set describing the impact of future resource changes on the time budget value, providing structured input for the subsequent construction of the probability mapping function.

[0066] In an exemplary application scenario, suppose the system is in a task scheduling environment of an industrial production line, where different work units need to respond to multiple processing instructions with strict timeliness within a short time window. At this time, the processing capacity of the execution unit may be affected by factors such as process switching and energy consumption protection strategies; the communication link may experience unstable occupancy due to peak data transmission periods; and the buffer may also experience short-term congestion as a large number of intermediate results are written.

[0067] In this context, the following actions can be taken to address the resource situation in the short term: For example, in S1051, the predicted resource release probability and new occupancy probability within the range of 3 to 5 seconds in the future can be time-seriested and organized into a probability sequence recorded at a fixed time resolution. This sequence can be characterized by: a higher probability of resource release at certain time points (such as when a certain job step is about to be completed), while the probability of new occupancy increases at other time points (such as when a new task is expected to be submitted in a short time).

[0068] For S1052, for example, when observing the above sequence, it may be found that some resource changes exhibit "short-period peak" characteristics, that is, the release and occupancy probabilities frequently alternate between adjacent time periods; it may also be found that some changes exhibit "block clustering" characteristics, that is, the resource occupancy probability remains at a high level at multiple consecutive time points. The system can summarize these characteristics into different typical distribution patterns.

[0069] If a typical distribution pattern indicates a concentrated surge in resource consumption between the 2nd and 3rd second, it can be inferred that this time period may affect the corresponding time budget value range. Therefore, this range can be marked as a time-limited area. If there is a concentrated release probability between the 0.5th and 1st second, it can constitute another time-limited area. Ultimately, this forms a set of time-limited areas containing multiple different durations and locations.

[0070] For S1054, for example in industrial scenarios, changes in the occupancy of execution units are typically more likely to cause actual latency than changes in communication links, so their importance can be set higher. Therefore, under the same distribution pattern, changes from execution units can be assigned a higher disturbance intensity weight, while changes from the buffer may have a lower weight. This creates a set of disturbance intensity weights with different time ranges.

[0071] By combining the aforementioned scope of effect with its corresponding disturbance weights, a disturbance prediction model can be constructed. For example, this model may exhibit a distribution pattern where the disturbance effect is concentrated in certain time periods (such as the processing switching phase), while the impact is weaker in other time periods (such as the stable operation phase).

[0072] Assuming a time granularity of 100 milliseconds, the above predictive perturbation model can be divided into several perturbation units of 100 milliseconds each. Each unit contains its corresponding time range information (such as whether it is in a high perturbation segment or a low perturbation segment) and corresponding perturbation weight data.

[0073] For example, if the perturbation weight is low in the first 0–100 milliseconds, medium in the first 100–200 milliseconds, and significantly higher in the second 200–300 milliseconds, it can be decomposed into three perturbation units.

[0074] For each disturbance unit, the disturbance impact factor can be calculated based on factors such as the weight of its time-bound scope and possible resource change patterns. For example: If a unit is in a period of rapid increase in resource consumption, its disturbance impact factor may be too high; If a unit is located in a region where resource release is concentrated, its disturbance impact factor may be relatively low.

[0075] Based on the set of influence quantities obtained in the foregoing embodiments, this application also provides an embodiment for constructing a probability mapping function. The embodiment is described below and includes: constructing a disturbance accumulation sequence for measuring the cumulative effect of disturbances based on the set of influence quantities; performing nonlinear normalization processing on the disturbance accumulation sequence to obtain a disturbance intensity scale; and establishing a probability mapping function for converting the disturbance influence factor into a time disturbance quantity according to the disturbance intensity scale and the time sensitivity parameters corresponding to different resource types.

[0076] In this embodiment, based on the set of influence quantities obtained in the preceding embodiments, a probability mapping function can be further constructed to map the resource disturbance influence quantity to the time disturbance quantity, thereby providing quantitative input for the calculation of the prediction correction term. In this embodiment, the set of influence quantities is first accumulated in chronological order to form a disturbance accumulation sequence. This sequence is used to measure the cumulative effect of each disturbance influence factor within a future time window, that is, the overall trend of the adverse effects that may be caused to event processing in different time periods.

