A wall-mounted charging pile monitoring method and system with voice interaction function

Abstract

The present invention relates to the field of charging pile monitoring technology, and discloses a wall-mounted charging pile monitoring method and system with voice interaction function. The system includes: an acquisition module that determines a charging vehicle based on a vehicle image; a judgment module that determines whether to obtain a thermal image of the charging vehicle based on input power and charging power; a processing module that extracts thermal pixels from the thermal image, determines a thermal comprehensive value of the charging vehicle based on thermal feature points, compares environmental data with qualified environmental data, determines an environmental coefficient based on the comparison result and an environmental coefficient model, adjusts the thermal comprehensive value based on the environmental coefficient, and an early warning module that determines whether the charging vehicle has an abnormality based on a target thermal comprehensive value, and determines the abnormality level of the charging vehicle based on the judgment result. The present invention determines the target thermal comprehensive value based on the thermal feature points and the environmental coefficient model, thereby improving the reliability of monitoring.

Description

Technical Field

[0001] The present invention relates to the technical field of charging pile monitoring, and in particular to a wall-mounted charging pile monitoring method and system with a voice interaction function. Background Art

[0002] New energy vehicles are generally equipped with large-capacity batteries to meet the growing demand for battery life. However, during charging, these batteries can heat up and pose a fire hazard. Existing wall-mounted chargers with voice interaction functions operate unattended, making it impossible to effectively monitor and provide early warnings if a vehicle overheats during charging.

[0003] Therefore, how to provide a wall-mounted charging pile monitoring method and system with voice interaction function is a technical problem that technicians in this field urgently need to solve. Summary of the Invention

[0004] In view of this, the present invention proposes a wall-mounted charging pile monitoring method and system with voice interaction function, aiming to solve the problem of being unable to effectively monitor and warn once a vehicle overheats during charging.

[0005] In one aspect, the present invention provides a wall-mounted charging pile monitoring system with voice interaction function, comprising:

[0006] an acquisition module configured to acquire a vehicle image, environmental data, and a charging power of a charging pile, and determine a charging vehicle based on the vehicle image;

[0007] a determination module configured to search for the input power of the charging vehicle in a preset vehicle lookup table, and determine whether to acquire a thermal image of the charging vehicle based on the input power and the charging power;

[0008] a processing module configured to, when acquiring a thermal image of the charging vehicle, extract thermal pixels from the thermal image, analyze all the thermal pixels to determine thermal feature points, determine a thermal comprehensive value of the charging vehicle based on the thermal feature points, compare the environmental data with qualified environmental data, determine an environmental coefficient based on the comparison result and an environmental coefficient model, adjust the thermal comprehensive value based on the environmental coefficient, and determine a target thermal comprehensive value;

[0009] The early warning module is configured to judge whether there is an abnormality in the charging vehicle according to the target thermal sensitivity comprehensive value, and determine the abnormality level of the charging vehicle according to the judgment result.

[0010] Furthermore, when determining the charging vehicle according to the vehicle image, the method includes:

[0011] The acquisition module performs image processing on the vehicle image, wherein the image processing includes image denoising, contrast adjustment, and pixel value normalization, and determines the target vehicle image according to the result of the image processing;

[0012] Key points of the front logo, headlights, air intake grille and rear logo of the target vehicle image are annotated, and scale-invariant feature transformation is used to extract features from the annotated key points to determine the charging vehicle.

[0013] Furthermore, searching the input power of the charging vehicle in a preset vehicle lookup table and determining whether to acquire the thermal image of the charging vehicle based on the input power and the charging power include:

[0014] The preset vehicle lookup table includes a mapping relationship between charging vehicles and corresponding input powers;

[0015] When the input power and the charging power are not equal, the judgment module determines to acquire a thermal image of the charging vehicle;

[0016] When the input power is equal to the charging power, the judgment module determines not to acquire the thermal image of the charging vehicle.

[0017] Furthermore, when extracting the thermal pixels of the thermal image, analyzing all the thermal pixels to determine thermal feature points, and determining the thermal comprehensive value of the charging vehicle based on the thermal feature points, the method includes:

[0018] The processing module converts all thermal pixel points into thermal coordinate points, and establishes a thermal coordinate system based on all thermal coordinate points;

[0019] All thermal coordinate points are fitted according to the least squares method to determine the thermal fitting curve.

[0020] Furthermore, when all thermal pixels are converted into thermal coordinate points and a thermal coordinate system is established based on all thermal coordinate points, the following steps are included:

[0021] The processing module uses the extraction time as the X-axis coordinate value of the thermal pixel point, obtains a thermal sensitivity index of the thermal pixel point, establishes a thermal sensitivity index set based on the thermal sensitivity index, and obtains a reference thermal sensitivity index set corresponding to the thermal sensitivity index set;

[0022] comparing the thermal sensation index set with the benchmark thermal sensation index set;

[0023] When the thermal sensation index in the thermal sensation index set is greater than the reference thermal sensation index in the reference thermal sensation index set, constructing the thermal sensation index greater than the reference thermal sensation index into a first thermal sensation set;

[0024] When the thermal sensation index in the thermal sensation index set is equal to the reference thermal sensation index in the reference thermal sensation index set, constructing the thermal sensation index equal to the reference thermal sensation index into a second thermal sensation set;

[0025] When the thermal sensation index in the thermal sensation index set is smaller than the reference thermal sensation index in the reference thermal sensation index set, constructing the thermal sensation index smaller than the reference thermal sensation index into a third thermal sensation set;

[0026] Calculate the Y-axis coordinate value of the thermal pixel point according to the first thermal sensing set, the second thermal sensing set, and the third thermal sensing set;

[0027] The thermal coordinate system is established according to the X-axis coordinate value and the Y-axis coordinate value.

