A method for monitoring thermal runaway protection of a charging pile

Abstract

The application relates to the technical field of charging pile protection, and discloses a monitoring method for thermal runaway protection of a charging pile. The method obtains a set of operation parameters through a charging pile monitoring platform, collects temperatures of temperature distribution centers to obtain center temperature values, collects temperature change characteristics and environmental humidity characteristics of each temperature distribution area, uses the center temperature values to learn thermal runaway response characteristics of the characteristics, forms a static thermal runaway response array, obtains an operable path of a protection device, combines each response value in the static thermal runaway response array and a corresponding area, controls an operation track of cooling intervention of the protection device, collects real-time environmental parameters of a current area to determine a dynamic thermal runaway response value during the cooling intervention, and corrects an intervention direction of the protection device according to the dynamic and static response values. The method can improve the accuracy of thermal runaway risk identification and the effectiveness of protection intervention of the charging pile, and guarantees the operation safety of the charging pile.

Description

Technical Field

[0001] This invention relates to the field of charging pile protection technology, specifically a monitoring method for thermal runaway protection of charging piles. Background Technology

[0002] With the rapid development of the new energy vehicle industry, the operational safety of charging piles, as a key infrastructure for energy replenishment, has received widespread attention. During long-term operation, charging piles are prone to localized abnormal temperature increases due to factors such as charging load fluctuations, aging of internal components, and changes in ambient temperature and humidity. If these issues are not monitored and addressed in a timely manner, they may lead to thermal runaway, resulting in equipment damage, fires, and other safety accidents. This not only affects the normal operation of the charging piles but also poses a threat to the safety of surrounding personnel and property.

[0003] Most monitoring and protection measures for thermal runaway in charging piles remain at the level of single-parameter monitoring or fixed-mode intervention. Some monitoring methods only focus on the overall temperature of the charging pile or the temperature value of specific key components, failing to comprehensively collect and analyze the temperature change patterns and environmental influencing factors in different temperature distribution areas of the charging pile. This results in inaccurate identification of thermal runaway risks, often leading to misjudgments or missed detections. For example, when the temperature in a certain temperature distribution area slowly rises but has not yet reached the preset threshold, if it is accompanied by abnormal changes in ambient humidity, it may already have the potential conditions for thermal runaway. However, existing single-temperature monitoring methods cannot capture this complex risk signal, thus missing the opportunity for early intervention.

[0004] In terms of protective intervention, existing technologies mostly adopt fixed cooling operation modes. That is, after activating the cooling device according to a preset temperature threshold, cooling is carried out according to a fixed path or method, failing to dynamically adjust the operation trajectory based on the thermal runaway risk characteristics of different temperature distribution areas. This fixed intervention mode may result in some high-risk areas not being adequately cooled, while low-risk areas are over-cooled, which not only affects the protective effect but also wastes energy. In addition, during the cooling intervention process, existing methods lack dynamic acquisition and feedback of real-time environmental parameters, and cannot promptly correct the intervention direction according to the actual changes in the temperature distribution area during the intervention process. This leads to insufficient adaptability and accuracy of protective operations, making it difficult to effectively curb the development trend of thermal runaway.

[0005] With the increasing number of charging piles and the growing complexity of their operating conditions, existing monitoring and protection technologies are no longer sufficient to meet the needs of accurate identification and efficient intervention of thermal runaway risks in practical applications. There is an urgent need for a charging pile thermal runaway protection monitoring method that can comprehensively collect multi-dimensional features, dynamically adjust protection strategies, and has real-time correction capabilities to improve the safety and reliability of charging pile operation. Summary of the Invention

[0006] The purpose of this invention is to provide a monitoring method for thermal runaway protection of charging piles, so as to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides a monitoring method for thermal runaway protection of charging piles, the method comprising:

[0008] The charging pile operation parameter set is obtained through the charging pile monitoring platform, and the temperature of the charging pile temperature distribution center is collected based on the charging pile operation parameter set to obtain the center temperature value.

[0009] Based on the feature acquisition of each temperature distribution area in the charging pile operation parameter set, the temperature change characteristics and environmental humidity characteristics corresponding to each temperature distribution area are obtained.

[0010] By learning the thermal runaway response characteristics of each temperature distribution area based on the central temperature value, the temperature change characteristics and environmental humidity characteristics of each temperature distribution area are obtained, thus obtaining a static thermal runaway response array.

[0011] The operable path of the charging pile protection device is obtained, and based on each static thermal runaway response value in the static thermal runaway response array and the corresponding temperature distribution area, the operating trajectory of the charging pile protection device when performing cooling intervention on the operable path is controlled.

[0012] During the cooling intervention process, real-time environmental parameters of the current temperature distribution area are collected to determine the dynamic thermal runaway response value. Based on the dynamic and static thermal runaway response values ​​of the current temperature distribution area, the intervention direction of the charging pile protection device is corrected.

