Bridge deck ice and snow removal method, system, terminal and medium based on neural network

Abstract

The bridge ice and snow removing method, system, terminal and medium based on a neural network provided by the application, the method comprises the following steps: collecting temperature data and image data of a bridge surface at a preset frequency by using a data collection device; inputting the image data into a trained neural network model, processing the image data by the neural network model, and outputting a heating level and a temperature threshold corresponding to the image data; comparing the temperature data with the temperature threshold, and if the temperature data is lower than the temperature threshold, turning on a heating device of the bridge surface and heating according to the heating level to remove ice and snow on the bridge surface. By combining the data collection device and the neural network model, the heating device of the bridge surface can be automatically controlled to heat, and the purpose of safely and efficiently removing ice and snow on the bridge is effectively achieved.

Description

Technical Field

[0001] This invention relates to the field of road ice and snow control technology, and in particular to a method, system, terminal and medium for removing ice and snow from bridge surfaces based on neural networks. Background Technology

[0002] The safe passage of bridges faces severe challenges in the face of snow and ice disasters. Traditional methods mainly rely on manual snow removal, chemical de-icing agents, and mechanical snow removal, but these methods all have certain limitations and shortcomings in practical applications.

[0003] While manual snow removal is relatively inexpensive, it is inefficient and requires a large workforce. Furthermore, prolonged outdoor work in extreme cold poses safety risks. Chemical snow-melting agents, although faster, can corrode bridge structures, affecting their lifespan. Melted snow also carries chemicals into the environment, polluting soil, water, and vegetation, thus harming the ecological environment. Mechanical snow removal is relatively efficient, but its effectiveness is significantly reduced on complex bridge structures with limited space. In conclusion, traditional methods for removing snow and ice from bridges are insufficient for safe and efficient operation.

[0004] Therefore, existing technologies have shortcomings and need to be improved and developed. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a method, device, terminal and storage medium for removing ice and snow from bridge surfaces based on neural networks, in order to address the above-mentioned deficiencies of the prior art and solve the problem that the prior art is unable to achieve safe and efficient removal of ice and snow from bridges.

[0006] The technical solution adopted by this invention to solve the technical problem is as follows:

[0007] In a first aspect, embodiments of the present invention provide a method for removing ice and snow from bridge surfaces based on neural networks, the method comprising:

[0008] Temperature and image data of the bridge deck are collected using data acquisition equipment at a preset frequency.

[0009] The image data is input into a trained neural network model, processed by the neural network model, and the heating level and temperature threshold corresponding to the image data are output.

[0010] The temperature data is compared with the temperature threshold. If the temperature data is lower than the temperature threshold, the bridge surface heating device is turned on and heated according to the heating level to remove the ice and snow from the bridge surface.

[0011] In one embodiment, the data acquisition device includes a temperature sensor and a camera, which are evenly arranged on the bridge surface.

[0012] In one embodiment, the neural network model is a convolutional neural network model; the step of inputting the image data into the trained neural network model, processing it through the neural network model, and outputting the heating level and temperature threshold corresponding to the image data includes:

[0013] The image data is input into the input layer of the trained convolutional neural network model, and processed by the input layer to obtain normalized image data.

[0014] The normalized image data is input into the convolutional layer of the convolutional neural network model, and the convolutional layer performs feature extraction on the normalized image data to obtain the first feature data.

[0015] The first feature data is input into the activation layer and transformed by a nonlinear activation function to obtain the second feature data;

[0016] The second feature data is input into the pooling layer of the convolutional neural network, and downsampled by the pooling layer to obtain the third feature data;

[0017] The third feature data is input into the fully connected layer of the convolutional neural network model, and processed by the fully connected layer to obtain the heating level and temperature threshold corresponding to the image data;

[0018] The heating level and temperature threshold corresponding to the image data are output through the output layer of the convolutional neural network model.

[0019] In one implementation, the training steps of the neural network model include:

[0020] Construct an initial neural network model, which is a convolutional neural network model;

[0021] Obtain a training dataset, which includes image data of multiple bridge surfaces;

[0022] The initial neural network model is trained using the training dataset until the loss function converges to a preset value, thus obtaining the trained neural network model.

[0023] In one embodiment, the method further includes:

[0024] If the temperature data is higher than the temperature threshold, the bridge deck heating device is turned off.