[0028] Furthermore, when extracting the thermal pixels of the thermal image, analyzing all the thermal pixels to determine thermal feature points, and determining the thermal comprehensive value of the charging vehicle based on the thermal feature points, the method further includes:

[0029] The processing module removes the unfitted thermal coordinate points and determines the thermal coordinate points on the thermal fitting curve as the thermal feature points;

[0030] The thermal sensing temperature values ​​of all thermal sensing feature points are obtained, and the thermal sensing comprehensive value is determined according to the thermal sensing temperature values.

[0031] Furthermore, when comparing the environmental data with qualified environmental data, determining an environmental coefficient according to the comparison result and an environmental coefficient model, adjusting the thermal sensitivity comprehensive value based on the environmental coefficient, and determining a target thermal sensitivity comprehensive value, the method includes:

[0032] When the environmental data and the qualified environmental data are not equal, the processing module determines the environmental coefficient according to the environmental coefficient model, and the target thermal sensitivity comprehensive value is the product of the environmental coefficient and the thermal sensitivity comprehensive value;

[0033] When the environmental data and the qualified environmental data are equal, the processing module determines the thermal sensation comprehensive value as the target thermal sensation comprehensive value.

[0034] Furthermore, when the processing module determines the environmental coefficient according to the environmental coefficient model, it includes:

[0035] The processing module obtains an environmental data set, divides the environmental data set into a training set and a test set, trains a data model using the training set, and tests the trained data model using the test set, and ultimately determines the environmental coefficient model whose input is environmental data and output is environmental coefficient;

[0036] The data model includes a BP neural network model and a RBF neural network model.

[0037] Furthermore, when judging whether the charging vehicle has an abnormality according to the target thermal sensitivity comprehensive value, and determining the abnormality level of the charging vehicle according to the judgment result, the method includes:

[0038] A first preset target thermal sensation comprehensive value and a second preset target thermal sensation comprehensive value are preset, and the first preset target thermal sensation comprehensive value is greater than the second preset target thermal sensation comprehensive value;

[0039] A first preset abnormality level, a second preset abnormality level, and a third preset abnormality level are preset, and the urgency of the first preset abnormality level, the second preset abnormality level, and the third preset abnormality level decreases in sequence;

[0040] When the target thermal sensitivity comprehensive value is less than the second preset target thermal sensitivity comprehensive value, the early warning module determines that there is no abnormality in the charging vehicle;

[0041] When the target thermal sensitivity comprehensive value is greater than or equal to the second preset target thermal sensitivity comprehensive value, the early warning module determines that there is an abnormality in the charging vehicle;

[0042] When the target thermal sensitivity comprehensive value is equal to the second preset target thermal sensitivity comprehensive value, the abnormality level of the charging vehicle is determined to be the third preset abnormality level;

[0043] When the target thermal sense comprehensive value is greater than the second preset target thermal sense comprehensive value and less than or equal to the first preset target thermal sense comprehensive value, the abnormality level of the charging vehicle is determined to be the second preset abnormality level;

[0044] When the target thermal sensation comprehensive value is greater than the first preset target thermal sensation comprehensive value, the abnormality level of the charging vehicle is determined to be the first preset abnormality level.

[0045] Compared with the prior art, the beneficial effects of the present invention are: the input power of the charging vehicle is accurately locked from the preset vehicle lookup table and compared with the charging power, which can preliminarily determine the condition of the charging vehicle and obtain thermal images in a targeted manner, avoiding blind monitoring of the charging vehicle, thereby improving monitoring efficiency and enabling reasonable allocation of monitoring resources. By extracting thermal pixel points to determine thermal feature points and calculating thermal comprehensive values, while comparing environmental data with qualified environmental data, the environmental coefficient model is used to comprehensively measure the impact of environmental factors to determine the target thermal comprehensive value, fully considering the complexity of the charging environment, ensuring that the monitoring results fit the actual thermal conditions of the charging vehicle, and judging whether the charging vehicle has abnormalities based on the target thermal comprehensive value, it makes up for the shortcomings of unmanned wall-mounted charging pile monitoring, effectively reduces accident losses, and improves the reliability and stability of the system in new energy vehicle charging safety monitoring.

