A multi-dimensional information intelligent detection method and system based on a direct-current charging pile

Abstract

The present application relates to the technical field of direct current charging pile detection, and particularly relates to a multi-dimensional information intelligent detection method and system based on a direct current charging pile.The present application comprises a charging pile multi-dimensional information acquisition module, a charging pile multi-dimensional information analysis module and a charging pile state regulation module.The charging pile multi-dimensional information acquisition module is used to obtain current charging space environment data of the charging pile in the charging space dimension, current pile body structure operation data of the pile body structure dimension and current output interface state data of the output interface dimension.The charging pile multi-dimensional information analysis module is used to perform multi-level evaluation analysis on the current data of the above three dimensions respectively.The charging pile state regulation module is used to perform charging pile operation parameter adjustment operations based on the multi-level evaluation analysis results.The system can comprehensively acquire multi-dimensional information of the charging pile and perform in-depth analysis, accurately control and timely regulate the operation state of the charging pile, and ensure stable and safe operation of the charging pile.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of direct current charging pile detection, and particularly relates to a multi-dimensional information intelligent detection method and system based on a direct current charging pile. BACKGROUND

[0002] With the rapid development of the new energy vehicle industry, as a core energy supplement facility, the operation stability and safety of the direct current charging pile directly affect the user experience and public safety. At present, the detection methods of the direct current charging pile are mostly concentrated on single-dimensional data collection and analysis, which is difficult to comprehensively cover various key influencing factors in the operation process of the charging pile. In terms of charging space environment, the existing detection means often ignores the dynamic changes of environmental parameters such as temperature, humidity and dust concentration, and abnormal fluctuations of these parameters may cause accelerated aging of internal components of the charging pile, and even cause safety hazards such as short circuit and fire. For example, in a high temperature and high humidity environment, the heat dissipation efficiency of the charging pile will decrease significantly, and if it is not detected and intervened in time, it may cause overload damage of the charging module; in a high dust concentration scene, dust adheres to the internal circuit board of the charging pile, which easily leads to poor circuit contact and affects the normal realization of the charging function.

[0003] In the pile body structure dimension, the traditional detection method mostly relies on artificial regular inspection, which not only consumes a large amount of labor cost, but also has the problems of long detection period and untimely fault discovery. Key components in the pile body structure, such as shell connecting structure and internal support frame, may appear loose and deformation in the long-term use process due to factors such as vibration and external impact, and if it is not found in time, it may cause the stability of the pile body structure to decrease, and even cause safety accidents such as the collapse of the charging pile. At the same time, artificial inspection is difficult to accurately detect the fastening state and insulation performance of the internal key electrical connection points of the pile body, and cannot comprehensively master the operation health status of the pile body structure.

[0004] In the output interface dimension, the existing detection system mostly only focuses on electrical parameters such as charging voltage and current, and ignores the detection of the physical state of the interface, such as the wear and tear degree of the interface plug, the oxidation condition of the contact point and the waterproof sealing performance. As a key component connecting the charging pile and the new energy vehicle, the physical state of the output interface directly affects the stability and safety of the charging connection. Excessive wear and tear of the interface plug will cause the contact resistance to increase, and electric sparks are easy to occur during the charging process, which has safety hazards; oxidation of the contact point will affect the current transmission efficiency, causing the charging speed to decrease, and even causing the charging to be interrupted; failure of the waterproof sealing performance will cause rainwater and dust to enter the interface, causing short circuit failure. In addition, the existing detection system lacks the ability of collaborative analysis of multi-dimensional information, and the detection data of each dimension is independent of each other, which cannot realize comprehensive evaluation of the overall operation state of the charging pile, leading to difficulty in accurately judging the fault root cause, low fault troubleshooting efficiency and affecting the normal operation and use of the charging pile. SUMMARY

[0005] The main purpose of the present application is to provide a direct current charging pile multi-dimensional information intelligent detection method and system, aiming at solving the technical problems in the prior art.

[0006] The present application provides a direct current charging pile multi-dimensional information intelligent detection system, comprising:

[0007] The charging pile multi-dimensional information acquisition module is used to obtain the current charging space environment data of the charging pile in the charging space dimension, the current pile body structure operation data of the pile body structure dimension, and the current output interface state data of the output interface dimension.

[0008] The charging pile multi-dimensional information analysis module is used to perform multi-level evaluation analysis on the current charging space environment data of the charging space dimension, the current pile body structure operation data of the pile body structure dimension, and the current output interface state data of the output interface dimension.

[0009] The charging pile state regulation module is used to perform charging pile operation parameter adjustment operation based on the multi-level evaluation analysis result.

[0010] Preferably, the current charging space environment data of the charging space dimension includes the current environment temperature value, the current environment humidity value, the current environment dust concentration value, and the current charging pile radiator surface temperature data.

[0011] The current pile body structure operation data of the pile body structure dimension includes the current charging pile internal core component temperature value, the current charging pile vibration frequency value, and the current charging pile noise decibel value.

[0012] The current output interface state data of the output interface dimension includes the current charging interface contact resistance value and the current charging interface wear degree data.

[0013] Preferably, the current charging pile radiator surface temperature data is specifically the radiator surface maximum temperature value and the radiator surface temperature distribution gradient value.

[0014] The current charging interface wear degree data is specifically the charging interface metal reed deformation amount and the charging interface insulation layer crack area value.

[0015] Preferably, the charging pile multi-dimensional information analysis module comprises:

[0016] The charging space dimension temperature evaluation unit is used to perform normalization processing on the current environment temperature value and the current environment dust concentration value, and perform weighted fusion analysis on the radiator surface maximum temperature value and the radiator surface temperature distribution gradient value to generate a charging space temperature comprehensive evaluation index.

