Intelligent inspection control system and method for unmanned battery replacement cabinet based on safety monitoring

Abstract

The application relates to the technical field of intelligent device monitoring, and particularly discloses an unmanned battery replacement cabinet intelligent inspection control system and method based on safety monitoring. The application acquires multi-source sensing data in real time by arranging a multi-source sensing network in the battery replacement cabinet, carries out preprocessing operation on the multi-source sensing data, extracts multi-source sensing features based on the multi-source sensing data, and forms a core factor set. A two-factor coupling analysis model is constructed according to the core factor set, and the allowed change range of the battery position is output, so that a type of battery position information is preliminarily screened and classified. A battery stability evaluation model is established according to the preliminary screening and classification result, the battery stability coefficient is secondarily evaluated, whether the battery has a displacement hidden danger is judged, and the corresponding displacement hidden danger grade is output. The application realizes the whole-process optimization from data acquisition, intelligent analysis to risk decision, and significantly improves the safety, reliability and intelligent level of the unmanned battery replacement cabinet system.

Description

Technical Field

[0001] This invention relates to the field of intelligent device monitoring technology, and more specifically, to an intelligent inspection control system and method for unmanned battery swapping cabinets based on security monitoring. Background Technology

[0002] With the rapid development of electric two-wheelers and shared battery swapping models, unmanned battery swapping cabinets are widely deployed in open or semi-open environments such as roadsides, community parking areas, and public transportation nodes. While these application scenarios are convenient for users, they also expose the equipment to complex external environments for extended periods, especially near driveways or high-traffic areas. Ground-transmitted vibrations from vehicle movement, braking, and loading / unloading continuously act on the battery swapping cabinet structure. Since battery swapping cabinets typically employ multi-layered battery compartments and automated pushing and clamping mechanisms for battery storage and retrieval, external vibrations transmitted through the cabinet to the internal structure can easily cause micro-displacement or even cumulative shifting of the batteries within the compartments if there are gaps in the battery compartment's limiting structure, the locking mechanism is not fully and reliably closed, or wear and loosening occurs after long-term use.

[0003] Under the aforementioned vibration coupling effect, the actual position of the battery may gradually deviate from the preset pick-up and drop position, resulting in problems such as grasping deviation, jamming, or even pick-up and drop failure during subsequent automatic battery swapping. In severe cases, it may also cause safety hazards such as battery falling, interface damage, or abnormal electrical connections. In addition, existing unmanned battery swapping cabinets lack effective means of sensing and assessing the structural response, battery displacement state, and its evolution caused by environmental vibration, making it difficult to identify potential risks in a timely manner and provide early warning or intervention control.

[0004] Therefore, it is necessary to provide an intelligent inspection and control system and method for unmanned battery swapping cabinets based on safety monitoring to solve the above-mentioned technical problems. In order to solve the above problems, a technical solution is provided. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the prior art, this invention provides an intelligent inspection control system and method for unmanned battery swapping cabinets based on safety monitoring. This system addresses the problem that existing unmanned battery swapping cabinets are susceptible to micro-displacement and cumulative deviation of batteries due to structural gaps or loose locking mechanisms caused by external environmental vibrations, which can lead to retrieval and placement failures and safety hazards. Furthermore, there is currently a lack of effective sensing and early warning methods.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] The intelligent inspection and control system for unmanned battery swapping cabinets based on safety monitoring includes a multi-source data sensing module, a multi-source sensing feature extraction module, a two-factor coupling analysis module, a battery stability assessment module, and a displacement hazard identification module.

[0008] The multi-source data sensing module is used to acquire multi-source sensing data in real time by deploying a multi-source sensor network inside the battery swapping cabinet and to perform preprocessing operations on the multi-source sensing data. The multi-source sensing data includes vibration response data and battery location information. The battery location information includes Class I battery location information and Class II battery location information.

[0009] The multi-source sensing feature extraction module is used to extract multi-source sensing features based on multi-source sensing data to form a core factor set.

[0010] The two-factor coupling analysis module is used to construct a two-factor coupling analysis model based on the core factor set, output the allowable range of battery location variation, and perform preliminary screening and classification of a class of battery location information;

[0011] The battery stability assessment module is used to establish a battery stability assessment model based on the initial screening and classification results, and to conduct a secondary assessment of the battery stability coefficient.

[0012] The displacement hazard detection module is used to determine whether there is a displacement hazard in the battery based on the battery stability coefficient, output the corresponding displacement hazard level, and provide a hazard warning.

[0013] As a further aspect of the present invention, the multi-source sensing features include vibration sensing features and battery position sensing features; the core factor set includes an excitation factor set and a response factor set; the excitation factor set is constructed based on the vibration sensing features, and the response factor set is constructed based on the battery position sensing features.

[0014] As a further aspect of the present invention, multi-source sensing features are extracted based on multi-source sensing data. These multi-source sensing features include vibration sensing features and battery position sensing features, specifically:

[0015] Vibration sensing characteristics include vibration change values;

[0016] Battery position sensing features include a first type of position sensing feature and a second type of position sensing feature; the first type of position sensing feature includes a first type of position offset and a first type of position change rate; the second type of position sensing feature includes a second type of position offset and a second type of position change rate.

[0017] As a further aspect of the present invention, a two-factor coupled analysis model is constructed based on the core factor set to output the allowable range of variation in battery location, and to perform preliminary screening and classification of a class of battery location information. The specific steps are as follows:

[0018] The set of incentive factors and the set of response factors in the core factor set are extracted and processed according to time series to establish a first-class and a second-class coupled change relationship.

[0019] Based on the two types of coupling change relationship, the two types of position sensing features corresponding to the vibration sensing features within the monitoring period are obtained, and the allowable change range of battery position is analyzed.

[0020] Based on the allowable range of battery position variation, a primary screening of a class of position sensing features is performed. The class of position sensing features that meet the conditions are summarized as primary non-abnormal sensing features; the class of position sensing features that do not meet the conditions are summarized as primary abnormal sensing features.

[0021] As a further aspect of the present invention, one type of coupling change relationship is as follows: , This represents a class of position-sensing features corresponding to the vibration-sensing characteristics at time t. The vibration sensing characteristics at time t are... The first type of position-aware feature at time t; the second type of coupling change relationship is... , These are the second-class position sensing features corresponding to the vibration sensing features at time t. The second type of position sensing feature at time t.