[0077] For example, assuming the disturbance impact factors within the next 500 milliseconds are [0.2, 0.5, 0.3, 0.7, 0.4], the disturbance accumulation sequence could be [0.2, 0.7, 1.0, 1.7, 2.1]. This accumulation sequence can intuitively reflect the disturbance superposition at different time points, showing that some time periods may have high accumulation disturbances, while other time periods have lower accumulation.

[0078] Nonlinear normalization is applied to the cumulative perturbation sequence to obtain the perturbation intensity scale. This process maps cumulative perturbations of different magnitudes to a uniform scale range, taking into account the sensitivity of high perturbation segments while preserving the effective discriminability of low perturbation segments.

[0079] For the cumulative sequence [0.2, 0.7, 1.0, 1.7, 2.1], a nonlinear function (such as a sigmoid or exponential mapping) can be used to normalize it to the range [0, 1], thereby generating a perturbation intensity scale sequence [0.1, 0.33, 0.44, 0.78, 1.0]. This processing can highlight the potential impact of high-cumulative perturbation segments on the time budget, while suppressing uncertain noise in low-perturbation segments.

[0080] Based on the disturbance intensity scale and the time sensitivity parameters corresponding to different resource types, a probabilistic mapping function is established to convert disturbance impact factors into time disturbance quantities. This function can be understood as a rule or mapping relationship used to transform disturbance impact factors of different sources and importance into quantitative indicators that can be used for time-level calculations.

[0081] Specifically, for the disturbance impact factor of the execution unit, the system can assign a higher time sensitivity parameter, so that the corresponding time disturbance amount is relatively amplified; for the disturbance impact factor of the buffer, a lower sensitivity parameter is assigned according to the actual importance, so that the corresponding time disturbance amount is appropriately adjusted; for the disturbance impact factor of the communication link, the time disturbance amount is calculated based on the current load trend and latency risk.

[0082] In a more specific embodiment, the cumulative perturbation sequence can be calculated using the following formula: ; in, This represents the cumulative perturbation sequence. represents the i-th disturbance influence factor, and k represents the position or index of the currently accumulated disturbance sequence.

[0083] The probability mapping function can be expressed as: ; Where S i To normalize the perturbation intensity scale, γ r ΔT is the time sensitivity parameter corresponding to the resource type. i This represents the corresponding time perturbation.

[0084] S106. Calculate the prediction correction term through the probability mapping function. The prediction correction term is determined based on the resource release probability, the resource new occupation probability, and the corresponding revenue and loss. After obtaining the resource change probability and probability mapping function, the system can use it as input to calculate a predictive correction term for revising the time budget. The predictive correction term can mean that if resources are more likely to be released in the future, it will positively improve event processing time; if resources are more likely to be occupied in the future, it may bring additional delay risks. The predictive correction term can be understood as a time offset after comprehensively assessing future resource change trends. Its calculation process can comprehensively consider the positive or negative impacts of resource release and occupation on time and integrate them appropriately.

[0085] See Figure 4 Based on the foregoing embodiments, this application provides an embodiment for determining the amount of gain and loss. In this embodiment, based on the aforementioned constructed probability mapping function, the disturbance influence factor is processed to generate a quantitative indicator that can be used for time correction, so as to make fine adjustments to the handling of future events. The specific steps are as follows: S1061. Based on the probability mapping function, each disturbance influence factor is mapped to obtain the corresponding time disturbance sequence. In this step, the system transforms each disturbance factor into a corresponding temporal disturbance using a probability mapping function. This temporal disturbance reflects the potential time effect of each disturbance factor on event processing within the future prediction window.