[0025] In one embodiment, after activating the bridge deck heating device and heating according to the heating level, the method further includes:

[0026] Obtain the operating duration of the bridge deck heating device;

[0027] If the activation duration exceeds a preset time threshold and the current temperature data is lower than the temperature threshold, an alarm message will be generated.

[0028] The alarm information is sent to the target sanitation workers so that they can carry out manual snow removal.

[0029] In one embodiment, the heating device is a heating mesh, which is uniformly embedded in the bridge surface.

[0030] Secondly, embodiments of the present invention also provide a bridge surface ice and snow removal system based on a neural network, comprising:

[0031] The data acquisition module is used to collect temperature data and image data of the bridge deck at a preset frequency using data acquisition equipment;

[0032] The data processing module is used to input the image data into a trained neural network model, process the image data through the neural network model, and output the heating level and temperature threshold corresponding to the image data.

[0033] The ice and snow removal module is used to compare the temperature data with the temperature threshold. If the temperature data is lower than the temperature threshold, the bridge surface heating device is turned on and heated according to the heating level to remove the ice and snow from the bridge surface.

[0034] Thirdly, embodiments of the present invention also provide a terminal, the terminal comprising: a memory, a processor, and a bridge surface ice and snow removal program based on a neural network stored in the memory and executable on the processor, wherein when the bridge surface ice and snow removal program based on a neural network is executed by the processor, it implements the steps of the bridge surface ice and snow removal method based on a neural network as described above.

[0035] Fourthly, embodiments of the present invention also provide a computer-readable storage medium storing a neural network-based bridge surface ice and snow removal program, which can be executed to implement the steps of the neural network-based bridge surface ice and snow removal method described above.

[0036] The beneficial effects of this invention are as follows: This invention utilizes a data acquisition device to collect temperature data and image data of the bridge surface at a preset frequency; the image data is input into a trained neural network model, which processes the data and outputs the heating level and temperature threshold corresponding to the image data; the temperature data is compared with the temperature threshold, and if the temperature data is lower than the temperature threshold, the bridge surface heating device is activated and heated according to the specified heating level to remove ice and snow from the bridge surface. This invention, by combining a data acquisition device and a neural network model, can automatically control the bridge surface heating device, effectively achieving the goal of safely and efficiently removing ice and snow from bridges. Attached Figure Description

[0037] Figure 1 This is a flowchart of a preferred embodiment of the bridge surface ice and snow removal method based on neural networks in this invention.

[0038] Figure 2 This is a functional principle block diagram of a preferred embodiment of the bridge surface ice and snow removal system based on neural networks in this invention.

[0039] Figure 3 This is a block diagram of the terminal principle of the present invention. Detailed Implementation

[0040] To make the objectives, technical solutions, and advantages of this invention clearer and more explicit, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0041] The safe passage of bridges faces severe challenges in the face of snow and ice disasters. Traditional methods mainly rely on manual snow removal, chemical de-icing agents, and mechanical snow removal, but these methods all have certain limitations and shortcomings in practical applications.

[0042] While manual snow removal is relatively inexpensive, it is inefficient and requires a large workforce. Furthermore, prolonged outdoor work in extreme cold poses safety risks. Chemical snow-melting agents, although faster, can corrode bridge structures, affecting their lifespan. Melted snow also carries chemicals into the environment, polluting soil, water, and vegetation, thus harming the ecosystem. Mechanical snow removal is relatively efficient, but its effectiveness is significantly reduced on complex bridge structures with limited space. In conclusion, traditional bridge de-icing methods are insufficient for safe and efficient snow removal from bridges.

[0043] To address the aforementioned shortcomings of existing technologies, this invention provides a method, system, terminal, and medium for removing ice and snow from bridge surfaces based on a neural network. The method includes: acquiring temperature data and image data of the bridge surface using a data acquisition device at a preset frequency; inputting the image data into a trained neural network model, processing the data, and outputting a heating level and temperature threshold corresponding to the image data; comparing the temperature data with the temperature threshold; if the temperature data is lower than the temperature threshold, activating the bridge surface heating device and heating according to the specified heating level to remove ice and snow from the bridge surface. This invention, by combining a data acquisition device and a neural network model, can automatically control the bridge surface heating device, effectively achieving the goal of safely and efficiently removing ice and snow from bridges.

[0044] Please see Figure 1 The bridge surface ice and snow removal method based on neural networks described in this embodiment of the invention includes the following steps:

[0045] Step S100: Use data acquisition equipment to collect temperature data and image data of the bridge deck at a preset frequency.