[0022] As a further aspect of the present invention, a battery stability evaluation model is established based on the preliminary screening and classification results, and the battery stability coefficient is evaluated in a second step. The specific steps are as follows:

[0023] Obtain the first-level no-anomaly sensing features and the corresponding vibration sensing features of the second-level position sensing features, and calculate the sensing feature difference; the sensing feature difference includes the offset sensing feature difference and the change sensing feature difference;

[0024] A battery stability analysis model is constructed based on the difference in perceived features, and the battery stability coefficient is calculated.

[0025] As a further aspect of the present invention, the presence of potential displacement risk in the battery is determined based on the battery stability coefficient, the corresponding displacement risk level is output, and a risk warning is issued. The specific steps are as follows:

[0026] The battery stability coefficient is compared with a preset stability threshold. If the battery stability coefficient is greater than or equal to the preset stability threshold, the battery does not have a risk of displacement. If the battery stability coefficient is less than the preset stability threshold, the battery has a risk of displacement.

[0027] For batteries with potential displacement risks, the corresponding displacement risk level is determined according to the battery stability coefficient, and a risk warning is issued.

[0028] As a further aspect of the present invention, for batteries with potential displacement risks, the corresponding displacement risk level is determined according to the battery stability coefficient, and a risk warning is issued. The specific steps are as follows:

[0029] For batteries with potential displacement risks, extract the corresponding battery stability coefficients and arrange the battery stability coefficients in descending order to obtain a stability sequence.

[0030] The stability sequence is compared with the preset risk range in order. If the battery stability coefficient is lower than the lower limit of the preset risk range, the corresponding displacement risk level is Level 1; if the battery stability coefficient is within the preset risk range, the corresponding displacement risk level is Level 2; if the battery stability coefficient is greater than the upper limit of the preset risk range, the corresponding displacement risk level is Level 3.

[0031] The intelligent inspection and control method for unmanned battery swapping cabinets based on security monitoring includes the following steps:

[0032] By deploying a multi-source sensor network inside the battery swapping cabinet, multi-source sensing data is acquired in real time, and preprocessing operations are performed on the multi-source sensing data.

[0033] Multi-source sensing features are extracted from multi-source sensing data to form a core factor set;

[0034] A two-factor coupled analysis model is constructed based on the core factor set to output the allowable range of battery location variation and to perform preliminary screening and classification of a class of battery location information.

[0035] A battery stability assessment model was established based on the preliminary screening and classification results, and the battery stability coefficient was evaluated in a second stage.

[0036] The system determines whether there is a potential displacement risk based on the battery stability coefficient, outputs the corresponding displacement risk level, and provides a risk warning.

[0037] A readable storage medium storing a computer program, which, when executed by a processor, is used to implement the steps of the intelligent inspection and control system for unmanned battery swapping cabinets based on security monitoring as described in any of the above claims.

[0038] The technical effects and advantages of this invention—an intelligent inspection control system and method for unmanned battery swapping cabinets based on safety monitoring—are as follows: This invention integrates vibration information with battery location information through multi-source data sensing and feature extraction, overcoming the limitations of traditional single-parameter monitoring and thus improving the comprehensiveness and accuracy of monitoring. By constructing a two-factor coupled analysis model and establishing the allowable variation range of battery location based on two types of location sensing features, the system can dynamically adjust the judgment criteria according to the actual vibration conditions, avoiding the problem of false alarms easily generated by fixed threshold methods in complex environments, and significantly improving the robustness and adaptability of the judgment.

[0039] This invention separates obviously abnormal data from potentially normal data through primary screening of a type of location-sensing features, improving data processing efficiency and providing high-quality input for subsequent refined analysis. It introduces a battery stability assessment model based on the difference in sensing features, performing secondary analysis on seemingly normal data to deeply uncover early-stage potential problems, enabling the system to provide early warning capabilities rather than just post-event detection. Through a displacement hazard identification and grading mechanism, the stability assessment results are transformed into specific risk levels and early warning information, achieving closed-loop control from data analysis to practical application. This invention can adopt differentiated processing strategies based on different hazard levels, such as early warning prompts, key monitoring, or mandatory intervention, thereby reducing maintenance costs while ensuring safety. It achieves full-process optimization from data collection and intelligent analysis to risk decision-making, significantly improving the safety, reliability, and intelligence level of the unmanned battery swapping cabinet system. Attached Figure Description

[0040] Figure 1 This is a system block diagram of an intelligent inspection and control system for unmanned battery swapping cabinets based on security monitoring, provided in an embodiment of the present invention.

[0041] Figure 2 This is a flowchart illustrating the intelligent inspection and control method for unmanned battery swapping cabinets based on safety monitoring, provided in an embodiment of the present invention. Detailed Implementation

[0042] The technical solutions of this invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described technical solutions are only a part of this invention, and not all of it. All other technical solutions obtained by those skilled in the art based on the technical solutions of this invention without inventive effort are within the scope of protection of this invention.

[0043] Example 1: As Figure 1 The diagram shown is a system block diagram of an intelligent inspection and control system for unmanned battery swapping cabinets based on safety monitoring, provided in an embodiment of the present invention. The intelligent inspection and control system for unmanned battery swapping cabinets based on safety monitoring includes a multi-source data sensing module, a multi-source sensing feature extraction module, a two-factor coupling analysis module, a battery stability assessment module, and a displacement hazard identification module. The multi-source data sensing module is connected to the multi-source sensing feature extraction module, the multi-source sensing feature extraction module is connected to the two-factor coupling analysis module, the two-factor coupling analysis module is connected to the battery stability assessment module, and the battery stability assessment module is connected to the displacement hazard identification module.

[0044] The multi-source data sensing module is used to acquire multi-source sensing data in real time by deploying a multi-source sensor network inside the battery swapping cabinet and to perform preprocessing operations on the multi-source sensing data. The multi-source sensing data includes vibration response data and battery position information. The battery position information includes Class I battery position information and Class II battery position information. Class I battery position information is the battery position information collected in real time, and Class II battery position information is the battery position information under standard operating conditions.

[0045] The multi-source sensing feature extraction module is used to extract multi-source sensing features based on multi-source sensing data to form a core factor set. Among them, the multi-source sensing features include vibration sensing features and battery position sensing features; the core factor set includes an excitation factor set and a response factor set; the excitation factor set is formed based on the vibration sensing features, and the response factor set is formed based on the battery position sensing features.