[0086] Assuming the disturbance impact factors are [0.1, 0.3, 0.5, 0.2, 0.4] within the next 500 milliseconds, a time disturbance sequence of [5ms, 12ms, 20ms, 8ms, 16ms] can be obtained after processing with a probability mapping function. This time disturbance sequence is used to characterize the potential delays or deviations in event processing caused by resource disturbances at different future time points.

[0087] S1062. Based on the resource release probability and the resource new occupation probability, calculate the amount of revenue brought by resource release and the amount of loss caused by resource new occupation, respectively. In this step, the system quantifies each resource event based on the predicted probability of resource release and new occupancy. Specifically: The revenue reflects the potential time gain brought about by resource release, such as the schedulability of processing units after release; The loss reflects the time consumption caused by the additional occupation of resources, such as the potential delay caused by new tasks occupying execution units or communication links.

[0088] For example, in the future prediction window, if the release probability of an execution unit is 0.6, it may contribute about 10ms of benefit; if the new occupancy probability is 0.4, it may cause about 8ms of loss.

[0089] S1063. The time disturbance sequence is weighted and fused with the revenue and the loss to obtain a time correction sequence for the future prediction window. In this embodiment, the system combines the time perturbation sequence obtained in step S1061 with the gain and loss amounts obtained in step S1062, and generates a time correction sequence through weighted fusion. The weights can be set according to resource type, perturbation intensity, or task priority, but the specific algorithm is not exposed.

[0090] For example, if the perturbation sequence is [5, 12, 20, 8, 16] ms, the corresponding gain is [2, 3, 1, 4, 2] ms, and the loss is [1, 2, 3, 1, 2] ms, then the time-corrected sequence obtained after weighted fusion may be [6, 13, 18, 11, 16] ms.

[0091] S1064. Perform interval compression on the time-corrected sequence to obtain a single prediction correction term.

[0092] To facilitate subsequent reachability calculations, the system performs interval compression or aggregation on the entire time correction sequence to generate a single prediction correction term, which characterizes the comprehensive impact of overall disturbances on the time budget within the future prediction window. This single prediction correction term can be directly input into the reachability function, supporting the selection of event handling strategies.

[0093] For example, based on the previous example, the time correction sequence [6, 13, 18, 11, 16] ms can be weighted and averaged or otherwise compressed to obtain a prediction correction term of approximately 12.8 ms. This value is used to adjust the event processing priority or select the processing strategy in subsequent steps.

[0094] In a more specific embodiment, the prediction correction term can be calculated using the following formula: ; Among them, T 修正 The prediction correction term represents the time adjustment for overall event processing within the future prediction window. It comprehensively considers the impact of future resource disturbances on processing time, as well as the potential gains from resource release and the losses from new resource occupancy. This prediction correction term can be directly used in reachability function calculations to adjust the probability of event completion within the target processing window.

[0095] This represents the summation of n time periods or disturbance units divided into a future forecast window. Each disturbance unit corresponds to a disturbance impact factor, a gain, and a loss.

[0096] ΔT i B represents the time disturbance. i The term "gain" represents the potential time gain resulting from resource release within the i-th time period. For example, releasing an execution unit can increase available processing time, and releasing a communication link can lead to faster data transmission.

[0097] L i The term "loss amount" represents the time loss caused by increased resource usage or load during the i-th time period. For example, a new task occupies execution units or cache resources, potentially delaying pending events.

[0098] wD, wB, and wL are preset weights, which represent the relative contributions of time disturbance, gain, and loss to the prediction correction term, respectively.

[0099] S107. Using a preset reachability function, calculate the reachability value of the event to be processed based on the time margin, the system resistance, and the prediction correction term, and compare the reachability value with a preset threshold set. In this embodiment, after obtaining the following three factors—time margin, system resistance, and prediction correction term—they are substituted into a preset reachability function, and a quantitative result, namely the "reachability value," is obtained through a prescribed calculation method. The reachability value defined in this application reflects the likelihood of the event being completed on time under current and expected conditions. The system then compares the reachability value with a preset set of thresholds, and the comparison results are as follows: exceeding a specific threshold indicates a high probability of success; falling into the middle range indicates a certain risk; and falling below the minimum threshold indicates a high probability that the time requirement cannot be met. This comparison result directly determines the selection of the next processing strategy.