[0017] The pile body structure dimension health evaluation unit is used for discretizing the current charging pile internal core component temperature value, the current charging pile vibration frequency value and the current charging pile noise decibel value and calculating a component operation deviation coefficient.

[0018] The output interface dimension safety evaluation unit is used for standardizing the current charging interface contact resistance value, and performing multi-factor coupling analysis on the charging interface metal reed deformation amount and the charging interface insulating layer crack area value.

[0019] As preferred, the charging space dimension temperature evaluation unit specifically performs:

[0020] Based on the charging space dimension acquisition module, the charging pile current radiator surface temperature data is obtained, and the highest temperature value of the radiator surface is standardized and converted;

[0021] The discrete coefficient of the radiator surface temperature distribution gradient value is calculated;

[0022] The highest temperature value of the radiator surface after the standardization conversion is superimposed with the discrete coefficient;

[0023] The temperature compensation correction processing is performed in combination with the current environment temperature value;

[0024] The current environment dust concentration value is fused to calculate the heat dissipation efficiency attenuation coefficient;

[0025] The heat dissipation efficiency attenuation coefficient and the temperature compensation correction processing result are integrated to generate the charging space temperature comprehensive evaluation index.

[0026] As preferred, the pile body structure dimension health evaluation unit specifically performs:

[0027] The current charging pile vibration frequency value obtained by the pile body structure dimension acquisition module is processed by frequency domain conversion;

[0028] The feature harmonic component in the frequency domain conversion processing result is extracted;

[0029] The current charging pile noise decibel value is processed by time domain slicing;

[0030] The variance coefficient of the time domain slicing processing result is calculated;

[0031] The feature harmonic component is matched with the variance coefficient in correlation;

[0032] The thermal coupling effect calculation is performed in combination with the current charging pile internal core component temperature value;

[0033] The component operation deviation coefficient is generated based on the thermal coupling effect calculation result.

[0034] As preferred, the output interface dimension safety evaluation unit specifically performs:

[0035] Logarithmic conversion processing is performed on the current charging interface contact resistance value obtained by the output interface dimension acquisition module;

[0036] The charging interface metal spring deformation variable is subjected to geometric topology analysis;

[0037] The expansion rate of the charging interface insulating layer crack area value is calculated;

[0038] The geometric topology analysis result and the expansion rate are subjected to time sequence correlation processing;

[0039] The interface safety risk index is generated by integrating the logarithmic conversion processing result and the time sequence correlation processing result.

[0040] As preferred, the charging pile state regulation module comprises:

[0041] The charging space regulation unit is configured to compare and analyze the charging space temperature comprehensive evaluation index with the preset temperature safety threshold, and trigger a heat dissipation system power adjustment instruction according to the comparison and analysis result;

[0042] The pile body structure regulation unit is configured to calculate the deviation degree of the component operation deviation coefficient from the preset health threshold interval, and trigger a mechanical structure fastening operation according to the deviation degree calculation result;

[0043] The output interface regulation unit is configured to map and match the interface safety risk index with the preset risk level scale, and trigger a charging interface power-off protection operation according to the mapping and matching result.

[0044] As preferred, a data interaction channel is arranged between the charging space regulation unit, the pile body structure regulation unit and the output interface regulation unit;

[0045] The data interaction channel is configured to transmit the associated matching parameters of the charging space temperature comprehensive evaluation index, the component operation deviation coefficient and the interface safety risk index;

[0046] The associated matching parameters are configured to coordinate the execution priority of the heat dissipation system power adjustment instruction, the mechanical structure fastening operation and the charging interface power-off protection operation.

[0047] As preferred, the present application further comprises a direct current charging pile multi-dimensional information intelligent detection method, which comprises all the modules and method processes of the above-mentioned direct current charging pile multi-dimensional information intelligent detection system.

[0048] The present application has the following advantages:

[0049] The direct current charging pile multi-dimensional information intelligent detection system can simultaneously obtain current charging space environment data of the charging space dimension, current pile body structure operation data of the pile body structure dimension and current output interface state data of the output interface dimension, breaks the limitation of single dimension data collection of the traditional detection system, and realizes comprehensive coverage of key influencing factors in the operation process of the charging pile. From the charging space environment, real-time collection of environmental parameters such as temperature, humidity and dust concentration can realize real-time mastering of the dynamic change of the charging space environment, timely discovery of environmental factors that may have adverse effects on the operation of the charging pile, avoidance of safety hazards such as accelerated aging of internal components of the charging pile, short circuit and fire caused by environmental factors, and guarantee of stable operation of the charging pile under suitable environmental conditions.

[0050] In the pile body structure dimension, the collection of operation data such as the connection structure of the pile body shell, the internal support frame, the fastening state of the key electrical connection points and the insulation performance can realize real-time monitoring of the operation state of the pile body structure without relying on regular manual inspection, greatly reducing the labor cost and shortening the fault discovery time. By mastering the operation state of the key components of the pile body structure in real time, problems such as looseness, deformation, poor contact of electrical connection points and decrease of insulation performance of the pile body structure can be found in time, safety accidents such as tilting and electrical failure caused by decrease of stability of the pile body structure can be avoided, the service life of the pile body structure can be prolonged, and the long-term healthy operation of the pile body structure can be guaranteed.

[0051] In the output interface dimension, not only the electrical parameters such as charging voltage and current are concerned, but also the physical state data such as interface plug-in wear degree, contact oxidation condition and waterproof sealing performance are collected to comprehensively master the operation state of the output interface. Through real-time monitoring of the physical state of the interface, problems such as increase of contact resistance, decrease of current transmission efficiency and failure of waterproof sealing of the interface can be found in time, safety hazards such as charging interruption, electric spark generation and short circuit caused by interface problems can be avoided, the stability and safety of charging connection can be guaranteed, and the user charging experience can be improved.