[0046] Specifically, temperature data reflects the temperature of the bridge deck, while image data reflects the environmental conditions. By collecting temperature and image data from the bridge deck at a preset frequency using data acquisition equipment, and then processing this data, timely monitoring of ice and snow conditions on the bridge deck can be achieved. This method effectively avoids the need for manual inspections for monitoring ice and snow conditions, enabling dynamic control of the distribution of ice and snow hazards on the bridge deck and improving the speed of response to such hazards. The preset frequency can be set according to actual needs and environmental conditions. For example, in blizzard conditions, the preset frequency can be 5 minutes per instance, and in moderate snow conditions, the preset frequency can be 10 minutes per instance.

[0047] In one implementation, the preset frequency is generated by dynamically adjusting based on historical snow and ice data and weather forecasts.

[0048] Specifically, when snowfall or low temperatures are predicted, the data collection frequency is automatically increased; while when the weather improves or the snow and ice conditions ease, the data collection frequency is reduced.

[0049] In one implementation, the data acquisition device includes a temperature sensor and a camera, both of which are mounted on the bridge deck.

[0050] Specifically, this invention incorporates temperature sensors and cameras on the bridge deck. The cameras and temperature sensors collect data at a preset frequency to monitor the icy and snowy conditions on the bridge deck. The camera's field of view covers the entire bridge deck area, including but not limited to the bridge's driveways, sidewalks, and guardrails on both sides. The camera features high-definition night vision, anti-fog, and waterproof capabilities, ensuring stable operation and capturing clear image information under any weather conditions.

[0051] Please see Figure 1 The bridge surface ice and snow removal method based on neural networks described in this embodiment of the invention includes the following steps:

[0052] Step S200: Input the image data into the trained neural network model, process it through the neural network model, and output the heating level and temperature threshold corresponding to the image data.

[0053] Specifically, after image data is collected at a preset frequency, the image data is input into a trained neural network model, which is then used to generate the heating level and temperature threshold corresponding to the current image data. The neural network model in this invention is a convolutional neural network (CNN). CNNs extract and learn features from input data through convolution operations, enabling tasks such as classification and regression of the input data, and are widely used in image recognition. This invention utilizes a CNN model to effectively learn complex features in images and accurately predict the heating level and temperature threshold corresponding to the image data. The neural network model in this invention can be deployed on a cloud server or a physical server; no limitation is imposed.

[0054] In one implementation, the neural network model is a convolutional neural network model; the step of inputting the image data into the trained neural network model, processing it through the neural network model, and outputting the heating level and temperature threshold corresponding to the image data includes:

[0055] The image data is input into the input layer of the trained convolutional neural network model, and processed by the input layer to obtain normalized image data.

[0056] The normalized image data is input into the convolutional layer of the convolutional neural network model, and the convolutional layer performs feature extraction on the normalized image data to obtain the first feature data.

[0057] The first feature data is input into the activation layer and transformed by a nonlinear activation function to obtain the second feature data;

[0058] The second feature data is input into the pooling layer of the convolutional neural network, and downsampled by the pooling layer to obtain the third feature data;

[0059] The third feature data is input into the fully connected layer of the convolutional neural network model, and processed by the fully connected layer to obtain the heating level and temperature threshold corresponding to the image data;

[0060] The heating level and temperature threshold corresponding to the image data are output through the output layer of the convolutional neural network model.

[0061] Specifically, the Convolutional Neural Network (CNN) model includes an input layer, convolutional layers, activation layers, pooling layers, fully connected layers, and an output layer. The input layer normalizes the input image data to obtain normalized image data. The convolutional layer extracts local features from the image data to obtain the first feature data. The activation layer transforms the first feature data using a non-linear activation function to obtain the second feature data. The pooling layer reduces the spatial dimension of the second feature data, increasing its uniqueness to the image, and obtains the third feature data through downsampling. In the fully connected layer, the third feature data is flattened into a one-dimensional feature vector to obtain the third feature vector. A weighted summation activation function is then applied to the third feature vector to finally obtain the heating level and temperature threshold corresponding to the image data. The heating level and temperature threshold corresponding to the image data are then output through the output layer. In this invention, both the first and second feature data are in the form of feature maps. The second feature data is the activated feature map, which is sparser than the first feature data, removing unnecessary redundant information while retaining important feature information. The third feature data is a pooled set of feature maps, which are smaller in size but contain more abstract and advanced feature information. Furthermore, the heating levels and temperature thresholds generated in this invention can be displayed on the control system interface.