[0013] Preferably, feature acquisition is performed based on the temperature distribution regions of the charging pile operating parameter set to obtain the temperature change characteristics and environmental humidity characteristics corresponding to each temperature distribution region. Specifically, this includes:

[0014] The charging pile operation parameter set is obtained, and the region is segmented based on the distribution density of the charging pile operation parameter set to obtain multiple temperature distribution regions;

[0015] The temperature change trend is obtained, and an initial temperature distribution area is selected based on the temperature change trend to collect temperature change features. The next temperature distribution area is selected based on the temperature change trend to collect temperature change features, until the temperature change features corresponding to each temperature distribution area are determined.

[0016] Obtain the environmental wind direction vector, reselect the initial temperature distribution area based on the environmental wind direction vector to collect environmental humidity features, select the next temperature distribution area based on the environmental wind direction vector to collect environmental humidity features, until the environmental humidity features corresponding to each temperature distribution area are determined.

[0017] Preferably, before collecting features from the various temperature distribution areas in the charging pile operating parameter set, the method further includes: obtaining the current temperature change trend and wind direction vector through the data center of the charging pile monitoring platform.

[0018] Preferably, the static thermal runaway response array is obtained by learning the thermal runaway response characteristics of each temperature distribution region based on the central temperature value and the corresponding temperature change characteristics and environmental humidity characteristics, specifically including:

[0019] The depth of the reference area is determined based on the center temperature value;

[0020] The temperature change characteristics and environmental humidity characteristics corresponding to the first temperature distribution area are obtained. Based on the location coordinates of the first temperature distribution area and the depth of the reference area, a reference distribution area is selected for thermal runaway response feature learning to obtain the static thermal runaway response value corresponding to the first temperature distribution area.

[0021] After obtaining the temperature change characteristics and environmental humidity characteristics corresponding to other temperature distribution areas, the static thermal runaway response values ​​corresponding to each temperature distribution area are determined in the same way, and the array is filled according to the charging pile operation parameter set to obtain the static thermal runaway response array.

[0022] Preferably, the operable path of the charging pile protection device is obtained, and based on each static thermal runaway response value in the static thermal runaway response array and their corresponding temperature distribution regions, the operating trajectory of the charging pile protection device when performing cooling intervention on the operable path specifically includes:

[0023] Obtain each static thermal runaway response value and its corresponding temperature distribution region in the static thermal runaway response array;

[0024] Obtain all operable path segments of the charging pile protection device; obtain each operable path segment and the adjacent temperature distribution area corresponding to the operable path segment;

[0025] The starting point of the operation trajectory of the charging pile protection device is obtained, and the static thermal runaway response values ​​of all adjacent temperature distribution areas corresponding to the operable path segment are accumulated as the response concentration corresponding to the operable path segment. The path planning algorithm is used to determine the operation trajectory of the charging pile protection device when it starts cooling intervention from the starting point of the operation trajectory.

[0026] Preferably, correcting the intervention direction of the charging pile protection device based on the dynamic and static thermal runaway response values ​​of the current temperature distribution area specifically includes:

[0027] Obtain the response difference between the dynamic thermal runaway response value and the static thermal runaway response value of the current temperature distribution area;

[0028] When the response difference is higher than a preset threshold, the correction range is determined based on the temperature parameters of the current temperature distribution area, and the dynamic thermal runaway response value and static thermal runaway response value of the adjacent temperature distribution area are obtained based on the correction range.

[0029] The intervention direction of the charging pile protection device is corrected based on the dynamic and static thermal runaway response values ​​of the adjacent temperature distribution area.

[0030] Preferably, before obtaining the charging pile operation parameter set through the charging pile monitoring platform, the method further includes: obtaining charging pile application demand data, and determining target safe temperature value data based on the charging pile application demand data;

[0031] The target safe temperature value data is used for subsequent feature acquisition and thermal runaway response feature learning.

[0032] Preferably, obtaining charging pile application demand data specifically includes:

[0033] Acquire charging pile usage scenario data; the charging pile usage scenario data includes charging power and heat dissipation configuration.

[0034] Obtain charging pile performance data; the charging pile performance data includes rated voltage and maximum load;

[0035] Based on the charging pile usage scenario data and the charging pile performance data, the charging pile application demand data is obtained.

[0036] Preferably, determining the target safe temperature value based on the charging pile application requirement data specifically includes:

[0037] The initial safe temperature value is obtained by matching the charging pile application requirement data with the pre-stored safety threshold library.

[0038] The initial safe temperature value is adjusted according to the optimization criteria to obtain the target safe temperature value data;

[0039] The target safe temperature value data is used to define the benchmark for temperature acquisition and feature acquisition.

[0040] Preferably, before acquiring real-time environmental parameters of the current temperature distribution area to determine the dynamic thermal runaway response value during the cooling intervention process, the following steps are also included:

[0041] The timestamp information in the charging pile operation parameter set is parsed to generate a dynamic timeline, and key temperature event nodes are marked on the dynamic timeline;

[0042] Based on the dynamic time axis and position mapping system, the temperature distribution area is spatially mapped to obtain an enhanced temperature visualization map;

[0043] The enhanced temperature visualization map is used to identify different risk zones, and the identified data is used to determine the dynamic thermal runaway response value.