[0046] On the other hand, the present application also provides a wall-mounted charging pile monitoring method with voice interaction function, which is used to apply the above-mentioned wall-mounted charging pile monitoring system with voice interaction function, including:

[0047] Acquire a vehicle image, environmental data, and the charging power of a charging pile, and determine the charging vehicle based on the vehicle image;

[0048] searching for the input power of the charging vehicle in a preset vehicle lookup table, and determining whether to acquire a thermal image of the charging vehicle based on the input power and the charging power;

[0049] When acquiring a thermal image of the charging vehicle, extracting thermal pixels from the thermal image, analyzing all of the thermal pixels to determine thermal feature points, determining a thermal comprehensive value of the charging vehicle based on the thermal feature points, comparing the environmental data with qualified environmental data, determining an environmental coefficient based on the comparison result and an environmental coefficient model, adjusting the thermal comprehensive value based on the environmental coefficient, and determining a target thermal comprehensive value;

[0050] It is determined whether the charging vehicle has an abnormality according to the target thermal sensitivity comprehensive value, and the abnormality level of the charging vehicle is determined according to the determination result.

[0051] It is understandable that the above-mentioned wall-mounted charging pile monitoring method and system with voice interaction function have the same beneficial effects and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present invention. The same reference symbols are used throughout the drawings to represent the same components. In the drawings:

[0053] Figure 1 This is a functional block diagram of a wall-mounted charging pile monitoring system with voice interaction function provided by an embodiment of the present invention;

[0054] Figure 2 This is a flow chart of a method for monitoring a wall-mounted charging pile with voice interaction function provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0055] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art. It should be noted that, unless there is a conflict, the embodiments of the present disclosure and the features in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.

[0056] See Figure 1 As shown, in some embodiments of the present application, this embodiment provides a wall-mounted charging pile monitoring system with voice interaction function, including:

[0057] The acquisition module is configured to obtain vehicle images, environmental data and the charging power of the charging pile, and determine the charging vehicle based on the vehicle image.

[0058] The judgment module is configured to search for the input power of the charging vehicle in a preset vehicle search table, and judge whether to obtain a thermal image of the charging vehicle according to the input power and the charging power.

[0059] The processing module is configured to, when acquiring a thermal image of a charging vehicle, extract thermal pixels of the thermal image, analyze all the thermal pixels to determine thermal feature points, determine a thermal comprehensive value of the charging vehicle based on the thermal feature points, compare the environmental data with qualified environmental data, determine the environmental coefficient based on the comparison result and the environmental coefficient model, adjust the thermal comprehensive value based on the environmental coefficient, and determine a target thermal comprehensive value.

[0060] The early warning module is configured to judge whether there is any abnormality in the charging vehicle according to the target thermal sensitivity comprehensive value, and determine the abnormality level of the charging vehicle according to the judgment result.

[0061] Specifically, the acquisition module captures vehicle images using a camera and other image acquisition devices, and uses temperature, humidity, and pressure sensors to obtain environmental data. The power collector then obtains charging power from the charging station. Using recognition technology, the module identifies the model of the charging vehicle based on features such as the vehicle's outline and brand logo. Battery characteristics and safety parameters vary between different vehicle models, making accurate identification essential for effective monitoring and laying the foundation for subsequent targeted monitoring. The judgment module searches a pre-set vehicle lookup table for the vehicle's input power and compares the current charging power with the input power in the lookup table. A discrepancy indicates a potential battery charge state deviation, potentially posing a risk of overheating. This triggers the acquisition of a thermal image. The relatively stable power level during normal battery charging allows for rapid determination of the vehicle's condition, avoiding the waste of resources associated with continuous thermal image acquisition. Thermal images are captured by a thermal imaging sensor and then processed by the processing module. In the real-world charging station environment, the charging vehicle may be surrounded by various reflective sources, such as walls, the ground, and other vehicles. When the thermal imaging sensor captures the thermal image, these reflected infrared radiation is also captured. For example, when sunlight strikes a nearby metal object and reflects infrared radiation, it enters the thermal imaging sensor, creating some false thermal pixels in the image. These pixels do not originate from the charging vehicle itself and do not represent the vehicle's true thermal state, making them invalid. Therefore, the processing module extracts and analyzes all thermal pixels in the thermal image, identifying thermal feature points. These feature points represent valid thermal pixels in the thermal image. This analysis calculates a thermal composite value, which reflects the overall thermal state of the charging vehicle. The environmental data is then compared with qualified environmental data. An environmental coefficient is determined based on an environmental coefficient model (a model that considers the impact of different environmental factors on battery heating). For example, high temperatures accelerate battery heating. The thermal composite value is adjusted based on the environmental coefficient to obtain a target thermal composite value, ensuring that the result more accurately reflects the actual charging vehicle's condition. The warning module then uses the target thermal composite value to determine whether the charging vehicle has an anomaly, assign a corresponding anomaly level, and issue a corresponding voice alert. When the emergency level is high, the voice reminder uses a sharp, loud tone and directly issues a serious warning, allowing the driver and surrounding people to take timely action. When the emergency level is low, the voice reminder uses a calmer tone to guide the driver's attention. Once the warning module determines the emergency level, the voice reminder can be issued within a short time.Whether the vehicle suddenly encounters an abnormality during charging or the abnormality level changes during monitoring, voice reminders can follow up in real time to effectively ensure the safety of the charging process. In addition, considering the needs of different user groups, the voice reminder has a multi-language switching function. When facing car owners with different language backgrounds, the system provides voice reminders in Chinese, English, Japanese and other languages ​​to ensure that every car owner can clearly understand the reminder content so that they can take timely measures, effectively reducing accident losses, thereby making up for the real-time and reliability of monitoring in unattended mode.