[0052] The charging pile multi-dimensional information analysis module respectively performs multi-level evaluation and analysis on the data of each dimension, not simply judges single data, but deeply mines and analyzes the data from different levels and different angles, can more accurately identify abnormal conditions existing in the data of each dimension, and accurately judges the health degree of the operation state of each dimension. This multi-level evaluation and analysis method can effectively avoid fault misjudgment caused by single data judgment deviation, improve the accuracy of evaluation of the operation state of each dimension. At the same time, through separate analysis of multi-dimensional data, the specific situation of the operation state of each dimension can be clearly mastered, and a clear direction for subsequent targeted adjustment can be provided.

[0053] The charging pile status control module adjusts charging pile operating parameters based on multi-level evaluation and analysis results, enabling proactive intervention and optimization of the charging pile's operating status. When the analysis results show anomalies in one or more dimensions, it can promptly adjust the corresponding operating parameters. For example, when the ambient temperature of the charging space is too high, it adjusts the operating parameters of the charging pile's heat dissipation system to improve heat dissipation efficiency and reduce the impact of ambient temperature on the charging pile; when key components of the pile structure become slightly loose, it adjusts relevant fastening control parameters to eliminate potential safety hazards; when oxidation of the output interface contacts leads to increased contact resistance, it adjusts the charging current parameters to avoid safety issues caused by excessive contact resistance. This precise control based on analysis results can promptly resolve problems that arise during charging pile operation, prevent the escalation of faults, ensure the continuous and stable operation of the charging pile, reduce downtime caused by faults, and improve the operational efficiency and service quality of the charging pile. Furthermore, the entire system automates the data acquisition, analysis, and control process, requiring minimal manual intervention, further reducing labor costs and improving the timeliness and accuracy of detection and control, providing strong support for the safe, stable, and efficient operation of DC charging piles. Attached Figure Description

[0054] Figure 1 This is a timing diagram of a multi-dimensional information intelligent detection system for DC charging piles as described in this invention.

[0055] Figure 2 A schematic diagram illustrating the working principle of current charging space environment data in the charging space dimension;

[0056] Figure 3 This is a flowchart illustrating the working principle of the multi-dimensional analysis module.

[0057] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0058] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0059] like Figure 1As shown, this application provides a method and system for intelligent detection of multi-dimensional information of DC charging piles. This includes building a complete hardware detection and software analysis platform. At the hardware level, the system integrates various sensors and data acquisition units, while at the software level, it deploys dedicated data analysis algorithms and control logic execution units. The core of the system lies in the coordinated operation of three main modules: a multi-dimensional information acquisition module for charging piles is responsible for collecting raw data from the charging pile itself and its surrounding environment in real time; a multi-dimensional information analysis module for charging piles performs in-depth processing and fusion analysis on the massive amounts of heterogeneous data collected, generating evaluation indicators of different dimensions; and a charging pile status control module automatically executes corresponding control commands based on the evaluation results output by the analysis module, thereby achieving intelligent closed-loop management of the charging pile status. The operation of the entire system relies on preset thresholds and algorithm models, which can be configured and optimized through the system backend according to actual application scenarios.

[0060] In one embodiment, Example 1: See Figure 2 Multiple sets of high-precision sensors deployed at different locations on the charging pile systematically acquire real-time operational data across the charging space, the charging pile structure, and the output interface. All sensors are connected to a central data acquisition unit via shielded cables, which features signal conditioning, analog-to-digital conversion, and data preprocessing capabilities. For data acquisition in the charging space dimension, ambient temperature and humidity values ​​are obtained using an integrated temperature and humidity sensor installed on the outside of the charging pile's protective casing. This sensor is typically located near the air intake grille to avoid interference from internal heat sources, and its sampling frequency is set to once per minute, effectively capturing slow changes in environmental parameters. Environmental dust concentration monitoring utilizes a particulate sensor based on the laser scattering principle. Its optical cavity design is moisture-proof, and its installation position faces the direction of vehicle entry to better reflect the true concentration of suspended particulate matter in the air. This sensor outputs data in micrograms per cubic meter, updating the data once per second. The acquisition of heat sink surface temperature data was carried out using a non-contact infrared thermal imager. This device was fixed inside the charging pile and continuously scanned the heat sink fin area. After the thermal image was transmitted to the processor, two key parameters were extracted by a special algorithm: one is the highest temperature value of all pixels in the scanned area, which is the highest temperature value of the heat sink surface; the other is to calculate the temperature difference between adjacent pixels and take the average value to obtain the temperature distribution gradient value of the heat sink surface. This gradient value can effectively reflect the uniformity of heat dissipation efficiency.

[0061] Data acquisition at the structural level of the charging pile involves monitoring its internal physical state. Temperature monitoring of core components is achieved using PT100 platinum resistance temperature sensors attached to the IGBT substrate of the power module, the housing of the DC support capacitor, and the surface of the filter inductor coil. These sensors employ high-temperature resistant insulating encapsulation and use thermally conductive silicone grease to ensure measurement accuracy, with data refreshed once per second. Vibration frequency monitoring utilizes a triaxial MEMS accelerometer rigidly mounted inside the main support structure of the charging pile. Its measurement range covers 0 to 2000 Hz, capturing mechanical vibrations caused by internal power device switching, cooling fan rotation, and the external environment. The raw acceleration data output by the sensor is transformed using FFT to extract the dominant frequency as the current charging pile vibration frequency value. Noise monitoring is achieved through a waterproof microphone array installed at different locations inside the charging pile. This array consists of four sound pressure sensors, employing time-division multiplexing technology to eliminate fan noise interference. The acquired sound wave signals are A-weighted filtered, and the root mean square value is calculated, ultimately outputting the current charging pile noise level in decibels.