[0100] In one specific implementation, the method of calculating the reachability value may include: quantifying the time margin to obtain a quantified time margin value used to characterize the remaining available time corresponding to the event to be processed; The system resistance is normalized to obtain a resistance factor that characterizes the difficulty of system execution. Substitute the prediction correction term into the reachability function, and perform a joint operation on the time margin quantification value and the resistance factor to obtain a reachability value that represents the degree to which the event to be processed can be successfully completed within the target processing window.

[0101] In this specific implementation, the time margin is first quantified to obtain a quantified value that characterizes the remaining available time of the event to be processed within the current time window. This quantified value can be adjusted according to the timeliness and urgency of the task, reflecting higher sensitivity in high-urgency tasks and reducing sensitivity to short-term fluctuations in tasks with ample time. The system normalizes the calculated system resistance to obtain a resistance factor, which reflects the execution difficulty and potential delay risk that the system may cause to event processing within the future time window. By substituting the prediction correction term into a preset reachability function and combining the quantified value of the time margin with the resistance factor, the system can obtain a reachability value indicating the degree to which the event to be processed can be successfully completed within the target processing window. This reachability value, as a comprehensive indicator, not only reflects the current remaining time and system load but also implies the impact of future short-term resource disturbances, providing a quantitative basis for the selection of subsequent event processing strategies, thereby achieving intelligent scheduling and dynamic optimization of time-sensitive events in complex operating environments.

[0102] S108. Determine the target event handling strategy from multiple event handling strategies based on the comparison results.

[0103] Based on the comparison results above, the system selects the corresponding target strategy from several processing strategies.

[0104] In an optional embodiment, the preset threshold set includes a first threshold, a second threshold, and a third threshold, and the multiple event handling strategies include at least four options, wherein the four options include a normal handling strategy, a preemptive handling strategy, a fast path handling strategy, and a rejection handling strategy. When the reachability value is greater than or equal to the first threshold, the normal processing strategy is selected to perform event processing according to the standard procedure. When the reachability value is between the first threshold and the second threshold, the preemptive processing strategy is selected; When the reachability value is lower than the second threshold but higher than the third threshold, the fast path processing strategy is selected; When the reachability value is lower than the third threshold, the rejection handling strategy is selected.

[0105] In this optional implementation, a preset threshold set is first established. This preset threshold set may include a first threshold, a second threshold, and a third threshold, used to quantify the reachability level of an event to be successfully completed within the target processing window. The system also defines at least four event handling strategies, including a normal processing strategy, a preemptive processing strategy, a fast path processing strategy, and a rejection processing strategy, to address event scheduling requirements under different reachability conditions. After obtaining the reachability value of the event to be processed, it is compared with the preset threshold set to determine the appropriate processing strategy. Specifically, when the reachability value is greater than or equal to the first threshold, a normal processing strategy is selected, allowing the event to be processed according to the standard procedure, ensuring the stability and accuracy of task completion, and making full use of existing system resources. When the reachability value is between the first and second thresholds, a preemptive processing strategy is selected, adjusting task priorities, temporarily allocating idle resources, or reordering the task queue to accelerate the processing speed of critical events while also considering the execution of other tasks. When the reachability value is lower than the second threshold but higher than the third threshold, a fast path processing strategy is selected, skipping some non-critical processing modules, using pre-set fast execution paths, or reducing some processing precision to significantly shorten processing latency and cope with urgent time constraints. When the reachability value is lower than the third threshold, a rejection processing strategy is implemented, downgrading or temporarily suspending the event to avoid consuming too many system resources and affecting the completion of higher-priority tasks.

[0106] The foregoing embodiments have described in detail the method provided in this application. The following describes embodiments of the apparatus provided in this application.