[0062] It is understandable that the system not only pays attention to the temperature of the charging vehicle, but also adjusts the comprehensive thermal sensitivity value based on environmental data, taking into account the impact of the environment on battery heating and thus the temperature of the charging vehicle, making the monitoring results more in line with reality and improving the reliability and stability of monitoring.

[0063] In some embodiments of the present application, when determining a charging vehicle based on a vehicle image, it includes: an acquisition module performing image processing on the vehicle image, the image processing includes image denoising, adjusting contrast and normalizing pixel values, determining the target vehicle image based on the image processing results, annotating key points of the front logo, headlights, air intake grille and rear logo of the target vehicle image, and using scale-invariant feature transformation to extract features from the annotated key points to determine the charging vehicle.

[0064] Specifically, the collected vehicle images will produce noise due to factors such as environmental electromagnetic interference. These noises will blur the features of the vehicle. Image denoising can remove these interferences, making the outline and details of the vehicle clear, providing clean and accurate data for subsequent key point annotation and feature extraction. Adjusting the contrast improves the recognizability of information in the image. The images collected by cameras of different brands and models have different pixel value ranges and distributions. Normalized pixel values ​​can unify the pixel values ​​of various images to a standard range, eliminating the impact of equipment differences on image analysis, improving the stability and accuracy of identifying charging vehicles in different scenarios, and enhancing the system's adaptability to various environments. The front logo is an intuitive reflection of the vehicle brand. Different brand logos have unique shapes and styles. The shape, position and arrangement of the headlights vary depending on the model. The style of the air intake grille can reflect the style and model of the vehicle, and the rear logo contains specific information about the model. By marking the key points of these parts, the core features of the charging vehicle are accurately locked, providing a basis for vehicle model identification. The scale-invariant feature transform (SIFT) feature extraction has scale invariance and rotation invariance. The SIFT algorithm can stably extract feature descriptors of key points under complex conditions, allowing the system to accurately identify charging vehicles in complex scenarios such as densely populated parking lots and various parking angles, further enhancing the system's adaptability and laying the foundation for subsequent determination of input power.

[0065] In some embodiments of the present application, when searching for the input power of the charging vehicle in a preset vehicle lookup table and determining whether to obtain a thermal image of the charging vehicle based on the input power and the charging power, the method includes: the preset vehicle lookup table includes a mapping relationship between the charging vehicle and the corresponding input power; when the input power and the charging power are not equal, the judgment module determines to obtain a thermal image of the charging vehicle; when the input power and the charging power are equal, the judgment module determines not to obtain a thermal image of the charging vehicle.

[0066] Specifically, the corresponding input power is searched in the vehicle lookup table according to the obtained charging vehicle. The preset vehicle lookup table is obtained according to the vehicle operating instructions of different new energy vehicles. The acquisition and processing of thermal images requires a certain amount of computing resources and time. If thermal images are continuously acquired for charging vehicles, it will cause a waste of resources, especially when there are a large number of vehicles at the charging station. By comparing the input power and charging power, limited resources can be accurately invested in monitoring of possible overheating risks, avoiding unnecessary thermal image acquisition of normally charging vehicles, thereby improving the utilization efficiency of system resources and ensuring the stable operation of the system.

[0067] In some embodiments of the present application, when extracting thermal pixel points of a thermal image, analyzing all thermal pixel points to determine thermal feature points, and determining the thermal comprehensive value of the charging vehicle based on the thermal feature points, it includes: a processing module converting all thermal pixel points into thermal coordinate points, establishing a thermal coordinate system based on all thermal coordinate points, fitting all thermal coordinate points according to the least squares method, and determining a thermal fitting curve.

[0068] Specifically, the thermal pixel points are converted into thermal coordinate points, and then a thermal coordinate system is constructed, which gives a clear structure to the complex thermal data. In the thermal coordinate system, they can be arranged and analyzed in an orderly manner. The thermal coordinate points are fitted using the least squares method, and a large amount of discrete data is simplified into a thermal fitting curve. The process from thermal pixel points to thermal coordinate system and then to thermal fitting curve simplifies the complexity of data processing and improves computing efficiency, allowing the system to complete the analysis of a large amount of thermal data in a short time. In addition, invalid thermal pixel points are removed by fitting, thereby comprehensively considering the distribution pattern of thermal coordinate points, and then quantifying the dynamic changes in the thermal state of the charging vehicle, and comprehensively and in-depth considering the overall trend and change details of the thermal distribution. It provides data support for the subsequent early warning module to accurately determine whether the vehicle is abnormal and the abnormality level classification, thereby ensuring the stability and reliability of monitoring.