[0044] Compared with the prior art, the beneficial effects of the present invention are:

[0045] This monitoring method for thermal runaway protection of charging piles acquires operational parameter sets through a charging pile monitoring platform and collects temperature data at the temperature distribution center. Starting from the overall operational status, it accurately captures the basic temperature information of the core area of ​​the charging pile, providing comprehensive and crucial initial data support for subsequent thermal runaway risk analysis. This avoids the risk identification bias caused by incomplete information in traditional single-parameter monitoring. In the feature acquisition stage, temperature change characteristics and environmental humidity characteristics are acquired separately for each temperature distribution area. This breaks through the limitation of traditional monitoring focusing only on a single temperature parameter, fully considering the temperature change patterns of different areas and the impact of environmental factors on thermal runaway. It can more comprehensively reflect the potential thermal runaway risk status of each area, making the judgment of thermal runaway risk richer and helping to detect potential thermal runaway hazards earlier.

[0046] By learning the thermal runaway response characteristics of each region based on the central temperature value and the environmental humidity characteristics, a static thermal runaway response array is formed. This array integrates and analyzes multi-dimensional feature information, establishing a correlation between each temperature distribution area and the thermal runaway risk. This allows for precise classification of thermal runaway risk levels in different areas, enabling personnel or control systems to clearly understand the risk differences in each area and providing clear guidance for subsequent targeted protective interventions. During the protective intervention phase, based on the response values ​​and corresponding area information of each region in the static thermal runaway response array, the control protective device adjusts the cooling intervention trajectory along the operable path. This overcomes the "one-size-fits-all" drawbacks of traditional fixed-mode interventions, dynamically allocating cooling resources according to the risk level of different areas. This allows for more targeted cooling treatment in high-risk areas and avoids unnecessary energy consumption in low-risk areas, improving the protective effect while achieving rational energy utilization.

[0047] During cooling intervention, real-time environmental parameters of the current temperature distribution area are collected to determine the dynamic thermal runaway response value. This dynamic response value is then combined with the static thermal runaway response value to correct the intervention direction of the protective device, constructing a closed-loop protection mechanism of "monitoring-intervention-feedback-correction." This mechanism can capture the state changes of each area in real time during intervention, promptly identify deviations between the initial intervention strategy and the actual situation, and ensure that the protective operation always matches the current thermal runaway risk state through dynamic correction. This effectively avoids insufficient or excessive protection caused by a fixed intervention direction. For example, if the real-time ambient humidity of a certain area suddenly changes abnormally during cooling, causing a significant difference between the dynamic and static thermal runaway response values, this method can adjust the intervention direction of the cooling device in a timely manner based on this change, ensuring that the area receives appropriate cooling treatment, thereby more effectively curbing the thermal runaway trend.

[0048] This method features a tightly integrated workflow, encompassing parameter acquisition, feature analysis, response learning, trajectory control, and real-time correction. It forms a complete and logically rigorous thermal runaway protection system capable of adapting to the complex needs of charging piles under various operating conditions. Whether for routine risk monitoring during daily operation or emergency intervention during sudden temperature anomalies, it demonstrates excellent adaptability and effectiveness. The application of this method significantly enhances the early warning and response capabilities of charging piles to thermal runaway risks, reduces equipment failures and safety accidents caused by thermal runaway, extends the service life of charging piles, and ensures the stability of charging pile operation. It also provides strong support for the safe development of new energy vehicle charging infrastructure, further promoting the healthy progress of the new energy vehicle industry. Attached Figure Description

[0049] Figure 1 This is a schematic diagram illustrating the working principle of the monitoring method for thermal runaway protection of charging piles according to the present invention.

[0050] Figure 2 A flowchart for collecting temperature variation characteristics and environmental humidity characteristics of various temperature distribution areas;

[0051] Figure 3 Flowchart for obtaining static thermal runaway response array;

[0052] Figure 4 A flowchart for the intervention direction correction of the charging pile protection device. Detailed Implementation

[0053] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0054] Please see Figure 1 This invention provides a monitoring method for thermal runaway protection of charging piles, the method comprising:

[0055] The charging pile monitoring platform acquires a set of charging pile operating parameters. Based on these parameters, the temperature at the center of the charging pile's temperature distribution is collected to obtain the center temperature value. Feature acquisition is performed on each temperature distribution area within the charging pile operating parameter set to obtain the temperature change characteristics and ambient humidity characteristics corresponding to each area. Thermal runaway response characteristics are learned from the center temperature value and the corresponding temperature change and ambient humidity characteristics of each temperature distribution area, resulting in a static thermal runaway response array. The operable path of the charging pile protection device is obtained. Based on the static thermal runaway response values ​​and their corresponding temperature distribution areas in the static thermal runaway response array, the operating trajectory of the charging pile protection device during cooling intervention along the operable path is controlled. During the cooling intervention process, real-time environmental parameters of the current temperature distribution area are collected to determine the dynamic thermal runaway response value. The intervention direction of the charging pile protection device is corrected based on the dynamic and static thermal runaway response values ​​of the current temperature distribution area.