[0107] See Figure 5 This application provides an embodiment of a processing apparatus for time-sensitive events in a complex working environment, the embodiment comprising: The time budget determination unit 501 is used to determine the corresponding time budget value based on the deadline of the pre-acquired pending event and the current time. Decoupling unit 502 is used to perform heterogeneous decoupling of multi-source resource data from execution unit, communication link and buffer to obtain resource capability vector; The time margin calculation unit 503 is used to calculate, based on the resource capacity vector and the current task queue status of the system, a time margin that characterizes the effective processing capacity that the system can provide within the time budget value. The system resistance calculation unit 504 is used to calculate the system resistance that characterizes the adverse impact of the system on the processing of the event to be processed within the time budget value. Function building unit 505 is used to determine the probability of resource release and the probability of new resource occupancy based on short-term predictions of future resource changes and event arrival, and to establish a probability mapping function for time disturbances. The function calculation unit 506 is used to calculate the prediction correction term through the probability mapping function. The prediction correction term is determined based on the resource release probability, the resource new occupation probability, and the corresponding revenue and loss. The comparison unit 507 is used to calculate the reachability value of the event to be processed based on the time margin, the system resistance and the prediction correction term using a preset reachability function, and compare the reachability value with a preset threshold set. The strategy determination unit 508 is used to determine the target processing strategy from multiple event processing strategies based on the comparison results.

[0108] Optionally, it also includes a perturbation model building unit 509, used for: The probability of resource release and the probability of new resource occupancy are serialized using a short window to obtain a time series expression representing the probability distribution of future changes; Typical distribution patterns of possible resource change states are extracted from the time series expression; Based on the typical distribution pattern, the time budget value range that may be affected by resource changes is determined, and a set of time scopes containing multiple time scopes is formed. Based on the change factors and resource importance corresponding to different typical distribution patterns, calculate the disturbance intensity weight corresponding to each time range. Based on the time range and the disturbance intensity weight, a prediction disturbance model is constructed; The predicted perturbation model is divided into several perturbation units according to a preset time granularity. Each perturbation unit includes a corresponding time range and perturbation intensity weight. Based on the time range and disturbance intensity weight of each disturbance unit, the corresponding disturbance impact factor is calculated to form a set of quantities used to describe the impact of future resource changes on the time budget value.

[0109] Optionally, function building unit 505 is specifically used for: Based on the set of influence quantities, a perturbation accumulation sequence is constructed to measure the cumulative effect of perturbation; The perturbation accumulation sequence is subjected to nonlinear normalization to obtain the perturbation intensity scale; Based on the disturbance intensity scale and the time sensitivity parameters corresponding to different resource types, a probability mapping function is established to convert the disturbance influence factor into a time disturbance quantity.

[0110] Optionally, the function computation unit 506 is specifically used for: Based on the probability mapping function, each disturbance influence factor is mapped to obtain the corresponding time disturbance sequence. Based on the resource release probability and the resource new occupation probability, calculate the amount of revenue from resource release and the amount of loss caused by resource new occupation, respectively. The time perturbation sequence is weighted and fused with the gain and loss to obtain a time correction sequence for the future prediction window; The time-corrected sequence is subjected to interval compression to obtain a single prediction correction term.

[0111] Optionally, the comparison unit 507 is specifically used for: The time margin is quantified to obtain a quantified value of the time margin used to characterize the remaining available time corresponding to the event to be processed. The system resistance is normalized to obtain a resistance factor that characterizes the difficulty of system execution. Substitute the prediction correction term into the reachability function, and perform a joint operation on the time margin quantification value and the resistance factor to obtain a reachability value that represents the degree to which the event to be processed can be successfully completed within the target processing window.

[0112] Optionally, the strategy determination unit 508 is specifically used for: The preset threshold set includes a first threshold, a second threshold, and a third threshold. The multiple event handling strategies include at least four options, including a normal handling strategy, a preemptive handling strategy, a fast path handling strategy, and a rejection handling strategy. When the reachability value is greater than or equal to the first threshold, the normal processing strategy is selected to perform event processing according to the standard procedure. When the reachability value is between the first threshold and the second threshold, the preemptive processing strategy is selected; When the reachability value is lower than the second threshold but higher than the third threshold, the fast path processing strategy is selected; When the reachability value is lower than the third threshold, the rejection handling strategy is selected.