[0062] In one implementation, the image data is input into a trained neural network model, processed by the neural network model, and outputs the heating level and temperature threshold corresponding to the image data, as well as the weather forecast and expected heating plan.

[0063] Specifically, this invention can utilize image data to generate heating levels, temperature thresholds, weather forecasts, and anticipated heating plans. The weather forecasts and anticipated heating plans can be displayed on the control system interface.

[0064] In one implementation, the training steps of the neural network model include:

[0065] Construct an initial neural network model, which is a convolutional neural network model;

[0066] Obtain a training dataset, which includes image data of multiple bridge surfaces;

[0067] The initial neural network model is trained using the training dataset until the loss function converges to a preset value, thus obtaining the trained neural network model.

[0068] Specifically, before obtaining the training dataset, it needs to be constructed. This invention first obtains the original dataset, which includes images of bridge surfaces of different materials, structures, and aging levels under different seasons and weather conditions. Then, data augmentation techniques are used to process the original dataset to obtain the training dataset. These data augmentation techniques include random cropping, color dithering, and adding noise.

[0069] Random cropping increases the model's adaptability to changes in field of view by randomly selecting a portion of the original image as new training samples. In bridge image detection tasks, this means the model will learn to identify key bridge features from local images, even if the positions of these features change within the image. This is particularly important for handling image variations caused by different shooting angles and distances in real-world scenarios. Color dithering simulates bridge images under different lighting conditions by adjusting attributes such as brightness, contrast, and saturation. In bridge image detection tasks, since lighting conditions may change over time, weather, and season, color dithering helps the model learn how to accurately identify bridge conditions under different lighting conditions. This helps improve the model's stability and reliability in practical applications. Adding noise (such as Gaussian noise) simulates interference that may be introduced during image transmission or acquisition. Bridge images may be contaminated by various noises during transmission, and by adding noise during training, the model will learn to ignore these noise interferences and focus on extracting useful bridge features. This helps improve the model's robustness to noise. By comprehensively utilizing various data augmentation techniques and processing the original dataset, the neural network model in this invention can perform excellently in complex bridge deck detection tasks.

[0070] During training, supervised learning is used. The gradient descent optimization algorithm is employed to calculate the gradients of the model parameters, which are then backpropagated back through each layer of the model to update the parameters. Understandably, by continuously learning and optimizing its internal parameters, the model can achieve accurate understanding and evaluation of bridge surface images.

[0071] Please see Figure 1 The bridge surface ice and snow removal method based on neural networks described in this embodiment of the invention further includes the following steps:

[0072] Step S300: Compare the temperature data with the temperature threshold. If the temperature data is lower than the temperature threshold, turn on the bridge surface heating device and heat it according to the heating level to remove the ice and snow from the bridge surface.

[0073] Specifically, whenever temperature and image data are collected, the image data is input into a trained neural network model to obtain temperature thresholds and heating levels. The temperature data is then compared to the thresholds; if the temperature is below the threshold, an activation command is generated to turn on the bridge deck heating system, and the bridge deck is heated according to the specified heating level. This method enables more intelligent heating decisions, resulting in higher efficiency in removing ice and snow from the bridge deck. If the heating system fails to start for any reason, the system immediately triggers an anomaly reporting process, sending a startup failure message to management personnel through a pre-set communication channel to ensure the problem is quickly identified and addressed. This real-time feedback mechanism is crucial for ensuring safe bridge passage in winter, effectively preventing ice and snow accumulation due to heating system failure and reducing the risk of potential road accidents. Furthermore, all anomaly records are automatically archived by the system, forming a detailed event log, providing data support for subsequent equipment maintenance, performance evaluation, and process optimization.

[0074] In one implementation, the method further includes:

[0075] If the temperature data is higher than the temperature threshold, the bridge deck heating device is turned off.

[0076] Specifically, by collecting temperature data at a preset frequency, the system considers the task of removing ice and snow from the bridge surface to be complete once the temperature data exceeds a certain threshold, and promptly shuts off the bridge surface heating device, effectively avoiding energy waste. Through real-time monitoring and intelligent judgment, the system can accurately capture changes in the bridge surface condition and adjust the heating strategy accordingly, thereby ensuring road safety while minimizing unnecessary energy consumption and achieving efficient resource utilization and sustainable environmental protection.