[0046] The two-factor coupling analysis module is used to construct a two-factor coupling analysis model based on the core factor set, output the allowable range of battery location variation, and perform preliminary screening and classification of a class of battery location information;

[0047] The battery stability assessment module is used to establish a battery stability assessment model based on the initial screening and classification results, and to conduct a secondary assessment of the battery stability coefficient.

[0048] The displacement hazard detection module is used to determine whether there is a displacement hazard in the battery based on the battery stability coefficient, output the corresponding displacement hazard level, and provide a hazard warning.

[0049] It should be noted that all parameter data involved in the calculation need to be dimensionless beforehand.

[0050] Preferably, the multi-source data sensing module is used to acquire multi-source sensing data in real time by deploying a multi-source sensor network inside the battery swapping cabinet. Specifically, it is used to acquire vibration signals as vibration response data in real time by deploying vibration sensing nodes, such as triaxial accelerometers, inside the battery swapping cabinet, and to acquire battery position information based on position detection units, such as limit switches, displacement sensors, or vision modules, set at the battery compartment or tray.

[0051] Preprocessing operations are performed on multi-source sensing data, including bandpass filtering of vibration response data to remove low-frequency drift and high-frequency noise, unified timestamp alignment and denoising of multi-source sensing data, and resampling to unify the time scale of multi-source sensing data, ensuring that data from different sources are comparable under the same time reference.

[0052] Preferably, multi-source sensing features are extracted based on the multi-source sensing data. These multi-source sensing features include vibration sensing features and battery location sensing features, specifically:

[0053] Vibration sensing characteristics include vibration change values;

[0054] Battery position sensing features include a first type of position sensing feature and a second type of position sensing feature; the first type of position sensing feature includes a first type of position offset and a first type of position change rate; the second type of position sensing feature includes a second type of position offset and a second type of position change rate.

[0055] Preferably, a two-factor coupled analysis model is constructed based on the core factor set to output the allowable range of variation in battery location, and a preliminary screening and classification of a class of battery location information is performed. The specific steps are as follows:

[0056] The set of incentive factors and the set of response factors are extracted from the core factor set and subjected to time-series processing to construct two types of coupled change relationships: a first type and a second type. The first type of coupled change relationship is... , This represents a class of position-sensing features corresponding to the vibration-sensing characteristics at time t. The vibration sensing characteristics at time t are... The first type of position-aware feature at time t; the second type of coupling change relationship is... , These are the second-class position sensing features corresponding to the vibration sensing features at time t. The second type of position-aware features at time t;

[0057] Based on the two types of coupling change relationship, the two types of position sensing features corresponding to the vibration sensing features within the monitoring period are obtained, and the allowable change range of battery position is analyzed.

[0058] Based on the allowable range of battery position variation, a primary screening of a class of position sensing features is performed. The class of position sensing features that meet the conditions are summarized as primary non-abnormal sensing features; the class of position sensing features that do not meet the conditions are summarized as primary abnormal sensing features.

[0059] In one embodiment of the invention, the invention is deployed in an unmanned battery swapping cabinet system, which is particularly suitable for complex operating environments such as urban shared battery swapping stations, centralized battery swapping cabinets for electric two-wheelers, and energy storage and swapping equipment in underground parking lots. By integrating vibration and location information, dynamic monitoring and intelligent judgment of the battery installation status can be achieved, effectively identifying structural loosening, installation misalignment, and potential displacement risks, which is of great significance in ensuring the safe operation of the equipment.

[0060] Multi-source sensing data is collected through a multi-source sensor network deployed inside the battery swapping cabinet. For example, triaxial accelerometers are installed on the cabinet frame and battery compartment structure to acquire vibration response data in real time, reflecting external environmental disturbances such as ground vibrations caused by passing vehicles and impacts when users insert or remove batteries. Simultaneously, displacement sensors or visual measurement devices are installed on the battery trays, guide rails, or fixing clips to acquire battery position information. This position information is further divided into two categories: one category consists of real-time data collected during actual operation, reflecting the current actual position of the battery; the other category consists of baseline data acquired under standard operating conditions or ideal installation conditions, typically constructed from factory calibration or historical stable operating data, used as a comparative reference.

[0061] After data acquisition, the multi-source sensing feature extraction stage begins. Vibration response data is processed to extract vibration change values ​​as vibration sensing features, characterizing the excitation intensity experienced by the system at the current moment. Simultaneously, battery position information is decomposed and modeled, extracting first-class position offsets and first-class position change rates, as well as second-class position offsets and second-class position change rates, thus forming a complete battery position sensing feature system. Based on this, vibration sensing features are categorized into an excitation factor set, and battery position sensing features are categorized into a response factor set, constructing a core factor set to provide a data foundation for subsequent coupled analysis.

[0062] A two-factor coupled analysis model is constructed based on a core factor set. Through time-series correlation analysis of excitation and response factors, a mapping relationship between vibration input and position response is established at each time step. Specifically, at time step... Construct a class of coupled change relationships This is used to describe the relationship between vibration and a class of position responses; simultaneously, it constructs two types of coupled change relationships. This is used to describe the relationship between vibration and Type II position response. Since Type II position sensing features are derived from standard operating conditions, they provide a more accurate description of the normal state of the structure. Therefore, a "normal response model" is preferentially established based on the Type II coupling relationship.

[0063] In actual operation, the sequence of two types of coupling changes is continuously collected within a monitoring cycle, and a dynamic mapping interval between vibration and type II position response is established through statistical analysis. For example, based on historical anomaly-free data, the distribution range of type II position offset and rate of change under different vibration intensities can be statistically analyzed, thus forming the "allowable range of change" for the battery under normal conditions. This range is not a fixed value, but changes dynamically with vibration intensity: the allowable displacement range is appropriately widened when the vibration is strong, and tightened when the vibration is weak. This adaptive range construction method enables the model to more realistically reflect actual physical laws and avoids misjudgments caused by traditional fixed threshold methods under complex operating conditions.

[0064] After determining the permissible range of battery location variation, the first-level screening stage begins. Specifically, for any given moment... The system compares a type of position sensing feature collected in real time with the allowable variation range corresponding to the current vibration conditions. If a position sensing feature falls within this range, it indicates that the battery position response at that moment is consistent with the response under standard operating conditions, and it is classified as a Level 1 no-anomaly sensing feature. Conversely, if it exceeds this range, it indicates an abnormal battery position response, which may indicate structural loosening or installation deviation, and it is classified as a Level 1 abnormal sensing feature. Through this screening process, obvious anomalies can be quickly identified from a large amount of monitoring data, while data that appears normal but may pose potential risks are retained for further analysis.