[0113] Optionally, the system resistance calculation unit 504 is specifically used for: Obtain the status of the task queue and the estimated execution time for each task; Based on the number of tasks to be executed in the task queue, task priority, and resource usage information, calculate the potential delay contribution to the event to be processed. Calculate the load data of the execution unit, the communication link, and the buffer within a future time budget window; The system resistance is obtained by fusing the potential delay contribution and the load data.

[0114] Please see Figure 6 This application also provides a processing apparatus for time-sensitive events in complex working environments, comprising: Processor 601, memory 602, input / output unit 603, bus 604; The processor 601 is connected to the memory 602, the input / output unit 603, and the bus 604; The memory 602 stores a program, and the processor 601 calls the program to execute any of the methods described above.

[0115] This application also relates to a computer-readable storage medium on which a program is stored, which, when run on a computer, causes the computer to perform any of the methods described above.

[0116] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0117] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection between apparatuses or units through some interfaces, and may be electrical, mechanical, or other forms.

[0118] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0119] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0120] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A method for handling time-sensitive events in complex work environments, characterized in that, The method includes: The corresponding time budget value is determined based on the deadline of the pre-acquired pending events and the current time. Heterogeneous decoupling is performed on multi-source resource data from execution units, communication links, and buffers to obtain resource capability vectors; Based on the resource capability vector and the current task queue status of the system, calculate the time margin used to characterize the effective processing capacity that the system can provide within the time budget value; Calculate the system resistance used to characterize the adverse impact of the system on the processing of the event to be processed within the time budget value; Based on short-term predictions of future resource changes and event arrival conditions, the probability of resource release and the probability of new resource occupancy are determined, and a probability mapping function for time disturbances is established. The prediction correction term is calculated using the probability mapping function, and the prediction correction term is determined based on the resource release probability, the resource new occupation probability, and the corresponding revenue and loss. The reachability value of the event to be processed is calculated based on the time margin, the system resistance, and the prediction correction term using a preset reachability function, and the reachability value is compared with a preset threshold set. The target event handling strategy is determined from a variety of event handling strategies based on the comparison results.

2. The method for handling time-sensitive events in complex work environments according to claim 1, characterized in that, Before establishing the probability mapping function for the time perturbation, the method further includes: The probability of resource release and the probability of new resource occupancy are serialized using a short window to obtain a time series expression representing the probability distribution of future changes; Typical distribution patterns of possible resource change states are extracted from the time series expression; Based on the typical distribution pattern, the time budget value range that may be affected by resource changes is determined, and a set of time scopes containing multiple time scopes is formed. Based on the change factors and resource importance corresponding to different typical distribution patterns, calculate the disturbance intensity weight corresponding to each time range. Based on the time range and the disturbance intensity weight, a prediction disturbance model is constructed; The predicted perturbation model is divided into several perturbation units according to a preset time granularity. Each perturbation unit includes a corresponding time range and perturbation intensity weight. Based on the time range and disturbance intensity weight of each disturbance unit, the corresponding disturbance impact factor is calculated to form a set of quantities used to describe the impact of future resource changes on the time budget value.

3. The method for handling time-sensitive events in complex work environments according to claim 2, characterized in that, The establishment of the probability mapping function for the time perturbation includes: Based on the set of influence quantities, a perturbation accumulation sequence is constructed to measure the cumulative effect of perturbation; The perturbation accumulation sequence is subjected to nonlinear normalization to obtain the perturbation intensity scale; Based on the disturbance intensity scale and the time sensitivity parameters corresponding to different resource types, a probability mapping function is established to convert the disturbance influence factor into a time disturbance quantity.