[0056] Example 1: See Figure 2 This involves in-depth processing of the charging pile's operating parameter set to obtain characteristic information of various temperature distribution areas. The execution of this process relies on the charging pile monitoring platform's data center, which integrates multiple sensor data streams and historical records, providing real-time and historical time-series information. Initially, the system obtains the current temperature change trend and wind direction vector from the data center. The temperature change trend is a calculated data sequence reflecting the direction and rate of change in recent temperature readings, typically generated by analyzing temperature values ​​from multiple sampling points over a past period using moving average or regression analysis. The wind direction vector originates from meteorological sensors or environmental monitoring units and includes parameters such as wind direction angle and wind speed.

[0057] After acquiring these basic environmental parameters, the system begins processing the charging pile's operating parameter set. This parameter set contains raw data collected by multiple temperature sensors located inside and around the charging pile. This data includes not only temperature values ​​but also the spatial coordinates of the sensors. When performing region segmentation based on distribution density, the system employs spatial clustering algorithms, such as the density-based DBSCAN method or the K-means algorithm, to group sensors that are physically adjacent and have similar temperature values, forming multiple temperature distribution regions. Each region represents a physical space with similar thermal characteristics, such as the power module area, cable connection area, or heat sink perimeter area of ​​the charging pile. The result of region segmentation is the generation of a region mapping table, where each region has a unique identifier and a list of included sensors.

[0058] The system selects the initial temperature distribution area based on the temperature change trend. The selection logic is based on the strength of the trend; the system identifies the area with the most significant temperature change trend as the starting point. For example, if the temperature in a certain area rises continuously over a short period with a large slope, that area is selected as the initial area. When collecting temperature change characteristics of this initial area, the system calculates the area's characteristic indicators, including temperature gradient, temperature fluctuation frequency, and temperature extremes. The temperature gradient is obtained by calculating the temperature difference between adjacent sensors within the area, reflecting the direction of heat propagation; the temperature fluctuation frequency is obtained by analyzing the periodic characteristics of time series data, indicating temperature instability; and the temperature extremes are the highest and lowest temperature readings within the area. These characteristics together constitute the temperature change characteristics of the area.

[0059] Based on the analyzed initial region and temperature change trend, the system selects the next temperature distribution region. The selection criteria consider trend correlation and spatial proximity; for example, it selects other regions adjacent to the initial region with similar temperature change trends. The same temperature change feature acquisition process is performed on this region, calculating its temperature gradient, fluctuation frequency, and extreme values. The system iteratively executes this process, processing each temperature distribution region sequentially until temperature change features have been acquired for all regions. Ultimately, each region has a corresponding temperature change feature vector for subsequent analysis.

[0060] During the environmental humidity characteristic acquisition phase, the system obtains the environmental wind direction vector and reselects the initial temperature distribution area accordingly. The selection logic at this stage is based on the influence of wind direction; for example, an area upstream of the wind direction is chosen as the initial area because the humidity distribution in that area may affect the downstream area. When acquiring environmental humidity characteristics for this initial area, the system reads data from humidity sensors within that area and calculates relative humidity, absolute humidity, and humidity change rate. Relative humidity is the percentage of current water vapor pressure to saturated water vapor pressure; absolute humidity is the mass of water vapor per unit volume of air; and humidity change rate is the rate at which humidity changes over time. These indicators collectively constitute the environmental humidity characteristics of the area.

[0061] Based on the environmental wind direction vector, the system selects the next temperature distribution area, for example, the downstream area pointed to by the wind direction vector or an adjacent area on the airflow path. The same environmental humidity feature acquisition process is performed on this area, calculating its relative humidity, absolute humidity, and rate of change of humidity. The system iteratively processes each area until environmental humidity features have been acquired for all temperature distribution areas. Ultimately, each area has a corresponding environmental humidity feature vector.

[0062] Throughout the implementation process, the system maintains a feature database, storing the identifier, spatial coordinates, temperature variation characteristics, and environmental humidity characteristics of each temperature distribution area. This data provides input for subsequent thermal runaway response feature learning. Through this sequential feature acquisition method, the system can comprehensively capture the thermal behavior and environmental impact of the charging pile, providing a data foundation for thermal runaway protection.

[0063] Example 2: See Figure 3 This system focuses on learning thermal runaway response characteristics, transforming temperature and environmental features into an operable static response array. The process uses a center temperature as a benchmark, constructing a comprehensive spatial distribution map reflecting thermal runaway risk through feature analysis of various temperature distribution regions. The implementation begins with acquiring the center temperature value. This core parameter, collected from the center of the charging pile's temperature distribution, represents the baseline level of the overall thermal state. Based on this center temperature value, the system calculates the reference region depth. The reference region depth is a derived parameter whose value is positively correlated with the degree to which the center temperature value deviates from the safe range, while also considering the physical size of the temperature distribution region. This depth defines the spatial influence range during feature learning, used to determine the range of adjacent regions that need to be referenced for each temperature distribution region.

[0064] The system retrieves the temperature change characteristics and ambient humidity characteristics corresponding to the first temperature distribution region from the feature database. The first temperature distribution region can be any selected starting point, typically the region with the most significant temperature change or closest to the center point. The temperature change characteristics of this region include quantitative indicators such as temperature gradient, fluctuation frequency, and extreme values; the ambient humidity characteristics include parameters such as relative humidity, absolute humidity, and humidity change rate. Based on the location coordinates of the first temperature distribution region and the depth of the reference region, the system selects a reference distribution region. The location coordinates are obtained from a region mapping table, accurately marking the position of each region in two-dimensional or three-dimensional space. The depth of the reference region determines the size of the selected range, for example, a circular or spherical space centered on the first region with a depth value equal to the radius. All temperature distribution regions within this spatial range are listed as reference distribution regions.