[0069] In some embodiments of the present application, when all the thermal pixels are converted into thermal coordinate points and a thermal coordinate system is established based on all the thermal coordinate points, the processing module uses the extraction time as the X-axis coordinate value of the thermal pixel point, obtains the thermal index of the thermal pixel point, establishes a thermal index set based on the thermal index, obtains a benchmark thermal index set corresponding to the thermal index set, compares the thermal index set with the benchmark thermal index set, and when the thermal index in the thermal index set is greater than the benchmark thermal index in the benchmark thermal index set, the thermal index set greater than the benchmark thermal index is converted into the thermal coordinate system. The thermal sensitivity indicators of the indicators are constructed into a first thermal sensitivity set. When the thermal sensitivity indicators in the thermal sensitivity indicator set are equal to the benchmark thermal sensitivity indicators in the benchmark thermal sensitivity indicator set, the thermal sensitivity indicators equal to the benchmark thermal sensitivity indicators are constructed into a second thermal sensitivity set. When the thermal sensitivity indicators in the thermal sensitivity indicator set are less than the benchmark thermal sensitivity indicators in the benchmark thermal sensitivity indicator set, the thermal sensitivity indicators less than the benchmark thermal sensitivity indicators are constructed into a third thermal sensitivity set. The Y-axis coordinate values ​​of the thermal sensitivity pixels are calculated based on the first thermal sensitivity set, the second thermal sensitivity set, and the third thermal sensitivity set, and a thermal sensitivity coordinate system is established based on the X-axis coordinate values ​​and the Y-axis coordinate values.

[0070] Specifically, the Y-axis coordinate value is obtained by the following formula:

[0071] ;

[0072] Wherein, Y is the Y-axis coordinate value of the thermal pixel point, n represents the number of thermal sensitivity indicators in the first thermal sensitivity set, Fi represents the i-th thermal sensitivity indicator in the first thermal sensitivity set, Di represents the benchmark thermal sensitivity indicator corresponding to the i-th thermal sensitivity indicator, m represents the number of thermal sensitivity indicators in the third thermal sensitivity set, Rj represents the benchmark thermal sensitivity indicator corresponding to the j-th thermal sensitivity indicator, and Kj represents the j-th thermal sensitivity indicator in the third thermal sensitivity set.

[0073] Specifically, the extraction time is 1 second, 1.5 seconds, and 3 seconds, which is an extracted arithmetic progression, indicating the extraction time interval for each thermal pixel. A thermal pixel is extracted every 1.5 seconds. Thermal indicators include thermal brightness and thermal saturation. In addition, the benchmark thermal indicators and thermal indicators in the benchmark thermal indicator set correspond one to one. By comparing the thermal indicators of the thermal pixels with the benchmark thermal indicators and constructing different thermal sets accordingly, the difference between each thermal indicator and the benchmark thermal indicator can be clearly represented. The extraction time is used as the X-axis coordinate value of the thermal pixel, which closely associates the thermal data with the time dimension. The Y-axis coordinate value determined by the thermal indicator is used to establish a thermal coordinate system that can fully reflect the dynamic changes of the thermal data over time, thereby improving the accuracy of thermal image analysis and enhancing the reliability of the entire monitoring system.

[0074] In some embodiments of the present application, when extracting thermal pixel points of a thermal image, analyzing all thermal pixel points to determine thermal feature points, and determining the thermal comprehensive value of the charging vehicle based on the thermal feature points, it also includes: a processing module removes unfitted thermal coordinate points, determines the thermal coordinate points on the thermal fitting curve as thermal feature points, obtains the thermal temperature values ​​of all thermal feature points, and determines the thermal comprehensive value based on the thermal temperature values.

[0075] Specifically, the average thermal temperature of all thermal feature points is determined as the thermal comprehensive value. The thermal temperature values ​​of individual abnormal thermal pixel points will not cause excessive interference to the thermal temperature average value, so that the thermal comprehensive value can remain relatively stable when facing complex and changeable thermal image data, reducing the situation where the thermal comprehensive value fluctuates greatly due to accidental factors, improving the reliability and stability of the monitoring results, and helping to accurately determine whether there are abnormalities in the charging vehicle. In addition, removing the thermal coordinate points that are not fitted can screen out abnormal data in the thermal image that may be caused by noise interference, environmental reflection and other factors. These thermal coordinate points may deviate from the actual thermal distribution pattern of the charging vehicle. If included in the calculation, it will affect the accuracy of the thermal comprehensive value. Only the thermal coordinate points on the thermal fitting curve are retained as thermal feature points, ensuring that the data used to calculate the thermal comprehensive value can represent the thermal characteristics of the charging vehicle, so that the thermal comprehensive value accurately reflects the actual thermal state of the charging vehicle, and improves the accuracy and reliability of the system monitoring.

[0076] In some embodiments of the present application, when comparing environmental data with qualified environmental data, determining the environmental coefficient based on the comparison result and the environmental coefficient model, adjusting the thermal sensation comprehensive value based on the environmental coefficient, and determining the target thermal sensation comprehensive value, it includes: when the environmental data and the qualified environmental data are not equal, the processing module determines the environmental coefficient based on the environmental coefficient model, and the target thermal sensation comprehensive value is the product of the environmental coefficient and the thermal sensation comprehensive value; when the environmental data and the qualified environmental data are equal, the processing module determines the thermal sensation comprehensive value as the target thermal sensation comprehensive value.

