CN-122018365-A - Closed loop monitoring and control method, device, equipment and medium based on target drive

CN122018365ACN 122018365 ACN122018365 ACN 122018365ACN-122018365-A

Abstract

The application discloses a closed-loop monitoring and control method, device, equipment and medium based on target driving. The method comprises the steps of obtaining a system observation vector and a prediction vector, obtaining overall deviation based on deviation of the observation vector or the prediction vector relative to a monitoring target, decomposing the overall deviation based on a system context state to obtain controllable deviation, issuing an action strategy and an effect specification based on the context state and the controllable deviation, executing the action strategy, performing process monitoring and effect monitoring based on the effect specification, and driving maintenance or adjustment of the strategy based on a monitoring result until the controllable deviation is smaller than a preset threshold. The application provides a closed-loop monitoring and controlling method of target drive, which forms a closed loop of 'target-perception-evaluation-decision-execution monitoring-regeneration-mark-remaining', and realizes continuous monitoring and active intervention of a system.

Inventors

GUO SHENGMIN
XIA SHUDONG
Han xingguang
LIU YANGYANG
Song Chongxian
ZHANG RUILONG
ZHAO JUNWU
LI YUNCAI
LI CHENGBAO
ZHANG XIN

Assignees

北京掌行通信息技术有限公司

Dates

Publication Date: 20260512
Application Date: 20251226

Claims (11)

1. A target drive-based closed loop monitoring and control method, comprising: S101, determining a monitoring target, acquiring a system observation vector and a prediction vector, and obtaining overall deviation based on the deviation of the observation vector or the prediction vector relative to the monitoring target; S102, decomposing the overall deviation based on a system context state to obtain a controllable deviation; S103, issuing an action strategy and an effect specification based on the context state and the controllable deviation; S104, executing the action strategy, performing process monitoring based on the action strategy, performing effect monitoring based on the effect specification, and driving maintenance or adjustment of the strategy based on the monitoring result until the controllable deviation is smaller than a preset threshold.
2. The method of claim 1, further comprising, prior to obtaining the system observation vector and the prediction vector: Defining a target interval and an index weight for each monitoring index; and setting an entry threshold, an exit threshold and a time condition for each risk level, wherein the entry threshold is larger than the exit threshold, and the risk levels are classified according to the risk values.
3. The method of claim 2, wherein obtaining a system observation vector and a prediction vector, and deriving an overall deviation based on a deviation of the observation vector or the prediction vector from a monitored target, comprises: Collecting the current observation vector of the system, inputting the observation vector, the historical data and the localization parameters into a pre-training prediction model to obtain a prediction vector representing the future state of the system; for each monitoring index, calculating the deviation of the observation vector or the prediction vector relative to the monitoring target to obtain the independent deviation of each monitoring index; And synthesizing the independent deviation of each monitoring index into the total deviation based on the index weight.
4. The method of claim 1, wherein decomposing the overall bias based on system context state to obtain a controllable bias comprises: uniformly encoding the current environment information and the data caliber information of the system to obtain the context state; Decomposing the overall deviation based on the context state to obtain a controllable deviation and a reasonable deviation; And mapping the controllable deviation into a risk value to obtain an effective risk value.
5. The method of claim 1, wherein issuing an action policy and an effect specification based on the context state, controllable deviation, comprises: Dynamically determining a set of executable action policies based on current context state constraints; selecting an optimal action from the action strategy set based on the maximization of the controllable deviation reduction as a preferential criterion; Generating an effect specification corresponding to the optimal action based on the optimal action and the controllable deviation, wherein the effect specification comprises an expected effect track and an actionation target interval corresponding to the optimal action; And releasing the optimal action and the effect specification.
6. The method of claim 1, wherein executing the action strategy, performing process monitoring based on the action strategy, performing effect monitoring based on the effect specification, driving maintenance or adjustment of strategy based on monitoring results, until the controllable deviation is less than a preset threshold, comprises: Calculating a process compliance based on a time compliance, a work ready gap value, a progress gap value during policy execution; obtaining an effect compliance degree based on the deviation of the effect specification and the observation vector, and obtaining a combination criterion based on the process compliance degree and the effect compliance degree; Triggering a decision process after the combination criterion is greater than or equal to a preset entry threshold and a time condition is met, and re-optimizing by a policy engine; And after the combination criterion is smaller than or equal to a preset exit threshold and the time condition is met, entering the next time slice and continuing to monitor until the action strategy is executed.
7. The method of claim 6, wherein the re-preferential by the policy engine comprises: acquiring a context state and controllable deviation at the current moment, and determining a new action strategy and an effect specification based on the fact that the controllable deviation is reduced to the maximum degree to be a preferential criterion; and S104, executing the action strategy until the execution of the action strategy is completed.
8. The method of claim 1, wherein maintenance or adjustment of the strategy is driven based on the monitoring results; Collecting monitoring data during execution of an action strategy, wherein the monitoring data comprises context data, process compliance data and effect compliance data; determining the action policy execution effect based on the process compliance data and effect compliance data; when the action strategy execution effect does not reach the preset effect specification, adjusting the monitoring target according to the context data to obtain an adjusted monitoring target; and S101-S104 steps are executed according to the adjusted monitoring target traversal until the controllable deviation is smaller than a preset threshold value.
9. A target drive-based closed loop monitoring and control device, comprising: the difference evaluation module is used for determining a monitoring target, acquiring a system observation vector and a prediction vector, and obtaining overall deviation based on the deviation of the observation vector or the prediction vector relative to the monitoring target; The context attribution module is used for decomposing the overall deviation based on the system context state to obtain a controllable deviation; The strategy release module is used for releasing action strategies and effect specifications based on the context state and the controllable deviation; And the execution closed loop module is used for executing the action strategy, executing process monitoring based on the action strategy, executing effect monitoring based on the effect specification, and driving the maintenance or adjustment of the strategy based on the monitoring result until the controllable deviation is smaller than a preset threshold value.
10. An electronic device comprising a processor and a memory storing program instructions, the processor being configured, when executing the program instructions, to perform the target drive based closed loop monitoring and control method of any one of claims 1 to 8.
11. A computer readable medium having stored thereon computer readable instructions executable by a processor to implement a target drive based closed loop monitoring and control method according to any of claims 1 to 8.