[0065] The thermal runaway response feature learning process employs machine learning algorithms. The system inputs feature data from the first temperature distribution region and feature data from a reference distribution region into the prediction model. This model, trained on historical data, can identify the correlation between feature patterns and thermal runaway risk. The model output is the static thermal runaway response value corresponding to the first temperature distribution region, a standardized value that quantifies the thermal runaway risk level of that region. A higher value indicates a greater risk and requires higher priority intervention. After analyzing the first region, the system processes other temperature distribution regions sequentially. The same processing logic is used for each region: acquiring the temperature change and ambient humidity characteristics of that region; selecting a reference distribution region based on the region's location coordinates and the depth of the reference region; inputting the feature data into the machine learning model; and outputting the static thermal runaway response value for that region. This process traverses all temperature distribution regions, ensuring that each region has a corresponding quantified risk value.

[0066] The system fills the array based on the charging pile's operating parameter set, which contains the spatial coordinates of all temperature distribution areas. The system matches these coordinates with the corresponding static thermal runaway response values ​​to construct a complete spatial array. This static thermal runaway response array is stored in matrix form, where each element represents the degree of thermal runaway risk at the corresponding spatial location. The array's row and column structure corresponds to the physical layout of the charging piles, facilitating subsequent path planning and intervention control.

[0067] Throughout the implementation process, the system maintains a response value database, recording the identifier, spatial coordinates, and static thermal runaway response value for each temperature distribution area. This data provides direct evidence for cooling intervention operations, enabling protective devices to target high-risk areas. Through this systematic feature learning method, the charging pile thermal runaway protection system can transform complex temperature and environmental data into a concise and clear risk distribution map, providing data support for precise prevention and control.

[0068] Example 3: See Figure 4 This involves the operational trajectory planning of charging pile protection devices. This process, based on a static thermal runaway response array and operable path information, generates the optimal cooling intervention trajectory through a systematic path planning algorithm. The implementation begins with acquiring each static thermal runaway response value and its corresponding temperature distribution region from the static thermal runaway response array. This data comes from a previous feature learning process; each temperature distribution region has a corresponding risk quantification value, which is stored in the system's response value database.

[0069] The system acquires all operable path segments of the charging pile protection device. Each operable path segment is a discrete unit of the device's mechanical structure that can move, and each segment has spatial coordinates for its start and end points. These path segments are predefined by mechanical design parameters and stored in the system's path configuration file. The system reads this configuration file to obtain the geometric information and connectivity of all operable path segments. The system establishes a mapping relationship between operable path segments and temperature distribution areas. A spatial location matching algorithm associates each operable path segment with its adjacent temperature distribution areas. Adjacency is determined based on the spatial distance and orientation angle between the path segment and the area; typically, a distance threshold is set, and areas within this threshold are considered adjacent. The system generates a list of adjacent areas for each operable path segment, recording the identifiers of all adjacent temperature distribution areas. The system acquires the starting point of the charging pile protection device's operational trajectory. This starting point is usually the device's standby position or the end point of the last intervention operation, and its coordinates are obtained from the device's position sensors or control system records. Determining the starting point provides an initial reference for path planning.

[0070] For each operable path segment, the system calculates its response concentration value. The response concentration is the weighted sum of the static thermal runaway response values ​​of all adjacent temperature distribution regions within that path segment, reflecting the overall thermal runaway risk level around that path segment. The calculation uses the following formula:

[0071]

[0072] in: It represents the response concentration value of the currently operable path segment and is a dimensionless quantitative indicator. This represents the total number of temperature distribution areas adjacent to this path segment; This represents the weight coefficient of the i-th adjacent region. This coefficient is determined by the distance and orientation relationship between the region and the path segment. The closer the distance, the higher the weight. This represents the static thermal runaway response value of the i-th neighboring region, derived from the static thermal runaway response array. The system calculates and stores the response concentration value for each operable path segment.

[0073] Based on the calculated response concentration values, the system employs a path planning algorithm to determine the optimal operating trajectory. The algorithm uses the starting point of the operating trajectory as the starting point, the response concentration value as the weight index for each path segment, and a graph search algorithm to find the path with the optimal cumulative weight. Commonly used algorithms include Dijkstra's algorithm or A.S. algorithm. Algorithms are developed to effectively identify the trajectory with the highest cumulative response concentration among numerous possible paths, prioritizing paths that traverse high-risk areas. During algorithm execution, the system considers the mechanical constraints of the protective device, including maximum speed, turning radius, and operation duration. These constraints serve as boundary conditions, ensuring the generated operation trajectory is physically executable. The algorithm outputs a complete sequence of operation trajectories, including the execution order of path segments, speed, and dwell time.