[0065] In real-world applications, such as a group of shared battery swapping stations along a main urban road, users frequently experience vehicle vibrations during peak hours, while simultaneously swapping batteries. During operation, vibration and location information is continuously collected, and a normal response range is established using a two-factor coupling model. At a certain moment, the detected vibration change value is at a moderate level, corresponding to an allowable positional offset range of ±2mm. When the positional offset of one battery is 1.5mm, it falls within the allowable range and is judged as Level 1 (no anomaly); however, when the offset of another battery reaches 3mm, it exceeds the allowable range and is judged as Level 1 (abnormal). This method accurately distinguishes between normal fluctuations and abnormal changes in complex environments.

[0066] Furthermore, the embodiments of the present invention can effectively avoid false alarms. For example, in traditional methods, when the device is under high vibration, large positional changes are easily misjudged as abnormal. However, in this solution, since the allowable range of change is dynamically adjusted based on vibration intensity, the judgment criteria are automatically relaxed under high vibration conditions, thereby avoiding misjudging normal responses as abnormalities; while under low vibration conditions, the range is tightened to improve detection sensitivity, making it easier to identify minor anomalies. This dynamic screening mechanism based on coupling significantly improves the accuracy and robustness of the system.

[0067] Furthermore, the embodiments of this invention also possess excellent engineering scalability. For example, differentiated models can be established by combining different types of battery compartment structures, or the allowable range can be self-learned and optimized based on long-term operating data, ensuring good adaptability to different devices and environments. Simultaneously, the primary screening results can also serve as input for subsequent stability assessment and hazard identification modules, enabling multi-level risk analysis.

[0068] This invention, through the construction of an analysis model based on the coupling of vibration and position, establishes a dynamic allowable range of change using two types of position sensing features, and performs preliminary screening of one type of position sensing features accordingly, achieving high-precision identification of battery position status. In real-world applications, it can not only effectively detect obvious structural anomalies but also maintain high accuracy in judgment under complex operating conditions, providing reliable technical support for the safe operation of unmanned battery swapping cabinets.

[0069] Preferably, a battery stability assessment model is established based on the preliminary screening and classification results, and the battery stability coefficient is evaluated a second time. The specific steps are as follows:

[0070] Obtain the first-level anomaly-free sensing features and the corresponding vibration sensing features of the second-level position sensing features, and calculate the sensing feature difference; the sensing feature difference includes the offset sensing feature difference and the change sensing feature difference; the offset sensing feature difference... ,in, Let be the difference in offset sensing features at time t. For a type of position offset at time t, The second type of position offset at time t; the difference in change-sensing features. ,in, The difference in perceived features at time t. Let be a type of rate of change of position at time t. Let be the rate of change of the second kind of position at time t;

[0071] A battery stability analysis model is constructed based on the difference in perceived features, and the battery stability coefficient is calculated. The calculation formula for the battery stability analysis model is as follows:

[0072]

[0073] In the formula: Let be the battery stability coefficient at time t. These are the weighting coefficients for the difference in offset-sensory features. The weighting coefficients for the difference in change-perceived features. Let be the difference in offset sensing features at time t. The perceived feature difference is the change at time t.

[0074] In one embodiment of the present invention, the invention is deployed in scenarios such as unmanned battery swapping cabinet systems, shared battery swapping stations, and energy storage battery cabinets to perform refined analysis and dynamic evaluation of the structural stability of batteries under complex operating environments. Especially in scenarios such as urban roadsides, high-frequency usage areas, or environments with complex vibrations, batteries are easily affected by external vibrations, frequent insertion and removal operations, and structural wear during long-term operation, resulting in minute but continuous displacement changes. Traditional methods often struggle to identify these early potential hazards and achieve higher-precision safety monitoring.

[0075] The battery location information is initially screened using a two-factor coupling analysis module, dividing all collected data into first-level no-anomaly sensing features and first-level anomaly sensing features. First-level no-anomaly sensing features represent the data set where the overall battery response remains within a reasonable range under vibration-position coupling constraints. Traditional methods typically classify this type of data as normal and leave it unprocessed, but it may actually contain early signals of gradual structural degradation. Therefore, this embodiment of the invention further evaluates the first-level no-anomaly sensing features to achieve more refined stability analysis.

[0076] Within a specific monitoring period, the corresponding Type I position offset and Type I position change rate are extracted from the primary anomaly-free sensing features. Simultaneously, Type II position offset and Type II position change rate are acquired under the same vibration conditions. Type I position features typically reflect the macroscopic positional changes of the battery's overall structure, such as the battery's overall displacement within the tray. Type II position features, on the other hand, reflect the response under standard operating conditions or reference conditions and can be considered as the battery's reference behavior under ideal, fixed conditions. By comparing these two types of features, the degree of difference between the actual operating state and the ideal state can be effectively revealed.

[0077] At each sampling time The system calculates the offset sensing characteristic difference to describe the offset difference between the actual battery position and the standard position; simultaneously, it calculates the change sensing characteristic difference to reflect the difference between the battery's dynamic change trend and the standard change trend. These two differences characterize the consistency of the battery's position response from both static and dynamic dimensions. When the battery is in a good fixed state, its actual response should be basically consistent with the standard response, so both the offset sensing characteristic difference and the change sensing characteristic difference are close to zero. However, when the battery has slight looseness or structural wear, these two differences will gradually increase, reflecting a decrease in stability.

[0078] Based on this, the above differences are input into the battery stability analysis model to calculate the battery stability coefficient at each moment. In practical applications, parameters can be set according to different structural characteristics. For example, for rigid connection structures, such as snap-fit ​​structures, the weight of the offset difference can be appropriately increased; for structures with elastic support or sliding, the weight of the rate of change difference can be increased to more sensitively capture dynamic anomalies.

[0079] From a model structure perspective, the difference is first amplified by a squared term to highlight anomalous changes. Then, weighted summation and square root operations are used to merge the offset and variation differences into a comprehensive difference intensity. Finally, the difference is mapped to a stability coefficient by subtracting this intensity from 1, and the absolute value is used to ensure the result is non-negative. This design has significant engineering implications: when the difference is small, the square root term is close to zero, and the stability coefficient is close to 1, indicating a stable state; as the difference gradually increases, the stability coefficient gradually decreases, indicating reduced stability; when the difference is large, the stability coefficient is close to 0, indicating a clearly unstable state.

