CN-121977752-A - Liquid cooling system leakage detection method, device, equipment and storage medium

CN121977752ACN 121977752 ACN121977752 ACN 121977752ACN-121977752-A

Abstract

The invention discloses a detection method, a device, equipment and a storage medium for liquid leakage of a liquid cooling system, wherein a base line network for representing a normal coupling relation of the liquid cooling loop is constructed based on multi-node operation data by collecting the multi-node operation data reflecting the fluid transportation state; when the local area of the base line network is abnormal in structure, the corresponding detected area is determined, controlled perturbation input is applied to the detected area under the condition that the thermal safety constraint of the energy storage battery system is met, so that the liquid cooling loop forms a recognizable dynamic response process, fluid response data of the detected area are collected in the dynamic response process, dynamic characteristics are extracted, difference measurement is carried out between the dynamic characteristics and the base line characteristics, and a leakage risk detection result is output. According to the invention, effective identification of the liquid leakage risk of the liquid cooling system can be realized under complex operation conditions, and the detection reliability and early warning capability are improved.

Inventors

LI KONGZHENG
WANG XIAOMING
Jia Wenhang

Assignees

洛阳理工学院
广东百德朗科技有限公司

Dates

Publication Date: 20260505
Application Date: 20260401

Claims (10)

1. The liquid cooling system leakage detection method is characterized by comprising the following steps of: in the operation process of the liquid cooling loop, collecting multi-node operation data reflecting the fluid transportation state of the liquid cooling loop, and constructing a base line network representing the normal coupling relation of the liquid cooling loop based on the multi-node operation data, wherein the multi-node operation data at least comprises two types of pressure data, flow data and temperature data; When structural abnormality occurs in a local area of the base line network, determining a detected area corresponding to the local area, and applying preset controlled perturbation input to the detected area under the condition of meeting the thermal safety constraint of a lithium battery, so that a liquid cooling loop forms an identifiable dynamic response process, and synchronously collecting at least one fluid response data of the detected area; And extracting dynamic characteristics representing equivalent fluid characteristics of the liquid cooling loop from the fluid response data, carrying out difference measurement on the dynamic characteristics and baseline characteristics corresponding to the base line network, and outputting a leakage risk detection result.
2. The method for detecting liquid leakage of a liquid cooling system according to claim 1, wherein during the operation of the liquid cooling circuit, collecting multi-node operation data reflecting the fluid transportation state of the liquid cooling circuit, and constructing a base line network representing the normal coupling relationship of the liquid cooling circuit based on the multi-node operation data, specifically comprising: Determining a plurality of nodes in a plurality of space areas of the liquid cooling loop, wherein the plurality of space areas at least comprise a cooling area corresponding to the lithium battery module, a manifold area corresponding to the liquid return collecting structure and a pump valve area corresponding to the liquid cooling loop power component; in a sliding time window, collecting at least two types of operation parameters with different physical meanings as node operation data aiming at each node, and performing time alignment, outlier removal and short-time fluctuation suppression processing on the node operation data; determining an effective data segment according to the operation condition information in the sliding time window, wherein the effective data segment meets a preset condition consistency condition, and the condition consistency condition at least comprises that a liquid cooling loop is in a stable circulation state and no preset start-stop transient state or valve switching transient state exists; and constructing a node state vector based on the node operation data in the effective data segment.
3. The method for detecting liquid leakage of a liquid cooling system according to claim 2, wherein during the operation of the liquid cooling circuit, collecting multi-node operation data reflecting the fluid transportation state of the liquid cooling circuit, and constructing a base line network representing the normal coupling relationship of the liquid cooling circuit based on the multi-node operation data, specifically comprising: calculating the association strength between any two nodes based on the node state vector of each node in the effective data segment, wherein the association strength comprises a correlation coefficient, mutual information or a combination of the two; Determining a statistic set for all associated intensities within the valid data segment, the statistic set including at least a mean and a dispersion of associated intensities; Determining an uncertainty amount based on the node state vector, the uncertainty amount being used to characterize a fluctuation or uncertainty of the node state vector within the valid data segment; Constructing an adaptive criterion threshold value with the statistic set and the uncertainty amount, wherein the adaptive criterion threshold value at least comprises a basic threshold value item formed by the average value and the dispersion of the association strength, and a modulation item for modulating the basic threshold value item by the uncertainty amount; and determining the node pair meeting the self-adaptive criterion threshold as a network side of a base line network, or normalizing and weighting the association strength according to the self-adaptive criterion threshold to be used as the side weight of the base line network so as to form the base line network.