[0062] Data acquisition at the output interface level focuses on the electrical and mechanical state of the charging interface. Contact resistance measurement is achieved by injecting a small test current into the positive and negative power circuits of the charging gun interface and measuring the corresponding voltage drop. This measurement is performed during charging intervals to avoid large current interference. The measurement results are processed by a temperature compensation algorithm to obtain the current contact resistance value of the charging interface. Interface wear data acquisition utilizes a vision inspection system consisting of a high-resolution miniature camera integrated inside the charging gun handle and a ring-shaped LED supplementary light source. Image acquisition is automatically triggered when the charging gun is returned to its base. The acquired internal interface images are analyzed using image processing algorithms: first, the metal spring contour is located using an edge detection algorithm; then, it is compared with a standard contour database to calculate the deformation, thus obtaining the deformation of the charging interface metal spring. Simultaneously, texture analysis is performed on the surface of the insulation layer, identifying the pixel percentage of cracked areas and calculating the cracked area value of the charging interface insulation layer by combining it with actual size calibration. All acquired data is timestamped and labeled with a sensor ID, and transmitted to the central data processing unit via a CAN bus network. CRC checks are used during transmission to ensure data integrity, ultimately forming a structured multi-dimensional data packet for use by the analysis module. The entire acquisition module adopts a hierarchical power supply design, with key sensors equipped with backup power supplies to ensure that the final data acquisition and reporting can still be completed in the event of a main power failure.

[0063] In one embodiment, Example 2: See Figure 3The charging space temperature assessment unit specifically processes data streams from environmental sensors and thermal imaging equipment. Its processing begins with the normalization of the current ambient temperature and dust concentration. This step maps the raw data to a 0-1 scale to eliminate analytical obstacles caused by different physical dimensions. The processed normalized data, along with the highest surface temperature of the radiator and the gradient of the radiator surface temperature distribution, is input into a weighted fusion analysis algorithm. This algorithm linearly combines the four input parameters according to a preset weighting coefficient matrix, assigning the highest weight to the highest surface temperature of the radiator, followed by the ambient dust concentration. Finally, a non-linear activation function generates a comprehensive charging space temperature assessment index; a higher index value indicates a more severe heat dissipation environment.

[0064] The charging pile structure health assessment unit focuses on processing multi-source sensor data reflecting the mechanical operating status of the charging pile. This unit first discretizes the current temperature of the core components, the current vibration frequency, and the current noise level in decibels, converting continuous analog signals into a finite number of discrete state levels. This conversion is based on a pre-defined threshold interval. The discretized data is then fed into a Markov chain-based state transition probability model. This model quantifies the degree of abnormality in the operating state by calculating the difference in transition probabilities between the current state and the ideal operating state, ultimately outputting a dimensionless component operation deviation coefficient, which is negatively correlated with the equipment's health status. The output interface safety assessment unit processes the electrical and mechanical parameters of the charging interface. During the unit initialization phase, the current charging interface contact resistance value is standardized using the Z-score standardization method, converting it into a standardized value with a mean of zero and a standard deviation of one. The processed resistance value, along with the deformation of the charging interface metal spring and the area of ​​the crack in the charging interface insulation layer, is input into a multi-factor coupled analysis model. This model uses principal component analysis to reduce the dimensionality of the three parameters, extracts the principal component factors that can explain the parameter variance to the greatest extent, and then establishes a mapping relationship between the input parameters and the safety status through a radial basis function neural network, ultimately generating an interface safety score value between 0 and 100.

[0065] The three evaluation units employ a parallel pipelined processing architecture, with each unit equipped with an independent data buffer and processing core to ensure parallel processing when multiple data sources arrive simultaneously. Units exchange preliminary results via shared memory space. For example, the pile structure dimension health evaluation unit can share its calculated component operational deviation coefficients with the charging space dimension temperature evaluation unit to correct the evaluation model for the impact of ambient temperature on the internal temperature of the equipment. All evaluation results include confidence indices and timestamp information and are transmitted in real-time to the state control module via a data bus. The system also establishes a self-learning mechanism for the evaluation model parameters, dynamically adjusting normalization parameters, weighting coefficients, and state division thresholds based on historical data, making the evaluation results increasingly accurate as the equipment operates. The entire analysis module adopts a fault-tolerant design; when data from a certain sensor is abnormally missing, it can automatically switch to an estimation mode based on historical data, ensuring the continuity of the system's evaluation function.

[0066] Taking the operation scenario of a DC charging pile at a coastal charging station in a high-temperature and high-humidity environment during summer as an example, the ambient temperature of this charging pile reaches 35 degrees Celsius and the relative humidity is 85% in the afternoon. Simultaneously, due to its proximity to a road, the environment contains a certain amount of salt and dust. The charging space dimension temperature assessment unit begins processing the collected multi-source data. The unit first normalizes the current ambient temperature value of 35℃ and the current ambient dust concentration value of 115μg / m³, converting these two parameters with different dimensions into a unified value within the [0,1] interval. The processed environmental data is then combined with the radiator surface monitoring data. The highest radiator surface temperature reaches 72℃, and the surface temperature distribution gradient is 4.2℃ / cm. These two parameters are weighted and fused to produce an intermediate result. During the fusion process, the ambient temperature is assigned a weight coefficient of 0.3, the dust concentration weight is 0.2, the highest radiator surface temperature weight is 0.35, and the temperature distribution gradient weight is 0.15. These weight coefficients are pre-set according to the thermal design characteristics of this type of charging pile. The final comprehensive temperature evaluation index of the charging space was 0.76, which indicates that the heat dissipation environment is in a relatively severe state.