[0080] In actual operation, battery stability coefficients are not typically judged based on a single moment's data. Instead, statistical analysis of the stability coefficient is performed over a time window. For example, the average stability coefficient, minimum value, and fluctuation range over a certain period can be calculated to more comprehensively reflect the battery's condition. If the battery stability coefficient remains at a high level with minimal fluctuations over a period of time, it indicates that the battery structure is stable. If the battery stability coefficient shows a continuous downward trend or frequent fluctuations, it may indicate that the battery's fixed structure is gradually deteriorating and requires further investigation.

[0081] In real-world scenarios, such as a shared battery swapping station located next to a main urban road, the batteries inside are affected by vehicle vibrations, user impacts, and environmental changes during daily operation. In the initial stages, the batteries are securely installed, and the Type I and Type II positional responses are highly consistent. At this point, the difference between the offset perception feature and the change perception feature is close to zero, and the stability coefficient is close to 1. As usage time increases, the battery clips may experience slight wear or loosening. Under certain vibration conditions, localized shifts may occur, while the overall positional change remains relatively stable. At this point, the difference begins to increase, but it may still be within the "no anomaly range" of the first-level screening. However, calculations using a secondary stability assessment model reveal that the stability coefficient has gradually decreased from 0.95 to 0.8 or even lower, thus identifying potential risks in advance.

[0082] Furthermore, when structural loosening worsens, significant differences in positional response may occur even under relatively low vibration conditions. In this case, the difference between the offset sensing characteristic and the change sensing characteristic increases significantly, and the stability coefficient drops rapidly, for example, to 0.6 or below. This can trigger an early warning mechanism, such as sending a notification to maintenance personnel suggesting inspection of the battery mounting structure or replacement of worn components. In this way, preventative maintenance can be performed before a failure occurs, avoiding safety accidents such as battery displacement, poor contact, or even falling.

[0083] Furthermore, the embodiments of the present invention also possess good adaptability and scalability. For example, the weighting coefficients can be customized according to the structural characteristics of different devices, or the model can be optimized through self-learning based on historical operating data to better reflect actual working conditions. Simultaneously, the stability coefficient can be integrated with other monitoring indicators, such as temperature and current, to further enhance the system's comprehensive judgment capabilities.

[0084] This invention provides a battery stability assessment model based on perceived feature differences. By constructing a difference measurement mechanism between "actual response and standard response," it achieves continuous quantitative assessment of battery structural stability. In real-world applications, this invention can further identify potential anomalies beyond initial screening, improving detection accuracy and reducing false alarm rates. This is of great significance for enhancing the safety, reliability, and intelligent operation and maintenance level of unmanned battery swapping systems.

[0085] Preferably, the battery stability coefficient is used to determine whether there is a potential displacement risk, the corresponding displacement risk level is output, and a risk warning is issued. The specific steps are as follows:

[0086] The battery stability coefficient is compared with a preset stability threshold. If the battery stability coefficient is greater than or equal to the preset stability threshold, the battery does not have a risk of displacement. If the battery stability coefficient is less than the preset stability threshold, the battery has a risk of displacement.

[0087] For batteries with potential displacement risks, the corresponding displacement risk level is determined according to the battery stability coefficient, and a risk warning is issued.

[0088] Furthermore, for batteries with potential displacement risks, the corresponding displacement risk level is determined based on the battery stability coefficient, and a risk warning is issued. The specific steps are as follows:

[0089] For batteries with potential displacement risks, extract the corresponding battery stability coefficients and arrange the battery stability coefficients in descending order to obtain a stability sequence.

[0090] The stability sequence is compared with the preset risk range in order. If the battery stability coefficient is lower than the lower limit of the preset risk range, the corresponding displacement risk level is Level 1; if the battery stability coefficient is within the preset risk range, the corresponding displacement risk level is Level 2; if the battery stability coefficient is greater than the upper limit of the preset risk range, the corresponding displacement risk level is Level 3.

[0091] In one embodiment of the present invention, the invention is deployed in an unmanned battery swapping cabinet, energy storage battery warehouse, or shared battery management system for intelligent monitoring and risk warning of the battery's fixed state in complex operating environments. Especially in urban roadsides, underground parking garages, or high-frequency usage scenarios, the battery's fixing structure may gradually loosen under the combined effects of long-term vibration and impact, frequent insertion and removal operations, and environmental changes such as temperature fluctuations and humidity changes, potentially leading to battery displacement.

[0092] Based on the aforementioned model, the stability coefficient of each battery is calculated in real time at the current moment or within the monitoring period. This reflects the consistency between the battery's macroscopic position and microscopic constraint response. The value typically ranges from 0 to 1, with values ​​closer to 1 indicating greater structural stability. Once the stability coefficient of a battery is obtained, the first layer of discrimination logic is executed, comparing the stability coefficient with a preset stability threshold. For example, in practical engineering, the stability threshold can be set to 0.75 or 0.8 based on historical operating data or experimental calibration results. When the battery's position response still conforms to structural constraints under the current vibration and operating conditions, and there is no obvious risk of displacement, it is classified as a "no displacement risk state" and routine monitoring continues; while when If the battery fails to respond properly, it indicates an inconsistency between the macroscopic and microscopic responses, which may indicate problems such as structural loosening, clip wear, or installation deviation. In this case, the battery is marked as having a "potential for displacement" and enters the second-level grading evaluation process.

[0093] In the second-level assessment, the assessment no longer simply determines "whether it is abnormal," but further sorts and classifies all batteries with potential displacement risks to achieve differentiated management. Specifically, the stability coefficients of all batteries identified as having "potential displacement risks" are extracted and sorted in descending order to form a stability sequence. For example, in a battery swapping cabinet, if five batteries have stability coefficients of 0.72, 0.65, 0.58, 0.42, and 0.30, then after sorting, a sequence is formed. The significance of this sorting process lies in quantifying and arranging unstable states of different degrees, which allows for the intuitive identification of the "most unstable" batteries, thus prioritizing the handling of high-risk items.