[0074] The generated operational trajectory is converted into control commands and sent to the actuator of the charging pile protection device. The device moves and operates according to the planned trajectory, performing cooling intervention along the operable path. Throughout the intervention process, the system continuously monitors the trajectory execution and makes necessary adjustments based on real-time feedback. This implementation method transforms static thermal runaway risk data into dynamic intervention operations through a systematic path planning approach, enabling the protection device to efficiently prioritize high-risk areas. The combination of response concentration calculation and path planning algorithms ensures the targetedness and effectiveness of cooling intervention, providing operational-level technical support for the thermal safety of charging piles.

[0075] Example 4: Focusing on the dynamic correction mechanism during cooling intervention. This mechanism dynamically adjusts the intervention direction of the protective device by comparing and analyzing real-time environmental parameter acquisition with static prediction data. The implementation process begins with the parsing of the timestamp information of the charging pile's operating parameters. The system extracts the acquisition timestamps of all temperature data points and constructs a dynamic time axis framework in chronological order. This time axis records the temperature monitoring event sequence with millisecond-level precision, forming a continuous time coordinate system. Key temperature event nodes are marked on the dynamic time axis. These nodes are automatically identified based on the rate of temperature change: when the rate of temperature change in a certain area exceeds a set threshold, the system automatically marks it as a key node. For example, if the charging pile's power module area experiences a temperature rise of 2°C per second at 14:30:25, the system marks it as a "power module temperature rise event".

[0076] Based on a dynamic timeline and location mapping system, the system performs spatial mapping of temperature distribution areas. The location mapping system uses a three-dimensional coordinate system (X, Y, Z) to define the physical space of the charging station, with each temperature distribution area bound to a unique coordinate range. The system maps key event nodes on the timeline to corresponding spatial coordinates, generating an enhanced temperature visualization map. This visualization map uses layered rendering technology: the bottom layer displays a basic temperature distribution heatmap, the middle layer overlays a timeline event marker layer, and the top layer highlights the outline of high-risk areas. Temperature levels are distinguished by color gradients (blue indicates low temperature, red indicates high temperature), key event locations are marked by flashing identifiers, and high-risk area boundaries are marked by a semi-transparent overlay layer.

[0077] Using enhanced temperature visualization maps, the system performs identification processing on different risk zones. The processing involves three steps: first, temperature values ​​are normalized, mapping the actual temperature to the 0-100 range; second, risk weight coefficients for each region are calculated based on historical data; and finally, a regional risk index is generated using a weighted algorithm. The processed data forms the basis for calculating the dynamic thermal runaway response value. The system collects real-time environmental parameters (including temperature, humidity, and airflow velocity) for the current temperature distribution area and combines them with the identification processing results to calculate the dynamic thermal runaway response value. This value reflects the actual thermal risk status at the current moment.

[0078] The system synchronously acquires the static thermal runaway response value (from previous prediction data) of the same area and calculates the response difference between the dynamic and static values. The response difference is calculated using the absolute difference algorithm: |Dynamic response value - Static response value|. The preset threshold is set according to the safety level of the charging pile, for example, 15 for ordinary commercial charging piles and 10 for high-speed charging piles. When the system detects that the response difference in the power module area reaches 18 (exceeding the threshold of 15), the correction mechanism is triggered.

[0079] The correction range is determined based on the temperature parameters of the current temperature distribution area. These parameters include the current temperature value, the rate of temperature rise, and the thermal conductivity coefficient. The system calculates the heat diffusion range using a thermodynamic model and defines the correction range as an elliptical region centered on the current area, with its major axis extending along the main heat dissipation direction. For example, in a power module temperature rise event, the correction range covers the adjacent capacitor bank area and the terminal block area.

[0080] The system acquires the dynamic and static thermal runaway response values ​​of all adjacent temperature distribution areas within the correction range, forming a set of correction parameters, as shown in Table 1.

[0081] Table 1: Spatiotemporal Mapping Table of Temperature Events

[0082]

[0083] Based on the comparison of response values ​​in neighboring areas, the system performs intervention direction correction. The correction algorithm analyzes deviations in three dimensions: the magnitude of the response differences between areas indicates the direction of heat propagation; the rate of change of the differences reflects the speed of crisis development; and the gradient of the differences between areas shows the path of thermal runaway propagation. The system generates a three-dimensional correction vector (ΔX, ΔY, ΔZ) based on these parameters, which includes the direction adjustment angle and distance parameters. For example, if the system detects that the rate of increase in the difference in the capacitor bank area is higher than that in the terminal block area, it shifts the intervention direction 15 degrees towards the capacitor bank area and simultaneously increases the movement speed parameter in that direction.

[0084] After receiving the calibration command, the protective device adjusts the nozzle angle and coolant spray trajectory in real time. The adjustment process employs closed-loop control: the device collects new data every 5 centimeters of movement, dynamically updating the calibration parameters. When three consecutive samples show that the response difference has fallen back within the threshold, the system automatically releases the calibration state and returns to the basic intervention mode. The time accuracy of the entire calibration process is controlled within 200 milliseconds, ensuring rapid response to sudden temperature rise events.