[0077] In some embodiments of the present application, when the processing module determines the environmental coefficient based on the environmental coefficient model, it includes: the processing module obtains the environmental data set, and divides the environmental data set into a training set and a test set, uses the training set to train the data model, and uses the test set to test the trained data model, and finally determines the environmental coefficient model whose input is environmental data and output is the environmental coefficient, and the data model includes a BP neural network model and an RBF neural network model.

[0078] Specifically, environmental data includes data such as ambient temperature, humidity, and air pressure. Qualified environmental data is the standard data for the environmental requirements of charging piles. When the environmental data and qualified environmental data differ, it indicates that the current environment affects the thermal sensitivity value. For example, high temperatures increase the thermal sensitivity value, while high humidity decreases it. The environmental coefficient is determined based on the environmental coefficient model, ensuring that the monitoring results closely reflect actual environmental variations and avoiding misjudgments caused by ignoring environmental factors. When the environmental data and qualified environmental data are equal, the thermal sensitivity value is directly determined as the target thermal sensitivity value. Under ideal and stable environmental conditions, unnecessary calculations are reduced, enabling the system to process data and output results under different environmental conditions, improving the adaptability of the monitoring system. The environmental dataset includes key data such as air pressure, temperature, and humidity. This data records the operating conditions of charging piles and charging vehicles over different periods of time. The environmental dataset is divided into training and test sets. Typically, 70%-80% of the data is used as the training set, and the remainder as the test set. Ensuring that both the training and test sets contain data from a variety of operating conditions improves the model's generalization ability. By training and testing BP and RBF neural network models, the resulting environmental coefficient model achieves high accuracy. These neural network models learn and capture relationships between data during the training process. For example, under the combined influence of various environmental factors such as temperature, humidity, and air pressure, the model outputs the corresponding environmental coefficient. Adjusting the thermal sensitivity value based on the environmental coefficient improves the accuracy of the target thermal sensitivity value and effectively avoids false positives and false negatives caused by environmental factors.

[0079] It's understandable that as time goes by and environmental diversity increases, the environmental dataset can be continuously updated, and the environmental coefficient model can be retrained and optimized, allowing the system to adapt to environmental changes in different regions, seasons, and time periods. For example, during periods of high summer temperatures and low winter temperatures, the environmental coefficient model can adjust its output based on the new data, ensuring that the system can stably and accurately monitor the status of charging piles and charging vehicles in various environments.

[0080] In some embodiments of the present application, when judging whether there is an abnormality in the charging vehicle according to the target thermal sense comprehensive value, and determining the abnormality level of the charging vehicle according to the judgment result, it includes: presetting a first preset target thermal sense comprehensive value and a second preset target thermal sense comprehensive value, and the first preset target thermal sense comprehensive value is greater than the second preset target thermal sense comprehensive value, presetting a first preset abnormality level, a second preset abnormality level and a third preset abnormality level, and the urgency of the first preset abnormality level, the second preset abnormality level and the third preset abnormality level decreases in sequence, when the target thermal sense comprehensive value is less than the second preset target thermal sense comprehensive value, the early warning module judges that There is no abnormality in the charging vehicle. When the target thermal sense comprehensive value is greater than or equal to the second preset target thermal sense comprehensive value, the early warning module determines that there is an abnormality in the charging vehicle. When the target thermal sense comprehensive value is equal to the second preset target thermal sense comprehensive value, the abnormality level of the charging vehicle is determined to be the third preset abnormality level. When the target thermal sense comprehensive value is greater than the second preset target thermal sense comprehensive value and less than or equal to the first preset target thermal sense comprehensive value, the abnormality level of the charging vehicle is determined to be the second preset abnormality level. When the target thermal sense comprehensive value is greater than the first preset target thermal sense comprehensive value, the abnormality level of the charging vehicle is determined to be the first preset abnormality level.

[0081] Specifically, by setting the first preset target thermal sensitivity comprehensive value and the second preset target thermal sensitivity comprehensive value, as well as the first preset abnormality level, the second preset abnormality level and the third preset abnormality level, the status of the charging vehicle can be finely graded. When the target thermal sensitivity comprehensive value is in different intervals, it corresponds to abnormality levels of different urgency. For example: when the target thermal sensitivity comprehensive value is greater than the first preset target thermal sensitivity comprehensive value, it is determined to be the first preset abnormality level, indicating that the charging vehicle is in a high-risk overheating state and emergency measures need to be taken immediately. The graded early warning mechanism enables relevant management personnel to quickly and accurately understand the degree of abnormality of the charging vehicle, so as to arrange targeted response strategies and enable resources to be reasonably allocated. It improves the efficiency of handling abnormalities in charging vehicles and improves the stability and reliability of monitoring.