[0077] In one implementation, after the bridge deck heating device is turned off, the system further includes:

[0078] Generate task completion information;

[0079] The task completion information will then be sent to the relevant staff.

[0080] Specifically, after the heating device is turned off, a task completion message is promptly sent to relevant personnel. The message can be sent via, but is not limited to, SMS, email, or app notifications, ensuring rapid and accurate delivery. Upon receiving the task completion message, relevant personnel can review camera footage to confirm the snow and ice removal progress and visually assess the effectiveness. In heavy snowstorms, if snow and ice on the bridge surface are not completely removed, bridge closures may occur. This invention automatically sends task completion messages to relevant personnel, who can then review camera footage for confirmation, allowing for timely lifting of the bridge closure, restoring traffic flow, and minimizing the social impact and economic losses caused by the closure.

[0081] In one implementation, after activating the bridge deck heating device and heating according to the heating level, the method further includes:

[0082] Obtain the operating duration of the bridge deck heating device;

[0083] If the activation duration exceeds a preset time threshold and the current temperature data is lower than the temperature threshold, an alarm message will be generated.

[0084] The alarm information is sent to the target sanitation workers so that they can carry out manual snow removal.

[0085] Specifically, once the bridge deck heating device is activated, the activation duration is recorded. The preset time threshold is set based on a combination of factors, including bridge deck material, climate conditions, and historical heating effects, aiming to ensure the desired heating effect is achieved within a reasonable timeframe. If the activation duration exceeds the preset time threshold, it is further determined whether the current temperature data is still below the threshold. If the current temperature data is still below the threshold, it means the heating effect is unsatisfactory and the goal of removing ice and snow has not been achieved. In this case, an alarm is promptly generated and sent to the target sanitation worker for timely manual intervention. The heating device, the data processing module, and the data acquisition device in this invention are all connected via a distributed soft bus. Data such as temperature threshold, heating level, and activation duration can be displayed on the control system interface, allowing control personnel to intuitively understand the current ice and snow removal status on the bridge deck.

[0086] In one implementation, the preset time threshold can be dynamically adjusted based on the current weather conditions.

[0087] In one implementation, the method further includes:

[0088] Based on the heating level, determine the power consumption of the heating device corresponding to the heating level;

[0089] Obtain the first heating duration corresponding to the heating level, and multiply the first heating duration by the power consumption to obtain the first energy consumption corresponding to the heating level;

[0090] Adding up all the first energy consumptions gives the total energy consumption for the entire ice removal mission.

[0091] The total energy consumption is obtained based on the power consumption and the heating time corresponding to the power consumption.

[0092] Specifically, different heating levels correspond to different power consumption of the heating device. Once the heating level is determined, the corresponding power consumption can be identified. During the entire bridge ice and snow removal process (i.e., from the start to the shutdown of the heating equipment), the different heating levels during the ice and snow removal process are multiplied by the corresponding first heating duration to obtain the first energy consumption for each heating level. Adding all the first energy consumptions together yields the total energy consumption for the entire ice and snow removal task. The first energy consumption is the energy consumption of that heating level in the bridge ice and snow removal task. The total energy consumption is the energy consumption for the entire bridge ice and snow removal process. The total energy consumption can be displayed on the control system interface or stored in the system for subsequent data analysis to optimize energy efficiency, and can also serve as data support for subsequent adjustments to preset time thresholds. This approach helps improve the overall energy efficiency management level of the system.

[0093] In one implementation, the heating device is a heating mesh, which is uniformly embedded in the bridge surface.

[0094] Specifically, the heating mesh of this invention consists of multiple conductive wires, which can be made of metal alloys or special conductive materials, without limitation. When the conductive wires are energized, they release heat to heat the bridge deck. By uniformly embedding the heating mesh into the bridge deck structure, this invention ensures that heat is evenly distributed across the entire bridge deck, avoiding localized overheating or underheating. This design not only improves heating efficiency but also extends the service life of the heating mesh because the uniformly distributed heat reduces stress concentration caused by temperature gradients.