[0085] Example 5: This describes the pre-configuration process for monitoring thermal runaway protection in charging piles. This process determines core safety baseline parameters through systematic data collection and analysis. The implementation begins with acquiring charging pile application requirement data, which includes key information on equipment usage scenarios and performance specifications. The system first collects charging pile usage scenario data. This data category is obtained by accessing the charging pile configuration archive, which stores engineering parameters entered during equipment installation. Charging power data is extracted from the electrical parameter record table, including power values ​​in standard charging mode and peak power in fast charging mode. Heat dissipation configuration data comes from the cooling system design document, recording the heat sink type, heat dissipation medium properties, and heat dissipation channel layout parameters. For example, the file for a DC fast charging pile shows: standard charging power 60kW, fast charging peak power 180kW; the heat dissipation system adopts a dual-fan air-cooled structure, equipped with copper-based heat sinks and three independent air ducts.

[0086] Simultaneously, charging pile performance data is collected, retrieving core parameters from the equipment technical specification library. Rated voltage data is read from the charging pile nameplate parameters or control system registration information to distinguish the voltage range of DC piles and the voltage level of AC piles. Maximum load data is analyzed from electrical safety test reports to record the maximum current carrying capacity under continuous operation. A charging pile's technical document states: DC output rated voltage 500V, maximum continuous load current 300A; AC output rated voltage 380V, maximum continuous load current 150A. The system inputs charging pile usage scenario data and charging pile performance data into the data fusion module. This module performs data standardization processing: power units are uniformly converted to kilowatts, voltage units to volts, and current units to amperes. The processed data is stored in the application requirement database according to the equipment's unique identifier, forming structured charging pile application requirement data records. Each record contains three data segments: equipment number, scenario parameter set, and performance parameter set.

[0087] Based on the application requirements data of charging piles, the system initiates the process of determining the target safe temperature value. The system accesses a pre-stored safety threshold database, which stores safe temperature benchmark values ​​categorized by device type and power level. The matching process employs a multi-level retrieval mechanism: first, it matches the device category index based on the charging pile type; second, it matches the power range index based on the charging power; and finally, it matches the cooling efficiency coefficient based on the heat dissipation configuration. The matching algorithm outputs an initial safe temperature value, which represents a general safety benchmark for similar devices.

[0088] The system optimizes and adjusts the initial safe temperature value, following thermal efficiency and safety redundancy criteria. The thermal efficiency criterion analyzes the heat conversion efficiency curve in the charging pile's historical operating data; when the thermal efficiency is detected to be lower than the set standard, the system automatically lowers the initial safe temperature value. The safety redundancy criterion assesses environmental risk factors, including ventilation conditions and ambient temperature fluctuation range, and increases temperature redundancy based on the risk assessment level. The adjusted value is rounded to the nearest integer to generate the target safe temperature value. This target safe temperature value is written to the system's core parameter register as a benchmark for subsequent monitoring processes. This data is used to determine the normal range of the temperature distribution center during temperature acquisition; as an evaluation benchmark for temperature change characteristics during feature acquisition; and as the calculation origin for the static response value during thermal runaway response feature learning. When performing temperature-related operations, the system uses the target safe temperature value in this register for real-time comparison and analysis.

[0089] The entire implementation process establishes a closed-loop verification mechanism: the system automatically scans the charging pile operation logs monthly, and when it detects changes in equipment heat dissipation configuration or adjustments to power parameters, it triggers a process to re-collect application requirement data, thereby updating the target safe temperature value data. This dynamic maintenance mechanism ensures that safety benchmark parameters are always synchronized with the actual state of the equipment, providing an accurate benchmark reference for thermal runaway protection monitoring.

[0090] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0091] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A monitoring method for thermal runaway protection of charging piles, characterized in that, include: The charging pile operation parameter set is obtained through the charging pile monitoring platform, and the temperature of the charging pile temperature distribution center is collected based on the charging pile operation parameter set to obtain the center temperature value. Based on the feature acquisition of each temperature distribution area in the charging pile operation parameter set, the temperature change characteristics and environmental humidity characteristics corresponding to each temperature distribution area are obtained. By learning the thermal runaway response characteristics of each temperature distribution area based on the central temperature value, the temperature change characteristics and environmental humidity characteristics of each temperature distribution area are obtained, thus obtaining a static thermal runaway response array. The operable path of the charging pile protection device is obtained, and based on each static thermal runaway response value in the static thermal runaway response array and the corresponding temperature distribution area, the operating trajectory of the charging pile protection device when performing cooling intervention on the operable path is controlled. During the cooling intervention process, real-time environmental parameters of the current temperature distribution area are collected to determine the dynamic thermal runaway response value. Based on the dynamic and static thermal runaway response values ​​of the current temperature distribution area, the intervention direction of the charging pile protection device is corrected.