Description

Closed loop monitoring and control method, device, equipment and medium based on target drive Technical Field The application relates to the technical field of intelligent monitoring and control, in particular to a closed-loop monitoring and control method, device, equipment and medium based on target driving. Background With the penetration of digital transformation, the requirements of key fields such as intelligent transportation, data center operation and maintenance, industrial process control and the like on the running stability, efficiency and compliance of the system are remarkably improved. The system runs as a normal state with high concurrency, long period and cross scenes, continuously generates massive heterogeneous data, and if any state deviation cannot be timely and accurately monitored and interfered, the state deviation can be amplified as service interruption, performance degradation or compliance risk. At present, a monitoring paradigm commonly adopted in the industry is mainly based on data visualization and static threshold alarming, historical average or experience threshold is compared with real-time data, and if the historical average or experience threshold exceeds the threshold, the monitoring paradigm alarms, and attendees are manually treated through SOP (Standard Operating Procedure). The existing monitoring system only carries out early warning based on a static threshold value, is difficult to adaptively generate action strategies for control, and is difficult to adjust targets based on real-time running conditions of the system. Disclosure of Invention The embodiment of the application provides a closed-loop monitoring and control method, device, equipment and medium based on target drive, which at least solve the technical problem that the closed-loop monitoring and control of a system based on a target is difficult in the related technology. According to an aspect of the embodiment of the present application, there is provided a closed-loop monitoring and control method based on target driving, including: determining a monitoring target, acquiring a system observation vector and a prediction vector, and obtaining overall deviation based on the deviation of the observation vector or the prediction vector relative to the monitoring target; decomposing the overall deviation based on the system context state to obtain a controllable deviation; issuing an action strategy and an effect specification based on the context state and the effective risk value; And executing the action strategy, performing process monitoring based on the action strategy, performing effect monitoring based on the effect specification, and driving maintenance or adjustment of the strategy based on a monitoring result until the controllable deviation is smaller than a preset threshold. According to another aspect of the embodiment of the present application, there is also provided a closed-loop monitoring and control device based on target driving, including: the difference evaluation module is used for determining a monitoring target, acquiring a system observation vector and a prediction vector, and obtaining overall deviation based on the deviation of the observation vector or the prediction vector relative to the monitoring target; The context attribution module is used for decomposing the overall deviation based on the system context state to obtain a controllable deviation; the policy issuing module is used for issuing an action policy and an effect specification based on the context state and the controllable deviation; And the execution closed loop module is used for executing the action strategy, executing process monitoring based on the action strategy, executing effect monitoring based on the effect specification, and driving the maintenance or adjustment of the strategy based on the monitoring result until the controllable deviation is smaller than a preset threshold value. According to yet another aspect of the embodiments of the present application, there is also provided an electronic device including a memory, and a processor, wherein the memory stores a computer program, and the processor is configured to execute the target drive-based closed-loop monitoring and control method by using the computer program. According to yet another aspect of the embodiments of the present application, there is also provided a computer readable storage medium having a computer program stored therein, wherein the computer program is configured to perform the above-described target drive based closed loop monitoring and control method when run. The technical scheme provided by the embodiment of the application can have the following beneficial effects: The embodiment of the application provides a closed-loop monitoring and control method based on target driving, which constructs a closed-loop control system integrating target management, perception estimation, gap estimation, context processing, strategy d