4. The method for handling time-sensitive events in complex work environments according to claim 2, characterized in that, The step of calculating the prediction correction term through the probability mapping function, wherein the prediction correction term is determined based on the resource release probability, the resource new occupation probability, and the corresponding revenue and loss, includes: Based on the probability mapping function, each disturbance influence factor is mapped to obtain the corresponding time disturbance sequence. Based on the resource release probability and the resource new occupation probability, calculate the amount of revenue from resource release and the amount of loss caused by resource new occupation, respectively. The time perturbation sequence is weighted and fused with the gain and loss to obtain a time correction sequence for the future prediction window; The time-corrected sequence is subjected to interval compression to obtain a single prediction correction term.

5. The method for handling time-sensitive events in complex work environments according to claim 1, characterized in that, The step of calculating the reachability value of the event to be processed based on the time margin, the system resistance, and the prediction correction term using a preset reachability function includes: The time margin is quantified to obtain a quantified value of the time margin used to characterize the remaining available time corresponding to the event to be processed. The system resistance is normalized to obtain a resistance factor that characterizes the difficulty of system execution. Substitute the prediction correction term into the reachability function, and perform a joint operation on the time margin quantification value and the resistance factor to obtain a reachability value that represents the degree to which the event to be processed can be successfully completed within the target processing window.

6. The method for handling time-sensitive events in complex work environments according to claim 5, characterized in that, The step of determining the target handling strategy from multiple event handling strategies based on the comparison results includes: The preset threshold set includes a first threshold, a second threshold, and a third threshold. The multiple event handling strategies include at least four options, including a normal handling strategy, a preemptive handling strategy, a fast path handling strategy, and a rejection handling strategy. When the reachability value is greater than or equal to the first threshold, the normal processing strategy is selected to perform event processing according to the standard procedure. When the reachability value is between the first threshold and the second threshold, the preemptive processing strategy is selected; When the reachability value is lower than the second threshold but higher than the third threshold, the fast path processing strategy is selected; When the reachability value is lower than the third threshold, the rejection handling strategy is selected.

7. The method for processing time-sensitive events in a complex work environment according to any one of claims 1-6, characterized in that, The calculations used to characterize the system resistance that adversely affects the processing of the event to be processed within the time budget include: Obtain the status of the task queue and the estimated execution time for each task; Based on the number of tasks to be executed in the task queue, task priority, and resource usage information, calculate the potential delay contribution to the event to be processed. Calculate the load data of the execution unit, the communication link, and the buffer within a future time budget window; The system resistance is obtained by fusing the potential delay contribution and the load data.

8. A device for processing time-sensitive events in complex working environments, characterized in that, The device includes: The time budget determination unit is used to determine the corresponding time budget value based on the deadline of the pre-acquired pending events and the current time. The decoupling unit is used to heterogeneously decouple multi-source resource data from the execution unit, communication link and buffer to obtain resource capability vectors; The time margin calculation unit is used to calculate, based on the resource capacity vector and the current task queue status of the system, a time margin that characterizes the effective processing capacity that the system can provide within the time budget value. A system resistance calculation unit is used to calculate the system resistance that characterizes the adverse impact of the system on the processing of the event to be processed within the time budget value. The function building unit is used to determine the probability of resource release and the probability of new resource occupancy based on short-term predictions of future resource changes and event arrival conditions, and to establish a probability mapping function for time disturbances. The function calculation unit is used to calculate the prediction correction term through the probability mapping function. The prediction correction term is determined based on the resource release probability, the resource new occupation probability, and the corresponding revenue and loss. The comparison unit is used to calculate the reachability value of the event to be processed based on the time margin, the system resistance and the prediction correction term using a preset reachability function, and compare the reachability value with a preset threshold set. The strategy determination unit is used to determine the target processing strategy from multiple event processing strategies based on the comparison results.

9. A device for processing time-sensitive events in complex working environments, characterized in that, The device includes: Processor, memory, input / output units, and bus; The processor is connected to the memory, the input / output unit, and the bus; The memory stores a program, which the processor invokes to perform the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium contains a program that, when executed on a computer, performs the method as described in any one of claims 1 to 7.

Images