[0067] The structural health assessment unit of the charging pile simultaneously processes mechanical operation data. The unit receives the current temperature values ​​of the core components inside the charging pile: power module 88℃, DC support capacitor 65℃, and filter inductor 72℃. These temperature data, along with the vibration frequency of 125.6Hz and noise level of 68dB, enter a discretization process. The process maps continuous temperature readings to five discrete levels, divides the vibration frequency into ten frequency bands, and converts the noise level into four intensity levels. The discretized data is then input into the operational status assessment model. The model calculates the Mahalanobis distance between the current state and the standard operational state, yielding a component operational deviation coefficient of 1.85. This value indicates that the mechanical operation has begun to deviate from the normal range, but has not yet reached a dangerous level.

[0068] The output interface dimension safety assessment unit processes relevant data from the charging interface. The current contact resistance of the charging interface is 2.8 mΩ, which, after logarithmic transformation, becomes -5.88, making the transformed data more consistent with a normal distribution. Simultaneously, the visual inspection system measures the deformation of the charging interface metal spring to be 0.12 mm, and the area of ​​insulation layer cracking to be 3.2 mm². A multi-factor coupled analysis model processes these three parameters simultaneously. The model first establishes a nonlinear relationship curve between contact resistance and spring deformation, then analyzes the correlation between insulation cracking and contact resistance changes, and finally extracts key factors affecting interface safety through principal component analysis. The analysis results show that although the contact resistance value is within the allowable range, combined with the spring deformation and insulation aging trend, the interface safety condition exhibits a slow deterioration characteristic. The processing of the three assessment units involves data interaction and result reference. The high-temperature assessment results of the charging space dimension are shared with the pile structure dimension health assessment unit to correct the assessment model of the temperature's impact on vibration and noise. The output interface dimension safety assessment unit provides the interface temperature data to the charging space dimension unit to improve the overall thermal assessment of the heat dissipation system. All evaluation results are accompanied by timestamps and confidence levels, and are transmitted in real time to the status control module via a data bus, providing a quantitative basis for subsequent intelligent control decisions. The entire analysis process adopts a rolling time window mechanism, which not only analyzes the data characteristics at the current moment, but also considers the changing trends of recent historical data, so that the evaluation results reflect both the immediate state and have temporal continuity.

[0069] In one embodiment, Example 3: The processing flow begins with the standardization of the highest temperature value on the radiator surface. This step maps the actual measured temperature value to a numerical range of 0 to 1 through a linear transformation. The transformation process uses a dynamic threshold method, with the upper threshold dynamically adjusted according to the rated power level of the charging pile, ensuring that the standardization result can adapt to the operating characteristics of different equipment models. The standardized temperature value eliminates the evaluation bias caused by absolute values. Subsequently, the coefficient of variation of the temperature distribution gradient value on the radiator surface is calculated. This calculation is based on the temperature field data of the entire heat dissipation surface collected by a thermal imager. First, the standard deviation of the temperature values ​​at all measurement points is calculated, and then it is divided by the average temperature value to obtain the coefficient of variation. This coefficient quantifies the uniformity of the temperature distribution on the heat dissipation surface. A higher coefficient of variation indicates that the radiator may have local overheating or poor airflow.

[0070] The standardized temperature values ​​obtained in the preceding steps are superimposed with the dispersion coefficients. This superposition process is not a simple arithmetic addition, but rather a weighted geometric mean is used. The calculation formula is as follows:

[0071]

[0072] in: Indicates intermediate fusion index, This represents the highest temperature value on the standardized surface of the radiator. The coefficients represent the discreteness, and α and β are the weighting exponents of the two parameters, which are preset according to the heatsink material and structural characteristics. This fusion method can simultaneously take into account information from two dimensions: absolute temperature level and temperature distribution uniformity.

[0073] The superimposed results then enter the temperature compensation and correction stage. This stage introduces the current ambient temperature as a compensation factor and uses a multinomial fitting algorithm to establish a correlation model between ambient temperature and radiator temperature. This model eliminates the baseline influence of ambient temperature on the radiator surface temperature, making the evaluation results more focused on the performance of the heat dissipation system itself. The corrected data is further fused with the ambient dust concentration value. This fusion process uses a physics-based model to calculate the attenuation of heat dissipation performance. First, an empirical relationship model between dust concentration and the thermal conductivity between radiator fins is established. Then, the attenuation ratio of the current heat dissipation performance is calculated based on the real-time dust concentration value. This ratio reflects the actual impact of dust accumulation on heat dissipation capacity. The final comprehensive evaluation index is generated using a multilayer perceptron neural network structure. The intermediate results obtained from the previous stages are used as input features. Through nonlinear transformation of the hidden layers and linear combination of the output layers, the final comprehensive evaluation index of the charging space temperature is generated. This index ranges from 0 to 100; a higher value indicates a more severe heat dissipation environment, requiring more proactive thermal management measures. The entire processing flow adopts a sliding time window mechanism, which not only considers the data at the current moment, but also introduces the moving average of recent historical data as a reference, so that the evaluation results reflect the real-time state and have time continuity.