[0082] In summary, the beneficial effects of the present invention are: accurately locking the input power of the charging vehicle from the preset vehicle lookup table and comparing it with the charging power, which can preliminarily determine the condition of the charging vehicle and obtain thermal images in a targeted manner, avoiding blind monitoring of the charging vehicle, thereby improving monitoring efficiency and enabling reasonable allocation of monitoring resources. By extracting thermal pixel points to determine thermal feature points and calculating thermal comprehensive values, while comparing environmental data with qualified environmental data, the environmental coefficient model is used to comprehensively measure the impact of environmental factors to determine the target thermal comprehensive value, fully considering the complexity of the charging environment, ensuring that the monitoring results fit the actual thermal conditions of the charging vehicle, and judging whether the charging vehicle has abnormalities based on the target thermal comprehensive value, it makes up for the shortcomings of unmanned wall-mounted charging pile monitoring, effectively reduces accident losses, and improves the reliability and stability of the system in new energy vehicle charging safety monitoring.

[0083] In another preferred embodiment based on the above embodiment, refer to Figure 2 As shown, this embodiment provides a wall-mounted charging pile monitoring method with voice interaction function, which is used to apply the above-mentioned wall-mounted charging pile monitoring system with voice interaction function, including:

[0084] S100: Acquire a vehicle image, environmental data, and the charging power of a charging pile, and determine a charging vehicle based on the vehicle image.

[0085] S200: Searching for the input power of the charging vehicle in a preset vehicle lookup table, and determining whether to obtain a thermal image of the charging vehicle based on the input power and the charging power.

[0086] S300: When obtaining a thermal image of a charging vehicle, extract the thermal pixels of the thermal image, analyze all the thermal pixels to determine thermal feature points, determine a thermal comprehensive value of the charging vehicle based on the thermal feature points, compare the environmental data with the qualified environmental data, determine the environmental coefficient based on the comparison result and the environmental coefficient model, adjust the thermal comprehensive value based on the environmental coefficient, and determine the target thermal comprehensive value.

[0087] S400: Determine whether the charging vehicle has an abnormality based on the target thermal sensitivity comprehensive value, and determine the abnormality level of the charging vehicle based on the determination result.

[0088] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0089] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0090] These computer program instructions may also be stored in a computer-readable storage device that can direct a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable storage device produce an article of manufacture comprising an instruction device that implements the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0091] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0092] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A wall-mounted charging pile monitoring system with voice interaction function, characterized in that: include: an acquisition module configured to acquire a vehicle image, environmental data, and a charging power of a charging pile, and determine a charging vehicle based on the vehicle image; a determination module configured to search for the input power of the charging vehicle in a preset vehicle lookup table, and determine whether to acquire a thermal image of the charging vehicle based on the input power and the charging power; a processing module configured to, when acquiring a thermal image of the charging vehicle, extract thermal pixels from the thermal image, analyze all the thermal pixels to determine thermal feature points, determine a thermal comprehensive value of the charging vehicle based on the thermal feature points, compare the environmental data with qualified environmental data, determine an environmental coefficient based on the comparison result and an environmental coefficient model, adjust the thermal comprehensive value based on the environmental coefficient, and determine a target thermal comprehensive value; an early warning module configured to determine whether the charging vehicle has an abnormality according to the target thermal sensitivity comprehensive value, and determine an abnormality level of the charging vehicle according to the determination result; Searching for the input power of the charging vehicle in a preset vehicle lookup table, and determining whether to acquire a thermal image of the charging vehicle based on the input power and the charging power, including: The preset vehicle lookup table includes a mapping relationship between charging vehicles and corresponding input powers; When the input power and the charging power are not equal, the judgment module determines to acquire a thermal image of the charging vehicle; When the input power is equal to the charging power, the judgment module determines not to acquire the thermal image of the charging vehicle; Extracting thermal pixels of the thermal image, analyzing all thermal pixels to determine thermal feature points, and determining a thermal comprehensive value of the charging vehicle based on the thermal feature points includes: The processing module converts all thermal pixel points into thermal coordinate points, and establishes a thermal coordinate system based on all thermal coordinate points; All thermal coordinate points are fitted according to the least square method to determine the thermal fitting curve; When converting all thermal pixels into thermal coordinate points and establishing a thermal coordinate system based on all thermal coordinate points, it includes: The processing module uses the extraction time as the X-axis coordinate value of the thermal pixel point, obtains a thermal sensitivity index of the thermal pixel point, establishes a thermal sensitivity index set based on the thermal sensitivity index, and obtains a reference thermal sensitivity index set corresponding to the thermal sensitivity index set; comparing the thermal sensation index set with the benchmark thermal sensation index set; When the thermal sensation index in the thermal sensation index set is greater than the reference thermal sensation index in the reference thermal sensation index set, constructing the thermal sensation index greater than the reference thermal sensation index into a first thermal sensation set; When the thermal sensation index in the thermal sensation index set is equal to the reference thermal sensation index in the reference thermal sensation index set, constructing the thermal sensation index equal to the reference thermal sensation index into a second thermal sensation set; When the thermal sensation index in the thermal sensation index set is smaller than the reference thermal sensation index in the reference thermal sensation index set, constructing the thermal sensation index smaller than the reference thermal sensation index into a third thermal sensation set; Calculate the Y-axis coordinate value of the thermal pixel point according to the first thermal sensing set, the second thermal sensing set, and the third thermal sensing set; Establishing the thermal coordinate system according to the X-axis coordinate value and the Y-axis coordinate value; The Y-axis coordinate value is obtained by the following formula: ； Wherein, Y is the Y-axis coordinate value of the thermal pixel point, n represents the number of thermal indices in the first thermal set, Fi represents the i-th thermal index in the first thermal set, Di represents the reference thermal index corresponding to the i-th thermal index, m represents the number of thermal indices in the third thermal set, Rj represents the reference thermal index corresponding to the j-th thermal index, and Kj represents the j-th thermal index in the third thermal set; When extracting the thermal pixels of the thermal image, analyzing all the thermal pixels to determine thermal feature points, and determining the thermal comprehensive value of the charging vehicle based on the thermal feature points, the method further includes: The processing module removes the unfitted thermal coordinate points and determines the thermal coordinate points on the thermal fitting curve as the thermal feature points; Obtaining thermal temperature values ​​of all thermal characteristic points, and determining the thermal comprehensive value according to the thermal temperature values; When comparing the environmental data with qualified environmental data, determining an environmental coefficient according to the comparison result and an environmental coefficient model, adjusting the thermal sensation comprehensive value based on the environmental coefficient, and determining a target thermal sensation comprehensive value, the method includes: When the environmental data and the qualified environmental data are not equal, the processing module determines the environmental coefficient according to the environmental coefficient model, and the target thermal sensitivity comprehensive value is the product of the environmental coefficient and the thermal sensitivity comprehensive value; When the environmental data and the qualified environmental data are equal, the processing module determines the thermal sensation comprehensive value as the target thermal sensation comprehensive value.