2. The monitoring method for thermal runaway protection of charging piles as described in claim 1, characterized in that, Based on the feature acquisition of various temperature distribution regions in the charging pile operation parameter set, the temperature change characteristics and environmental humidity characteristics corresponding to each temperature distribution region are obtained, specifically including: The charging pile operation parameter set is obtained, and the region is segmented based on the distribution density of the charging pile operation parameter set to obtain multiple temperature distribution regions; The temperature change trend is obtained, and an initial temperature distribution area is selected based on the temperature change trend to collect temperature change features. The next temperature distribution area is selected based on the temperature change trend to collect temperature change features, until the temperature change features corresponding to each temperature distribution area are determined. Obtain the environmental wind direction vector, reselect the initial temperature distribution area based on the environmental wind direction vector to collect environmental humidity features, select the next temperature distribution area based on the environmental wind direction vector to collect environmental humidity features, until the environmental humidity features corresponding to each temperature distribution area are determined.

3. The monitoring method for thermal runaway protection of charging piles as described in claim 1, characterized in that, Before collecting features from the various temperature distribution areas in the charging pile operating parameter set, the process also includes: obtaining the current temperature change trend and wind direction vector through the data center of the charging pile monitoring platform.

4. The monitoring method for thermal runaway protection of charging piles as described in claim 1, characterized in that, By learning the thermal runaway response characteristics of each temperature distribution region based on the central temperature value, and considering the temperature change characteristics and ambient humidity characteristics, a static thermal runaway response array is obtained, which specifically includes: The depth of the reference area is determined based on the center temperature value; The temperature change characteristics and environmental humidity characteristics corresponding to the first temperature distribution area are obtained. Based on the location coordinates of the first temperature distribution area and the depth of the reference area, a reference distribution area is selected for thermal runaway response feature learning to obtain the static thermal runaway response value corresponding to the first temperature distribution area. After obtaining the temperature change characteristics and environmental humidity characteristics corresponding to other temperature distribution areas, the static thermal runaway response values ​​corresponding to each temperature distribution area are determined in the same way, and the array is filled according to the charging pile operation parameter set to obtain the static thermal runaway response array.

5. The monitoring method for thermal runaway protection of charging piles as described in claim 1, characterized in that, The operable path of the charging pile protection device is obtained. Based on the static thermal runaway response values ​​and their corresponding temperature distribution regions in the static thermal runaway response array, the specific operating trajectory of the charging pile protection device when performing cooling intervention on the operable path includes: Obtain each static thermal runaway response value and its corresponding temperature distribution region in the static thermal runaway response array; Obtain all operable path segments of the charging pile protection device; obtain each operable path segment and the adjacent temperature distribution area corresponding to the operable path segment; The starting point of the operation trajectory of the charging pile protection device is obtained, and the static thermal runaway response values ​​of all adjacent temperature distribution areas corresponding to the operable path segment are accumulated as the response concentration corresponding to the operable path segment. The path planning algorithm is used to determine the operation trajectory of the charging pile protection device when it starts cooling intervention from the starting point of the operation trajectory.

6. The monitoring method for thermal runaway protection of charging piles as described in claim 1, characterized in that, The intervention direction of the charging pile protection device is corrected based on the dynamic and static thermal runaway response values ​​of the current temperature distribution area. Specifically, this includes: Obtain the response difference between the dynamic thermal runaway response value and the static thermal runaway response value of the current temperature distribution area; When the response difference is higher than a preset threshold, the correction range is determined based on the temperature parameters of the current temperature distribution area, and the dynamic thermal runaway response value and static thermal runaway response value of the adjacent temperature distribution area are obtained based on the correction range. The intervention direction of the charging pile protection device is corrected based on the dynamic and static thermal runaway response values ​​of the adjacent temperature distribution area.

7. The monitoring method for thermal runaway protection of charging piles as described in claim 1, characterized in that, Before obtaining the charging pile operation parameter set through the charging pile monitoring platform, the process also includes: obtaining charging pile application demand data, and determining target safe temperature value data based on the charging pile application demand data. The target safe temperature value data is used for subsequent feature acquisition and thermal runaway response feature learning.

8. The monitoring method for thermal runaway protection of charging piles as described in claim 7, characterized in that, Obtaining data on charging pile application requirements specifically includes: Acquire charging pile usage scenario data; the charging pile usage scenario data includes charging power and heat dissipation configuration. Obtain charging pile performance data; the charging pile performance data includes rated voltage and maximum load; Based on the charging pile usage scenario data and the charging pile performance data, the charging pile application demand data is obtained.

9. The monitoring method for thermal runaway protection of charging piles as described in claim 7, characterized in that, The target safe temperature value data determined based on the charging pile application requirement data specifically includes: The initial safe temperature value is obtained by matching the charging pile application requirement data with the pre-stored safety threshold library. The initial safe temperature value is adjusted according to the optimization criteria to obtain the target safe temperature value data; The target safe temperature value data is used to define the benchmark for temperature acquisition and feature acquisition.

10. The monitoring method for thermal runaway protection of charging piles as described in claim 1, characterized in that, Before determining the dynamic thermal runaway response value by collecting real-time environmental parameters of the current temperature distribution area during the cooling intervention process, the following steps are also included: The timestamp information in the charging pile operation parameter set is parsed to generate a dynamic timeline, and key temperature event nodes are marked on the dynamic timeline; Based on the dynamic time axis and position mapping system, the temperature distribution area is spatially mapped to obtain an enhanced temperature visualization map; The enhanced temperature visualization map is used to identify different risk zones, and the identified data is used to determine the dynamic thermal runaway response value.

Images