[0074] Taking the operation scenario of a DC charging pile located in an industrial park in a dusty environment as an example, this charging pile is installed in the charging area for logistics vehicles, where there is a certain degree of industrial dust in the surrounding environment, and there is already slight dust accumulation on the surface of the radiator. The charging space dimension temperature assessment unit initiates the data processing flow. The unit first receives real-time monitoring data from multiple sources: the ambient temperature sensor transmits a current ambient temperature value of 32.8℃, the laser dust sensor detects a current ambient dust concentration of 156μg / m³, and the infrared thermal imager transmits radiator surface temperature distribution data. The unit performs feature extraction on the radiator surface temperature data, calculates the highest radiator surface temperature value to be 68.4℃, and obtains the radiator surface temperature distribution gradient value of 3.8℃ / cm through a temperature field analysis algorithm.

[0075] The unit begins by standardizing the highest surface temperature value of the radiator using a dynamic range mapping algorithm, mapping the actual temperature value of 68.4℃ to a value range of 0-1. During the mapping process, considering the rated power level and heat dissipation design standards of the charging pile, 65℃ is set as the reference temperature, corresponding to a standardized value of 0.5, resulting in a standardized temperature value of 0.72. This value reflects the degree of deviation of the current radiator surface temperature from the design reference. Simultaneously, the unit calculates the coefficient of variation (COP) of the radiator surface temperature distribution gradient, based on data from 256 temperature measurement points collected by a thermal imager. First, the standard deviation of all measurement points is calculated to be 2.7℃, and the average temperature is 61.3℃. The COP is then calculated by the ratio of the standard deviation to the average, yielding a COP of 0.044. This COP indicates that there is some non-uniformity in the radiator surface temperature distribution, but it is not yet severe. The standardized temperature value of 0.72 and the COP of 0.044 are then superimposed using a weighted geometric mean algorithm. The algorithm assigns a weighting coefficient of 0.6 to the standardized temperature value and a weighting coefficient of 0.4 to the discrete coefficients. Through exponential weighting and square root calculation, the intermediate fusion index value is obtained as 0.61. This fusion index considers information from both the absolute temperature level and the uniformity of temperature distribution. The intermediate result enters the temperature compensation correction stage, which introduces the current ambient temperature of 32.8℃ as a compensation factor. The processing algorithm is based on a pre-established ambient temperature-radiator temperature correlation model. This model, through historical data analysis, shows that every 1℃ increase in ambient temperature leads to a corresponding increase in radiator temperature of 0.8℃. Based on this relationship, the unit performs ambient temperature compensation correction on the intermediate fusion index, revising the index value to 0.58, thus eliminating the fundamental influence of ambient temperature on the evaluation results.

[0076] The corrected data was further fused with environmental dust concentration values ​​for analysis, and the unit employed a physics-based model to calculate heat dissipation efficiency attenuation. The calculation process followed the empirical relationship curve between dust concentration and heat dissipation efficiency; when the dust concentration reached 156 μg / m³, the corresponding heat dissipation efficiency attenuation coefficient was 0.86. This coefficient indicates that due to dust accumulation, the actual heat dissipation capacity of the radiator decreased to 86% of the ideal state. The final comprehensive evaluation index was generated using a multi-layer neural network model, taking the temperature compensation correction result of 0.58 and the heat dissipation efficiency attenuation coefficient of 0.86 as input features. The neural network contained a hidden layer, using a Sigmoid activation function for nonlinear transformation, and the output layer produced a final comprehensive evaluation index value of 0.67 for the charging space temperature. This index value falls within the range of 0.6-0.8, indicating that the current heat dissipation environment presents some risk but remains within a controllable range; the system needs enhanced heat dissipation management but has not yet reached the point where immediate load reduction is required. The entire processing employed a sliding time window mechanism; the unit not only analyzed the current data but also incorporated historical data from the previous 15 minutes to calculate a moving average. Time series analysis shows that the evaluation indicators exhibit a slow upward trend, providing a data foundation for predictive maintenance. All processing results are accompanied by timestamps and data processing logs, and are transmitted in real time to the status control module via the data bus, providing a basis for intelligent control of the cooling system.

[0077] In one embodiment, Example 4, the pile structure dimensional health assessment unit first processes the vibration frequency data. After receiving the raw time-domain signal collected by the vibration sensor, the unit initiates a frequency-domain conversion process. This process uses a Fast Fourier Transform algorithm to convert the time-domain vibration waveform into a frequency-domain spectrum. The converted spectrum data is then processed by a peak detection algorithm to identify the main vibration frequency components and extract the characteristic harmonic components with significant amplitudes and harmonic relationships. These components typically correspond to specific vibration modes of rotating mechanical components inside the charging pile (such as cooling fans and transformer cores). Referring to Table 1, the vibration frequency analysis results of the charging pile within a 10-second time window are shown, including the frequency values ​​and relative amplitudes of the three main characteristic harmonic components.

[0078] Table 1: Results of Harmonic Component Analysis of Vibration Frequency Characteristics

[0079]

[0080] Simultaneously, the noise decibel data enters the time-domain slicing process, which divides the continuous sound pressure signal into multiple time segments at 100-millisecond intervals and calculates the root mean square (RMS) value of the sound pressure level for each segment. Based on these slice data, the variance coefficient is further calculated. This coefficient reflects the stability of the noise fluctuation level; a higher variance coefficient indicates that the noise source has intermittent abrupt changes or impact events. The characteristic harmonic components obtained from vibration analysis are correlated with the variance coefficients from noise analysis. This matching process uses the Pearson correlation coefficient to calculate the statistical correlation between the two sets of data sequences. The matching results reveal the coupling relationship between mechanical vibration and acoustic noise.