[0094] The stability coefficients in this stability sequence are compared with the preset hazard range. The preset hazard range typically consists of two boundary values: an upper limit and a lower limit. and lower limit For example, it can be set to , Based on this range, displacement risks can be divided into three levels: when the stability coefficient is below the lower limit (i.e., This indicates a severe decline in battery stability, with significant differences between its macroscopic and microscopic positional responses, strongly suggesting significant loosening or displacement. This is classified as a Level 1 hazard, i.e., high risk. When the stability coefficient is within the hazard range, i.e. This indicates a medium-risk state, with potential initial loosening or localized anomalies. At this point, it is classified as a Level II hazard, i.e., medium risk. When the stability coefficient is higher than the upper limit of the hazard range but still lower than the initial threshold, i.e., and If the deviation is slight, it indicates that the overall situation is still close to stable. In this case, it is judged as a level three hazard, which is low risk.

[0095] In real-world applications, such as a shared battery swapping station located next to a main urban road, batteries are frequently subjected to vehicle vibrations during peak hours, while users continuously swap batteries. At a certain point, the stability coefficient of some batteries is detected to be below the threshold of 0.75, triggering a hazard identification process. After sorting and grading, one battery is found to have a stability coefficient of 0.30, far below the lower limit of 0.5, and is therefore classified as a Level 1 hazard. At this point, a high-priority warning can be immediately triggered, such as sending an emergency notification to maintenance personnel through the backend management platform, displaying a message on the device's local display screen stating "Battery fixation abnormal, please stop using and check," or even temporarily locking the battery compartment in conjunction with the control system to prevent further user operation and thus avoid safety accidents caused by batteries falling or poor contact.

[0096] Batteries with a stability coefficient between 0.5 and 0.7, such as 0.58 or 0.65, are classified as Level 2 hazards. While these batteries haven't reached a point of severe instability, they exhibit a clear deterioration trend and require focused monitoring. This includes increasing the sampling frequency, recording continuous trends, and generating maintenance recommendations in the background, such as "It is recommended to focus on checking the battery compartment latches and guide rail structure during the next inspection." Simultaneously, data analysis can be used to determine if their stability continues to decline, in order to decide whether to upgrade to Level 1 hazard.

[0097] Batteries with a stability coefficient slightly below the threshold but still above 0.7 (e.g., 0.72) are classified as Level 3 hazards. These situations typically occur during short-term disturbances or minor structural changes, such as when a user has just inserted the battery and it is not yet fully in place, or due to brief deviations caused by momentary vibrations. For these low-risk hazards, mandatory measures are generally not triggered immediately; instead, alerts or records are used, such as marking it as a "minor anomaly" in the backend, and a comprehensive judgment is made based on subsequent data to avoid false alarms.

[0098] Furthermore, this tiered mechanism can be integrated with operation and maintenance strategies. For example, it can automatically generate maintenance tasks with different priorities based on the level of hazard: Level 1 hazards require immediate attention, Level 2 hazards require attention within a specified time, and Level 3 hazards are included in the routine inspection plan. Simultaneously, big data analysis can be used to statistically analyze the distribution of hazards across different equipment and locations, identifying structural designs or installation methods prone to problems, thereby enabling reverse optimization of equipment design.

[0099] This invention enables refined identification and hierarchical management of potential battery displacement hazards. Compared to traditional single-threshold alarm methods, it not only determines the existence of a risk but also quantifies its degree, and implements differentiated response measures based on the risk level. This ensures system safety while reducing unnecessary maintenance costs and false alarm rates. It exhibits good scalability and engineering practical value, and can be widely applied to various battery management and structural safety monitoring scenarios.

[0100] Example 2: In this embodiment of the invention, an intelligent inspection and control system for unmanned battery swapping cabinets based on safety monitoring is deployed in shared battery swapping stations along urban roads, centralized battery swapping cabinets in industrial parks, or energy storage battery swapping equipment in underground parking lots. This system is used for full-process intelligent monitoring and evaluation of battery installation stability, location security, and potential displacement risks under complex operating environments. Based on multi-source data perception, and through the collaborative work of multiple modules, closed-loop control is achieved from data acquisition, feature extraction, coupled analysis, stability assessment to hazard identification and early warning, effectively improving the operational safety and intelligent maintenance level of battery swapping equipment.

[0101] The multi-source data sensing module deploys a multi-source sensor network at key structural locations within the battery swapping cabinet. For example, triaxial accelerometers, displacement sensors, or vision measurement units are placed at the cabinet frame, battery trays, guide rails, and snap-fit ​​connections to acquire multi-source sensing data in real time. This data primarily includes vibration response data and battery position information. Vibration response data reflects external environmental disturbances, such as ground vibrations caused by passing vehicles, impacts from users inserting or removing batteries, and mechanical vibrations during equipment operation. Battery position information is divided into two categories: real-time battery position information reflecting the actual position changes of the battery under current operating conditions, and battery position information under standard operating conditions, typically established through equipment factory calibration or historical stable operating data, used as a comparison benchmark. After data acquisition, the multi-source data sensing module also performs preprocessing operations on the raw data, such as filtering and denoising, outlier removal, time synchronization, and normalization, to ensure the accuracy and consistency of subsequent analysis.

[0102] After data acquisition and preprocessing, the multi-source sensing feature extraction module performs in-depth analysis and feature extraction on the multi-source sensing data, transforming vibration response data into vibration sensing features, such as vibration change values, vibration amplitudes, or vibration frequencies, to characterize the excitation intensity experienced by the system. Simultaneously, it transforms battery position information into battery position sensing features, including Type I position offset, Type I position change rate, and Type II position offset and Type II position change rate, thus comprehensively describing the battery's spatial positional state and its dynamic change trend. Based on this, all extracted features are used to form a core factor set, which is further divided into an excitation factor set and a response factor set. Vibration sensing features serve as excitation factors, reflecting external input effects, while battery position sensing features serve as response factors, reflecting the structure's response behavior to excitation. This factor division provides a clear physical basis for subsequent coupled analysis.

[0103] The dual-factor coupling analysis module constructs a vibration-position dual-factor coupling analysis model based on the aforementioned core factor set. Through time-series correlation analysis of excitation and response factors, it establishes the coupling relationship between vibration sensing characteristics and Class I and Class II position sensing characteristics at each moment, thereby characterizing the dynamic mapping law of "vibration input-position response." Specifically, the system prioritizes using Class II position sensing characteristics to establish a normal behavior model, i.e., the baseline response under standard operating conditions. Through statistical analysis of the reasonable position response range that the battery should have under different vibration conditions, the allowable range of battery position variation is obtained. This range has adaptive characteristics, allowing greater position fluctuations under stronger vibrations and narrowing the range under weaker vibrations, thus better conforming to actual physical laws. Based on this allowable range, Class I battery position information is initially screened: if the real-time position response falls within the allowable range, it is determined to be a Class I no-abnormal sensing characteristic; if it exceeds the range, it is determined to be a Class I abnormal sensing characteristic. Through this process, obvious abnormalities can be quickly filtered from massive amounts of data, while potentially hazardous data is retained for further analysis.