2. The wall-mounted charging pile monitoring system with voice interaction function according to claim 1 is characterized in that: When determining the charging vehicle according to the vehicle image, the method includes: The acquisition module performs image processing on the vehicle image, wherein the image processing includes image denoising, contrast adjustment, and pixel value normalization, and determines the target vehicle image according to the result of the image processing; Key points of the front logo, headlights, air intake grille and rear logo of the target vehicle image are annotated, and scale-invariant feature transformation is used to extract features from the annotated key points to determine the charging vehicle.

3. The wall-mounted charging pile monitoring system with voice interaction function according to claim 2 is characterized in that: When the processing module determines the environmental coefficient according to the environmental coefficient model, it includes: The processing module obtains an environmental data set, divides the environmental data set into a training set and a test set, trains a data model using the training set, and tests the trained data model using the test set, and ultimately determines the environmental coefficient model whose input is environmental data and output is environmental coefficient; The data model includes a BP neural network model and a RBF neural network model.

4. The wall-mounted charging pile monitoring system with voice interaction function according to claim 3 is characterized in that: When judging whether the charging vehicle has an abnormality according to the target thermal sensitivity comprehensive value, and determining the abnormality level of the charging vehicle according to the judgment result, the method includes: A first preset target thermal sensation comprehensive value and a second preset target thermal sensation comprehensive value are preset, and the first preset target thermal sensation comprehensive value is greater than the second preset target thermal sensation comprehensive value; A first preset abnormality level, a second preset abnormality level, and a third preset abnormality level are preset, and the urgency of the first preset abnormality level, the second preset abnormality level, and the third preset abnormality level decreases in sequence; When the target thermal sensitivity comprehensive value is less than the second preset target thermal sensitivity comprehensive value, the early warning module determines that there is no abnormality in the charging vehicle; When the target thermal sensitivity comprehensive value is greater than or equal to the second preset target thermal sensitivity comprehensive value, the early warning module determines that there is an abnormality in the charging vehicle; When the target thermal sensitivity comprehensive value is equal to the second preset target thermal sensitivity comprehensive value, the abnormality level of the charging vehicle is determined to be the third preset abnormality level; When the target thermal sense comprehensive value is greater than the second preset target thermal sense comprehensive value and less than or equal to the first preset target thermal sense comprehensive value, the abnormality level of the charging vehicle is determined to be the second preset abnormality level; When the target thermal sensation comprehensive value is greater than the first preset target thermal sensation comprehensive value, the abnormality level of the charging vehicle is determined to be the first preset abnormality level.

5. A method for monitoring a wall-mounted charging pile with voice interaction function, which is used to apply the wall-mounted charging pile monitoring system with voice interaction function according to any one of claims 1 to 4, characterized in that: include: Acquire a vehicle image, environmental data, and the charging power of a charging pile, and determine the charging vehicle based on the vehicle image; searching for the input power of the charging vehicle in a preset vehicle lookup table, and determining whether to acquire a thermal image of the charging vehicle based on the input power and the charging power; When acquiring a thermal image of the charging vehicle, extracting thermal pixels from the thermal image, analyzing all of the thermal pixels to determine thermal feature points, determining a thermal comprehensive value of the charging vehicle based on the thermal feature points, comparing the environmental data with qualified environmental data, determining an environmental coefficient based on the comparison result and an environmental coefficient model, adjusting the thermal comprehensive value based on the environmental coefficient, and determining a target thermal comprehensive value; It is determined whether the charging vehicle has an abnormality according to the target thermal sensitivity comprehensive value, and the abnormality level of the charging vehicle is determined according to the determination result.

Images