[0081] Combining real-time monitoring data of the core components inside the charging pile, the evaluation unit initiates a thermo-coupling effect calculation. This calculation is based on a pre-established temperature-vibration-noise coupling model, which considers physical parameters such as the thermal expansion coefficient of materials and the temperature characteristics of structural stiffness. During the calculation, the system identifies that when the power module temperature rises from 65℃ to 78℃, the fundamental vibration amplitude increases by approximately 15%, while the noise variance coefficient increases by 22%, indicating that the temperature increase exacerbates mechanical vibration and noise fluctuations. Based on all these analysis results, the evaluation unit generates a quantified component operation deviation coefficient, which comprehensively reflects the degree of deviation between the current state of the mechanical structure and the ideal operating state. The output interface dimension safety evaluation unit processes the relevant data of the charging interface in parallel. The unit first performs a logarithmic transformation on the contact resistance measurement values. The transformed resistance values ​​better conform to the normal distribution characteristics, facilitating subsequent statistical analysis. The transformed data, along with the deformation data of the charging interface metal spring, enters the analysis process. The deformation data comes from the image analysis results of the vision inspection system and represents the degree of deviation of the spring from the standard geometry.

[0082] Geometric topology analysis was performed on the reed deformation data. This analysis extracted geometric features such as curvature changes and angular deviations of the reed profile, quantifying the severity of mechanical wear through these features. Simultaneously, the system calculated the expansion rate of the crack area in the charging interface insulation layer. This calculation was based on crack area measurement data from multiple consecutive detection cycles, using a linear regression method to fit the trend of area change over time. The rate value reflects the speed of insulation aging. The geometric topology analysis results and crack expansion rate were time-series correlated using a time series similarity analysis algorithm to identify the synergistic relationship between mechanical wear and insulation aging in terms of development speed and time. The analysis revealed that when the reed deformation exceeded 0.15 mm, the crack expansion rate accelerated significantly, indicating that mechanical stress concentration accelerated the aging process of the insulation material.

[0083] All processing results are integrated to generate an interface safety risk index. The integration process uses a weighted fusion algorithm, with contact resistance having the highest weight, followed by deformation and crack rate. The generated risk index is a standardized value; a higher value indicates a greater interface safety risk. When the index exceeds a predetermined threshold, the system will trigger corresponding warning or protection actions. The entire analysis process employs multi-timescale data processing. Short-term data reflects the immediate state, while long-term trend data is used for predictive maintenance assessment, forming a complete equipment health status assessment system.

[0084] In one embodiment, Example 5: The charging space control unit continuously receives comprehensive temperature evaluation indicators from the analysis module. This unit has multiple preset temperature safety thresholds, forming a stepped response mechanism. When the evaluation indicator first exceeds the first-level threshold, the unit generates a power adjustment command for the cooling system, gradually increasing the setpoint speed of the cooling fan. This adjustment process uses a PID control algorithm to ensure the stability of temperature changes. If the indicator continues to rise and exceeds the second-level threshold, the unit will activate the auxiliary cooling device and simultaneously reduce the output power ratio of the charging pile to curb the temperature rise by reducing heat sources. All control commands are sent to the thermal management system actuator of the charging pile via the Modbus communication protocol.

[0085] The pile structure control unit operates based on real-time monitoring of component deviation coefficients. The unit's preset health threshold range is divided into three levels: normal, attention, and warning. When the deviation coefficient is in the attention range, the unit generates a mechanical structure tightening operation prompt, which is displayed to maintenance personnel via the human-machine interface, suggesting preventative tightening checks on connectors in designated areas. When the coefficient enters the warning range, the unit automatically triggers a high-frequency vibration suppression algorithm, adjusting the switching frequency of the power module to avoid mechanical resonance, and simultaneously sends an emergency maintenance request to the monitoring center. During the control process, the unit records the time and effect of all operations, forming a log of the equipment's health status evolution.

[0086] The output interface control unit employs a risk level mapping mechanism to process the interface safety risk index. The unit internally stores a preset risk level scale, mapping a risk index of 0-100 to five risk levels. When the risk index is at a low risk level, the unit only records interface status data without taking active intervention. Upon reaching a medium risk level, the unit generates a charging current limit command, gradually reducing the maximum allowable charging current. Upon entering a high risk level, the unit immediately triggers a charging interface power-off protection operation. This operation involves several sequentially executed steps: first, sending a charging termination signal to the vehicle; second, controlling the contactor to disconnect the DC power supply circuit; and finally, activating the grounding switch to ensure safe discharge of the equipment.

[0087] The three control units share information and coordinate their operations through a data exchange channel, which uses a publish-subscribe model to transmit associated matching parameters. The charging space control unit periodically publishes comprehensive temperature assessment indicators and their changing trends, while the pile structure control unit subscribes to this data to determine whether abnormal vibrations are related to temperature changes. The output interface control unit publishes real-time values ​​of the interface safety risk index for other units to assess the impact of electrical conditions on the mechanical cooling system. The central scheduler manages the execution priority of all control commands, and the scheduling algorithm dynamically adjusts the command queue based on multi-dimensional state assessment results. When both a cooling system power adjustment command and a charging interface power-off protection operation request are received simultaneously, the scheduler prioritizes the power-off protection operation because electrical safety takes precedence over thermal management requirements. The scheduler also has a conflict detection function; when it identifies conflicting commands from different units (such as simultaneously increasing and decreasing fan speed), it automatically enters a coordination decision-making mode, selecting the optimal control strategy based on the overall equipment state assessment results.

[0088] All control operations are equipped with a dual confirmation mechanism. Important commands must simultaneously meet automatic evaluation conditions and require manual confirmation (or default confirmation upon timeout) before execution, ensuring the reliability of system actions. The control module also has operation rollback capabilities. When it detects that the equipment status has not improved or has deteriorated after the execution of a control command, it can automatically cancel the current operation and try alternative solutions. The entire control process forms a closed-loop feedback loop. The system continuously monitors changes in the equipment status after the execution of control commands and dynamically adjusts subsequent control strategies, achieving truly adaptive intelligent control.