[0095] In summary, this invention utilizes a data acquisition device to collect temperature and image data of the bridge surface at a preset frequency; the image data is input into a trained neural network model, which processes the data and outputs the heating level and temperature threshold corresponding to the image data; the temperature data is compared with the temperature threshold, and if the temperature data is lower than the temperature threshold, the bridge surface heating device is activated and heated according to the specified heating level to remove ice and snow from the bridge surface. This invention, by combining a data acquisition device and a neural network model, can automatically control the bridge surface heating device, effectively achieving the goal of safely and efficiently removing ice and snow from bridges.

[0096] In one embodiment, such as Figure 2As shown, based on the above-mentioned bridge surface ice and snow removal method based on neural networks, the present invention also provides a bridge surface ice and snow removal system based on neural networks, comprising:

[0097] Data acquisition module 100 is used to acquire temperature data and image data of the bridge deck using data acquisition equipment at a preset frequency;

[0098] The data processing module 200 is used to input the image data into a trained neural network model, process it through the neural network model, and output the heating level and temperature threshold corresponding to the image data.

[0099] The ice and snow removal module 300 is used to compare the temperature data with the temperature threshold. If the temperature data is lower than the temperature threshold, the bridge surface heating device is turned on and heated according to the heating level to remove the ice and snow from the bridge surface.

[0100] In one embodiment, the neural network model is a convolutional neural network model; the data processing module includes:

[0101] The normalization processing unit is used to input the image data into the input layer of the trained convolutional neural network model, and process it through the input layer to obtain normalized image data;

[0102] The first feature data generation unit is used to input the normalized image data into the convolutional layer of the convolutional neural network model, and extract features from the normalized image data through the convolutional layer to obtain the first feature data.

[0103] The second feature data generation unit is used to input the first feature data into the activation layer and transform it through a nonlinear activation function to obtain the second feature data.

[0104] The third feature data generation unit is used to input the second feature data into the pooling layer of the convolutional neural network, and then downsample it through the pooling layer to obtain the third feature data;

[0105] The result generation unit is used to input the third feature data into the fully connected layer of the convolutional neural network model, and process it through the fully connected layer to obtain the heating level and temperature threshold corresponding to the image data.

[0106] The result transmission unit is used to output the heating level and temperature threshold corresponding to the image data through the output layer of the convolutional neural network model.

[0107] In one embodiment, the system further includes:

[0108] The first building unit is used to build an initial neural network model, wherein the neural network model is a convolutional neural network model.

[0109] The training set acquisition unit is used to acquire a training dataset, which includes image data of multiple bridge surfaces.

[0110] The model training unit is used to train the initial neural network model using the training dataset until the loss function converges to a preset value, thereby obtaining the trained neural network model.

[0111] In one embodiment, the system further includes:

[0112] The equipment shutdown unit is used to shut down the bridge deck heating device if the temperature data is higher than the temperature threshold.

[0113] In one embodiment, the system further includes:

[0114] The second data acquisition unit is used to acquire the operating duration of the bridge deck heating device;

[0115] An alarm generation unit is used to generate alarm information if the opening duration exceeds a preset time threshold and the current temperature data is lower than the temperature threshold.

[0116] An alarm sending unit is used to send the alarm information to the target sanitation worker so that manual snow removal can be carried out.

[0117] Based on the above embodiments, the present invention also provides a terminal, the principle block diagram of which can be as follows: Figure 3 As shown. The terminal includes a processor, memory, network interface, and display screen connected via a device bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores operating devices and a neural network-based bridge icing removal program. The internal memory provides an environment for the operation of the operating devices and the neural network-based bridge icing removal program stored in the non-volatile storage medium. The network interface is used for communication with external terminals via a network connection. When executed by the processor, the neural network-based bridge icing removal program implements the steps of any of the aforementioned intelligent building energy control methods. The display screen can be a liquid crystal display (LCD) or an e-ink display.

[0118] Those skilled in the art will understand that Figure 3 The block diagram shown is only a partial structural diagram related to the present invention and does not constitute a limitation on the terminal to which the present invention is applied. A specific terminal may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0119] In one embodiment, a terminal is provided, the terminal including a memory, a processor, and a neural network-based bridge surface ice and snow removal program stored in the memory and executable on the processor. When the neural network-based bridge surface ice and snow removal program is executed by the processor, it implements the steps of any of the intelligent building energy control methods provided in the embodiments of the present invention.

[0120] This invention also provides a computer-readable storage medium storing a bridge surface ice and snow removal program based on a neural network. When the neural network-based bridge surface ice and snow removal program is executed by a processor, it implements the steps of any of the neural network-based bridge surface ice and snow removal methods provided in this invention.