[0104] After initial screening, the battery stability assessment module performs a secondary, refined evaluation of the primary anomaly-free sensing features. By modeling the differences between Class I and Class II positional features, it calculates the differences in offset sensing features and change sensing features. Based on these differences, it constructs a battery stability analysis model and outputs a battery stability coefficient. This coefficient quantifies the structural consistency and stability of the battery under current operating conditions; specifically, when the responses of Class I and Class II positions are highly consistent, the stability coefficient is close to 1; when deviations occur, the stability coefficient gradually decreases. Through statistical analysis of the stability coefficient within a time window, the system can not only identify instantaneous anomalies but also capture long-term trend changes, thereby achieving continuous monitoring and evaluation of the battery's structural state. The introduction of this module enables the system to further uncover potential risks from "preliminarily normal" data, significantly improving early fault identification capabilities.

[0105] The displacement hazard detection module, based on battery stability assessment results, makes a final determination of whether a battery has a displacement hazard and outputs the corresponding hazard level and warning information. It compares the battery stability coefficient with a preset stability threshold; if the stability coefficient is higher than the threshold, it is determined that there is no displacement hazard; if it is lower than the threshold, it is determined that a displacement hazard exists. For batteries with potential hazards, the system further classifies them according to the magnitude of the stability coefficient, for example, into Level 1, Level 2, and Level 3 hazard levels. Level 1 hazard represents a severe instability state requiring immediate action; Level 2 hazard represents a medium risk requiring close monitoring and prompt maintenance; and Level 3 hazard represents a minor anomaly that can be included in routine inspections. Depending on the hazard level, the system can adopt differentiated alert and control strategies, such as sending alarm information to maintenance personnel through the backend management platform, prompting users on the device display interface, or restricting battery use in severe cases to prevent safety accidents.

[0106] In real-world applications, such as an unmanned battery swapping station located beside a main urban road, the system is frequently affected by vehicle vibrations during peak hours, while users continuously perform battery swapping operations. The system continuously collects vibration and position information through a multi-source data sensing module. After feature extraction and coupled analysis, it detects that a certain battery's position response under the current vibration conditions deviates slightly, but remains within acceptable limits, and is therefore classified as Level 1 data with no anomalies. However, in subsequent stability assessments, the system detects a gradually widening difference between the battery's Type 1 and Type 2 position responses, causing a continuous decrease in the stability coefficient. When this coefficient falls below a preset threshold, the displacement hazard detection module identifies it as a potential hazard and further classifies it as a Level 2 or Level 1 hazard, thus issuing a timely warning. Maintenance personnel can then inspect the battery compartment according to the system prompts, finding minor wear on the clips and replacing them promptly, thereby preventing potential battery drops or poor contact issues.

[0107] This invention achieves closed-loop management of the entire process from data acquisition to risk warning through multi-source data fusion and multi-module collaborative analysis. Its core advantage lies in the introduction of a vibration-position coupling analysis mechanism and a hierarchical stability assessment method, which can not only accurately identify obvious anomalies, but also discover potential hidden dangers in advance, thereby significantly improving the safety, reliability and intelligence level of battery swapping equipment, and has broad engineering application prospects.

[0108] Example 3: Intelligent inspection and control method for unmanned battery swapping cabinets based on safety monitoring. The specific steps are as follows:

[0109] Step S1: Acquire multi-source sensing data in real time by deploying a multi-source sensor network inside the battery swapping cabinet, and perform preprocessing operations on the multi-source sensing data.

[0110] Step S2: Extract multi-source sensing features from the multi-source sensing data to form a core factor set;

[0111] Step S3: Construct a two-factor coupled analysis model based on the core factor set, output the allowable range of battery location variation, and perform preliminary screening and classification of a class of battery location information;

[0112] Step S4: Establish a battery stability assessment model based on the preliminary screening and classification results, and conduct a secondary assessment of the battery stability coefficient.

[0113] Step S5: Determine whether there is a potential displacement risk in the battery based on the battery stability coefficient, output the corresponding displacement risk level, and issue a risk warning.

[0114] like Figure 2 The diagram shown is a flowchart illustrating the intelligent inspection and control method for unmanned battery swapping cabinets based on security monitoring, according to an embodiment of the present invention. This method can be used to execute... Figure 1 The steps in the method embodiments shown are implemented in a similar manner and have similar technical effects, and will not be repeated here.

[0115] A readable storage medium storing a computer program, which, when executed by a processor, is used to implement the steps of the intelligent inspection and control system for unmanned battery swapping cabinets based on security monitoring as described in any of the above claims.

[0116] The readable storage medium can be a computer storage medium or a communication medium. A communication medium includes any medium that facilitates the transfer of computer programs from one location to another. A computer storage medium can be any available medium accessible to a general-purpose or special-purpose computer. For example, a readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application-Specific Integrated Circuit (ASIC). Alternatively, the ASIC can be located in a user equipment. Of course, the processor and the readable storage medium can also exist as discrete components in a communication device. The readable storage medium can be a read-only memory (ROM), random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0117] The present invention also provides a program product including executable instructions stored in a readable storage medium. At least one processor of the device can read the executable instructions from the readable storage medium, and the at least one processor executes the executable instructions to cause the device to implement the methods provided in the various embodiments described above.

[0118] In the embodiments of the above-described device, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly manifested as execution by a hardware processor, or execution by a combination of hardware and software modules within the processor.

[0119] Through the above embodiments, this invention integrates vibration information with battery location information through multi-source data sensing and feature extraction, overcoming the limitations of traditional single-parameter monitoring. This allows the system to understand battery structural behavior from an "excitation-response" perspective, thereby improving the comprehensiveness and accuracy of monitoring. By constructing a two-factor coupled analysis model and establishing the allowable range of battery location variation based on two types of location sensing features, the system can dynamically adjust the judgment criteria according to actual vibration conditions. This avoids the problem of false alarms easily generated by fixed threshold methods in complex environments, significantly improving the robustness and adaptability of the judgment.