4. The method for detecting leakage of a liquid cooling system according to claim 1, wherein when the structural abnormality occurs in a local area of the base line network, determining a detected area corresponding to the local area specifically includes: Acquiring network structure parameters of the base line network in the sliding time window, and calculating regional risk metrics of each spatial region according to the network structure parameters, wherein the network structure parameters comprise at least one of the number change of node association edges, the change amplitude of association edge weights or the increment of node uncertainty; Comparing the regional risk metric of each spatial region with the corresponding regional baseline value, and determining that the spatial region is a detected region when the regional risk metric of at least one spatial region meets the preset trigger condition.
5. The method for detecting liquid leakage of a liquid cooling system according to claim 4, wherein a preset controlled perturbation input is applied to the detected region under the condition of meeting the thermal safety constraint of a lithium battery, so that the liquid cooling loop forms an identifiable dynamic response process, and the method specifically comprises the following steps: Acquiring operation condition information of a liquid cooling loop, and determining disturbance amplitude and disturbance duration of controlled perturbation input based on the operation condition information, so that the controlled perturbation input meets a preset thermal safety constraint condition during disturbance, wherein the thermal safety constraint condition at least comprises at least one of target temperature range constraint or temperature rise rate constraint of a lithium battery system; And maintaining at least one boundary condition of the liquid cooling loop constant or in a preset steady state during the disturbance applied to the controlled perturbation input, wherein the controlled perturbation input is a preset disturbance sequence applied to at least one operating parameter of the liquid cooling loop.
6. The method for detecting leakage of a liquid cooling system according to claim 1, wherein the synchronously collecting at least one fluid response data of the detected area specifically comprises: And synchronously observing the detected area according to a sampling period corresponding to the controlled perturbation input in a preset observation period after the disturbance of the controlled perturbation input is applied, and collecting fluid response data related to a fluid transportation state, wherein the fluid response data comprises at least one of pressure-related response data, flow-related response data and temperature difference-related response data, and the fluid response data is aligned with a disturbance sequence of the controlled perturbation input in time.
7. The method for detecting leakage of a liquid cooling system according to claim 6, wherein extracting a dynamic characteristic characterizing equivalent fluid characteristics of a liquid cooling circuit from the fluid response data, and performing a difference measurement on the dynamic characteristic and a baseline characteristic corresponding to the baseline network, specifically comprises: Based on the fluid response data and the disturbance sequence input by the controlled perturbation, extracting at least one type of characteristic quantity reflecting dynamic response speed, at least one type of characteristic quantity reflecting steady state change and at least one type of characteristic quantity reflecting dynamic hysteresis relation as dynamic characteristics; And carrying out normalization processing on the dynamic characteristics, and calculating a difference metric value of the dynamic characteristics relative to the baseline characteristics, wherein the difference metric value is a weighted difference value, a distance metric value or a combination of the two.
8. The utility model provides a detection device of liquid cooling system weeping which characterized in that includes: The data acquisition module is used for acquiring multi-node operation data reflecting the fluid transportation state of the liquid cooling loop in the operation process of the liquid cooling loop and constructing a base line network representing the normal coupling relation of the liquid cooling loop based on the multi-node operation data, wherein the multi-node operation data at least comprises two types of pressure data, flow data and temperature data; The controlled perturbation module is used for determining a detected area corresponding to a local area when the structural abnormality occurs in the local area of the base line network, applying preset controlled perturbation input to the detected area under the condition of meeting the thermal safety constraint of the lithium battery, enabling a liquid cooling loop to form an identifiable dynamic response process, and synchronously collecting at least one fluid response data of the detected area; And the detection result output module is used for extracting dynamic characteristics representing equivalent fluid characteristics of the liquid cooling loop from the fluid response data, carrying out difference measurement on the dynamic characteristics and baseline characteristics corresponding to the baseline network, and outputting a leakage risk detection result.
9. The liquid cooling system leakage detection device is characterized by comprising a memory and at least one processor, wherein the memory stores instructions; The at least one processor invokes the instructions in the memory to cause the liquid cooling system leakage detection apparatus to perform the steps of the liquid cooling system leakage detection method according to any one of claims 1 to 7.
10. A computer-readable storage medium having instructions stored thereon, wherein the instructions, when executed by a processor, implement the steps of the liquid cooling system leakage detection method according to any one of claims 1 to 7.