[0121] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0122] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the above device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this invention. The specific working process of the units and modules in the above device can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0123] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0124] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0125] In the embodiments provided by this invention, it should be understood that the disclosed apparatus / terminal devices and methods can be implemented in other ways. For example, the apparatus / terminal device embodiments described above are merely illustrative. For instance, the division of modules or units described above is only a logical functional division, and in actual implementation, it can be divided in other ways. For example, multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed.

[0126] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not mean that the essence of the corresponding technical solutions deviates from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for removing ice and snow from bridge surfaces based on neural networks, characterized in that, The method includes: Temperature and image data of the bridge deck are collected using data acquisition equipment at a preset frequency. The image data is input into a trained neural network model, processed by the neural network model, and the heating level and temperature threshold corresponding to the image data are output. The temperature data is compared with the temperature threshold. If the temperature data is lower than the temperature threshold, the bridge surface heating device is turned on and heated according to the heating level to remove the ice and snow from the bridge surface. The neural network model is a convolutional neural network model; the step of inputting the image data into the trained neural network model, processing it, and outputting the heating level and temperature threshold corresponding to the image data includes: The image data is input into the input layer of the trained convolutional neural network model, and processed by the input layer to obtain normalized image data. The normalized image data is input into the convolutional layer of the convolutional neural network model, and the convolutional layer performs feature extraction on the normalized image data to obtain the first feature data. The first feature data is input into the activation layer and transformed by a nonlinear activation function to obtain the second feature data; The second feature data is input into the pooling layer of the convolutional neural network, and downsampled by the pooling layer to obtain the third feature data; The third feature data is input into the fully connected layer of the convolutional neural network model, and processed by the fully connected layer to obtain the heating level and temperature threshold corresponding to the image data; The heating level and temperature threshold corresponding to the image data are output through the output layer of the convolutional neural network model.

2. The bridge surface ice and snow removal method based on neural networks according to claim 1, characterized in that, The data acquisition device includes a temperature sensor and a camera, both of which are mounted on the bridge deck.

3. The bridge surface ice and snow removal method based on neural networks according to claim 1, characterized in that, The training steps of the neural network model include: Construct an initial neural network model, which is a convolutional neural network model; Obtain a training dataset, which includes image data of multiple bridge surfaces; The initial neural network model is trained using the training dataset until the loss function converges to a preset value, thus obtaining the trained neural network model.

4. The bridge surface ice and snow removal method based on neural networks according to claim 1, characterized in that, The method further includes: If the temperature data is higher than the temperature threshold, the bridge deck heating device is turned off.

5. The bridge surface ice and snow removal method based on neural networks according to claim 1, characterized in that, After the bridge deck heating device is activated and heated according to the heating level, the system further includes: Obtain the operating duration of the bridge deck heating device; If the activation duration exceeds a preset time threshold and the current temperature data is lower than the temperature threshold, an alarm message will be generated. The alarm information is sent to the target sanitation workers so that they can carry out manual snow removal.

6. The bridge surface ice and snow removal method based on neural networks according to claim 1, characterized in that, The heating device is a heating mesh, which is evenly embedded in the bridge surface.

7. A bridge surface ice and snow removal system based on neural networks, characterized in that, The system is used to implement the steps of the bridge surface ice and snow removal method based on any one of claims 1-6, and the system comprises: The data acquisition module is used to collect temperature data and image data of the bridge deck at a preset frequency using data acquisition equipment; The data processing module is used to input the image data into a trained neural network model, process the image data through the neural network model, and output the heating level and temperature threshold corresponding to the image data. The ice and snow removal module is used to compare the temperature data with the temperature threshold. If the temperature data is lower than the temperature threshold, the bridge surface heating device is turned on and heated according to the heating level to remove the ice and snow from the bridge surface.

8. A terminal, characterized in that, The terminal includes a memory, a processor, and a neural network-based bridge surface ice and snow removal program stored in the memory and executable on the processor. When the neural network-based bridge surface ice and snow removal program is executed by the processor, it implements the steps of the neural network-based bridge surface ice and snow removal method as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a bridge surface ice and snow removal program based on a neural network. When the bridge surface ice and snow removal program based on a neural network is executed by a processor, it implements the steps of the bridge surface ice and snow removal method based on a neural network as described in any one of claims 1-6.

Images