[0120] By performing primary screening on a class of location-aware features, obviously abnormal data is separated from potentially normal data, which not only improves data processing efficiency but also provides high-quality input for subsequent refined analysis. A battery stability assessment model based on the difference in sensing features is introduced to perform secondary analysis on seemingly normal data, enabling in-depth discovery of early-stage potential problems and providing early warning capabilities, rather than just post-event detection. Simultaneously, the continuous quantitative expression of stability coefficients intuitively reflects the changing trends of battery structural state, providing a reliable basis for operation and maintenance decisions.

[0121] By employing a displacement hazard identification and grading mechanism, the stability assessment results are transformed into specific risk levels and early warning information, achieving closed-loop control from data analysis to practical application. The system can adopt differentiated handling strategies based on different hazard levels, such as early warning prompts, focused monitoring, or mandatory intervention, thereby reducing maintenance costs while ensuring safety. This invention optimizes the entire process from data acquisition and intelligent analysis to risk decision-making, significantly improving the safety, reliability, and intelligence level of the unmanned battery swapping cabinet system.

[0122] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

[0123] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An intelligent inspection and control system for unmanned battery swapping cabinets based on safety monitoring, characterized in that: It includes a multi-source data sensing module, a multi-source sensing feature extraction module, a two-factor coupling analysis module, a battery stability assessment module, and a displacement hazard detection module; The multi-source data sensing module is used to acquire multi-source sensing data in real time by deploying a multi-source sensor network inside the battery swapping cabinet, and to perform preprocessing operations on the multi-source sensing data. The multi-source sensing data includes vibration response data and battery position information. The battery position information includes Class I battery position information and Class II battery position information. Class I battery position information is data collected in real time during actual operation. Class II battery position information is reference data obtained under standard operating conditions or ideal installation conditions. The multi-source sensing feature extraction module is used to extract multi-source sensing features based on multi-source sensing data to form a core factor set. Among them, the multi-source sensing features include vibration sensing features and battery position sensing features; the core factor set includes the excitation factor set and the response factor set; the excitation factor set is formed based on the vibration sensing features, and the response factor set is formed based on the battery position sensing features. Specifically, the extraction of multi-source sensing features includes: Vibration sensing characteristics include vibration change values; Battery position sensing features include a first type of position sensing feature and a second type of position sensing feature; the first type of position sensing feature includes a first type of position offset and a first type of position change rate; the second type of position sensing feature includes a second type of position offset and a second type of position change rate. The two-factor coupling analysis module is used to construct a two-factor coupling analysis model based on the core factor set, output the allowable variation range of battery location, and perform preliminary screening and classification of a class of battery location information. The specific steps are as follows: The set of incentive factors and the set of response factors in the core factor set are extracted and processed according to time series to establish a first-class and a second-class coupled change relationship. Based on the two types of coupling change relationship, the two types of position sensing features corresponding to the vibration sensing features within the monitoring period are obtained, and the allowable change range of battery position is analyzed. Based on the allowable range of battery position variation, a primary screening of a type of position sensing features is performed. The type of position sensing features that meet the conditions are summarized as primary non-anomaly sensing features; the type of position sensing features that do not meet the conditions are summarized as primary anomaly sensing features. The battery stability assessment module is used to establish a battery stability assessment model based on the initial screening and classification results, and to conduct a secondary assessment of the battery stability coefficient. The specific steps are as follows: Obtain the first-level no-anomaly sensing features and the corresponding vibration sensing features of the second-level position sensing features, and calculate the sensing feature difference; the sensing feature difference includes the offset sensing feature difference and the change sensing feature difference; A battery stability analysis model is constructed based on the difference in perceived features, and the battery stability coefficient is calculated. The displacement hazard detection module is used to determine whether there is a displacement hazard in the battery based on the battery stability coefficient, output the corresponding displacement hazard level, and provide a hazard warning. The specific steps are as follows: The battery stability coefficient is compared with a preset stability threshold. If the battery stability coefficient is greater than or equal to the preset stability threshold, the battery does not have a risk of displacement. If the battery stability coefficient is less than the preset stability threshold, the battery has a risk of displacement. For batteries with potential displacement risks, the corresponding displacement risk level is determined according to the battery stability coefficient, and a risk warning is issued.

2. The intelligent inspection and control system for unmanned battery swapping cabinets based on security monitoring as described in claim 1, characterized in that, One type of coupling change relationship is , This represents a class of position-sensing features corresponding to the vibration-sensing characteristics at time t. The vibration sensing characteristics at time t are... The first type of position-aware feature at time t; the second type of coupling change relationship is... , These are the second-class position sensing features corresponding to the vibration sensing features at time t. The second type of position sensing feature at time t.

3. The intelligent inspection and control system for unmanned battery swapping cabinets based on security monitoring as described in claim 1, characterized in that, For batteries with potential displacement risks, the corresponding displacement risk level is determined according to the battery stability coefficient, and a risk warning is issued. The specific steps are as follows: For batteries with potential displacement risks, extract the corresponding battery stability coefficients and arrange the battery stability coefficients in descending order to obtain a stability sequence. The stability sequence is compared with the preset risk range in order. If the battery stability coefficient is lower than the lower limit of the preset risk range, the corresponding displacement risk level is Level 1; if the battery stability coefficient is within the preset risk range, the corresponding displacement risk level is Level 2; if the battery stability coefficient is greater than the upper limit of the preset risk range, the corresponding displacement risk level is Level 3.

4. A method for intelligent inspection and control of unmanned battery swapping cabinets based on safety monitoring, applied to the intelligent inspection and control system for unmanned battery swapping cabinets based on safety monitoring as described in any one of claims 1-3, wherein the specific steps are as follows: By deploying a multi-source sensor network inside the battery swapping cabinet, multi-source sensing data is acquired in real time, and preprocessing operations are performed on the multi-source sensing data. Multi-source sensing features are extracted from multi-source sensing data to form a core factor set; A two-factor coupled analysis model is constructed based on the core factor set to output the allowable range of battery location variation and to perform preliminary screening and classification of a class of battery location information. A battery stability assessment model was established based on the preliminary screening and classification results, and the battery stability coefficient was evaluated in a second stage. The system determines whether there is a potential displacement risk based on the battery stability coefficient, outputs the corresponding displacement risk level, and provides a risk warning.

5. A readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it is used to implement the steps of the intelligent inspection and control system for unmanned battery swapping cabinets based on security monitoring as described in any one of claims 1-3.

Images