Description

Liquid cooling system leakage detection method, device, equipment and storage medium Technical Field The invention relates to the technical field of liquid cooling system leakage detection, in particular to a liquid cooling system leakage detection method, device and equipment and a storage medium. Background The liquid cooling system is a cooling mode which takes heat generated in the running process of equipment away from a heating component and transfers the heat to an external heat radiating unit through liquid serving as a heat transfer medium. Compared with the traditional air cooling mode, the liquid cooling system has the advantages of high heat transfer efficiency, uniform temperature control, low noise and the like, and is widely applied to products with high power density and high energy integration level, such as data center servers, power electronic equipment, lithium battery systems and the like. In lithium battery systems, the liquid cooling system generally comprises components such as a cold plate, a pipeline, a manifold, a pump body, a valve element, a heat exchanger and the like, and is used for continuously radiating heat of a battery module or a battery core so as to ensure that the battery runs in a proper temperature range, thereby improving the safety, the service life and the performance stability of the battery. However, the liquid cooling system generally belongs to a closed circulation structure, and the inside of the liquid cooling system comprises a plurality of connection parts and a sealing structure, and the risk of leakage of cooling liquid exists under the action of long-term operation, vibration impact or thermal circulation. For lithium battery systems, once the liquid cooling system leaks, the cooling capacity is reduced, the local temperature is abnormal, and the cooling liquid can enter the battery pack to be in contact with electrical components or high-voltage connection structures, so that the insulation performance is reduced, corrosion damage is caused, and even potential safety hazards are caused. In addition, the liquid leakage of the liquid cooling system is often hidden in the early stage, is difficult to find in time, and if the liquid leakage cannot be effectively detected and judged in the early stage of the liquid leakage risk, the fault is possibly enlarged or the system is stopped passively. Therefore, it is necessary to perform liquid leakage detection on the liquid cooling system so as to effectively monitor the running state of the liquid cooling system and early warn the risk. Disclosure of Invention The embodiment of the application provides a detection method for liquid leakage of a liquid cooling system, which aims to solve the technical problems that the existing liquid cooling system is strong in liquid leakage concealment, easy to be interfered by working condition changes and difficult to identify liquid leakage risk timely and reliably in the operation process. The invention is realized in such a way that a liquid leakage detection method of a liquid cooling system comprises the following steps: in the operation process of the liquid cooling loop, collecting multi-node operation data reflecting the fluid transportation state of the liquid cooling loop, and constructing a base line network representing the normal coupling relation of the liquid cooling loop based on the multi-node operation data, wherein the multi-node operation data at least comprises two types of pressure data, flow data and temperature data; When structural abnormality occurs in a local area of the base line network, determining a detected area corresponding to the local area, and applying preset controlled perturbation input to the detected area under the condition of meeting the thermal safety constraint of a lithium battery, so that a liquid cooling loop forms an identifiable dynamic response process, and synchronously collecting at least one fluid response data of the detected area; And extracting dynamic characteristics representing equivalent fluid characteristics of the liquid cooling loop from the fluid response data, carrying out difference measurement on the dynamic characteristics and baseline characteristics corresponding to the base line network, and outputting a leakage risk detection result. The invention also provides a device for detecting liquid leakage of the liquid cooling system, which comprises: The data acquisition module is used for acquiring multi-node operation data reflecting the fluid transportation state of the liquid cooling loop in the operation process of the liquid cooling loop and constructing a base line network representing the normal coupling relation of the liquid cooling loop based on the multi-node operation data, wherein the multi-node operation data at least comprises two types of pressure data, flow data and temperature data; The controlled perturbation module is used for determining a detected area corresponding to a local area w