[0089] The above description is merely a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A multi-dimensional information intelligent detection system based on DC charging piles, characterized in that, include: The charging pile multi-dimensional information acquisition module is used to acquire the current charging space environment data in the charging space dimension, the current pile structure operation data in the pile structure dimension, and the current output interface status data in the output interface dimension. The charging pile multi-dimensional information analysis module is used to perform multi-level evaluation and analysis on the current charging space environment data in the charging space dimension, the current pile structure operation data in the pile structure dimension, and the current output interface status data in the output interface dimension. The charging pile status control module is used to perform charging pile operation parameter adjustment operations based on multi-level evaluation and analysis results; The current charging space environment data in the charging space dimension includes the current ambient temperature value, the current ambient humidity value, the current ambient dust concentration value, and the current charging pile radiator surface temperature data; The current operational data of the charging pile structure includes the current temperature value of the core components inside the charging pile, the current vibration frequency value of the charging pile, and the current noise decibel value of the charging pile. The current output interface status data at the output interface level includes the current charging interface contact resistance value and the current charging interface wear data. The current surface temperature data of the charging pile heat sink specifically includes the highest surface temperature value of the heat sink and the surface temperature distribution gradient value of the heat sink. The current wear data of the charging interface specifically includes the deformation of the metal spring of the charging interface and the area value of the crack in the insulation layer of the charging interface. The multi-dimensional information analysis module for the charging pile includes: The charging space temperature assessment unit is used to normalize the current ambient temperature value and the current ambient dust concentration value, and perform weighted fusion analysis by combining the highest temperature value of the radiator surface and the temperature distribution gradient value of the radiator surface to generate a comprehensive assessment index of the charging space temperature. The structural health assessment unit for charging piles is used to discretize the temperature value of the core components inside the current charging pile, the vibration frequency value of the current charging pile, and the noise decibel value of the current charging pile, and to calculate the component operation deviation coefficient. The output interface dimension safety assessment unit is used to standardize the current charging interface contact resistance value and perform multi-factor coupling analysis by combining the deformation of the charging interface metal spring and the crack area value of the charging interface insulation layer. The specific execution of the charging space dimension temperature assessment unit is as follows: Based on the current radiator surface temperature data of the charging pile obtained by the charging space dimension acquisition module, the highest radiator surface temperature value is standardized and converted. Calculate the coefficient of variation of the temperature distribution gradient on the surface of the radiator; The highest surface temperature value of the radiator after standardization and transformation is superimposed with the coefficient of variation. Temperature compensation and correction are performed based on the current ambient temperature value; The heat dissipation efficiency attenuation coefficient is calculated by integrating the current environmental dust concentration value; The comprehensive evaluation index of the charging space temperature is generated by integrating the heat dissipation efficiency attenuation coefficient and the temperature compensation correction results.

2. The intelligent detection system based on multi-dimensional information of DC charging piles according to claim 1, characterized in that, The specific implementation of the pile structure health assessment unit is as follows: The current vibration frequency value of the charging pile obtained by the pile structure dimension acquisition module is subjected to frequency domain transformation processing; Extract characteristic harmonic components from the frequency domain transformation result; Perform time-domain slicing on the current charging pile noise decibel value; Calculate the variance coefficient of the time-domain slice processing result; Correlation matching between characteristic harmonic components and variance coefficients; The thermal coupling effect is calculated by combining the temperature values ​​of the core components inside the current charging pile; The component's operational deviation coefficient is generated based on the calculation results of the thermo-coupling effect.

3. The intelligent detection system based on multi-dimensional information of DC charging piles according to claim 2, characterized in that, The specific execution of the output interface dimension security assessment unit is as follows: The current charging interface contact resistance value obtained by the output interface dimension acquisition module is logarithmically transformed. Perform geometric topological analysis on the deformation of the metal spring contacts of the charging interface; Calculate the rate of expansion of the cracked area value of the charging interface insulation layer; The geometric topology analysis results are temporally correlated with the expansion rate; The interface security risk index is generated by integrating the results of logarithmic transformation and time-series correlation processing.

4. The intelligent detection system based on multi-dimensional information of DC charging piles according to claim 3, characterized in that, The charging pile status control module includes: The charging space control unit is used to compare and analyze the comprehensive evaluation index of the charging space temperature with the preset temperature safety threshold, and trigger the power adjustment command of the heat dissipation system based on the comparison and analysis results. The pile structure control unit is used to calculate the deviation between the component operation deviation coefficient and the preset health threshold range, and trigger the mechanical structure fastening operation based on the deviation calculation result; The output interface control unit is used to map and match the interface safety risk index with a preset risk level scale, and trigger the charging interface power-off protection operation based on the mapping and matching result.

5. The intelligent detection system based on multi-dimensional information of DC charging piles according to claim 4, characterized in that, A data interaction channel is set between the charging space control unit, the pile structure control unit, and the output interface control unit. The data interaction channel is used to transmit the correlation matching parameters of the comprehensive evaluation index of charging space temperature, component operation deviation coefficient and interface safety risk index. The associated matching parameters are used to coordinate the execution priority of the heat dissipation system power adjustment command, the mechanical structure fastening operation, and the charging interface power failure protection operation.

6. A method for intelligent detection based on multi-dimensional information of DC charging piles, characterized in that, It includes all modules and method flows of the intelligent detection system based on multi-dimensional information of DC charging piles as described in any one of claims 1 to 5.

Images