An intelligent safety control system for rolling shutter doors

Abstract

The present invention provides an intelligent safety control system for rolling shutter doors, which relates to the technical field of rolling shutter door control, and comprises a plurality of sensor modules; an infrared data analysis module, which is used to judge whether an object enters under the rolling shutter door, and generates a first control instruction; an ultrasonic data analysis module, which is used to judge whether the distance between the rolling shutter door and the object is less than a first preset threshold value, and generates a second control instruction; a resistance analysis protection module, which is used to generate a third control instruction when the resistance exceeds a second preset threshold value; a prediction control module, which is used to predict the motion trajectory of the object, and generates a fourth control instruction, so that when the motion trajectory of the object passes under the rolling shutter door, the rolling shutter door reaches a height intersecting with the trajectory within a preset time and stops running; a drive control module, which is used to receive the control instruction and control the operation of the rolling shutter door; a fault feedback module; and a power supply module. The present invention can improve the intelligence level of the rolling shutter door and facilitate the operation of the rolling shutter door.

Description

Technical Field

[0001] The invention relates to the technical field of rolling shutter door control, in particular to an intelligent safety control system for rolling shutter doors. Background Art

[0002] Electric rolling shutters are widely used in shops, garages, warehouses and other places. They have gradually replaced traditional manual rolling shutters due to their convenience, speed and safety. However, in actual applications, electric rolling shutters may encounter various obstacles during operation. If there is no effective control system, the rolling shutters may cause motor damage, rolling shutter deformation and even safety accidents after encountering obstacles.

[0003] In the related art, mechanical resistance protection devices are generally used to control and achieve anti-pinch, but they rely on mechanical structures to sense resistance and cannot respond in time when facing slight resistance. In addition, if a moving object (such as a vehicle, a child riding a toy car, a pet, etc.) has not reached the bottom of the rolling door, the rolling door will continue to close, which may cause the moving object to fail to stop in time, resulting in a safety accident. Summary of the invention

[0004] The purpose of the present invention is to provide an intelligent safety control system for rolling shutter doors, aiming to solve the problems mentioned in the background technology.

[0005] In order to solve the above technical problems, the technical solution of the present invention is as follows:

[0006] An intelligent safety control system for rolling shutter doors, the system comprising:

[0007] A plurality of sensor modules are installed at the lower edge of the rolling door, wherein the sensor modules include an infrared sensor, an ultrasonic sensor and a pressure sensor. The infrared sensor is used to detect the movement of an object under the rolling door, the ultrasonic sensor is used to measure the distance between the object and the rolling door, and the pressure sensor is used to sense the resistance of the rolling door when it encounters an object;

[0008] The infrared data analysis module is used to receive the detection data of the infrared sensor, determine whether the object enters under the rolling door, and generate a first control instruction;

[0009] The ultrasonic data analysis module is used to receive the detection data of the ultrasonic sensor, determine whether the distance between the rolling door and the object is less than a set safety distance threshold, and generate a second control instruction;

[0010] The resistance analysis protection module is used to receive the resistance data of the pressure sensor and generate a third control instruction when the resistance exceeds a second preset threshold value, so as to control the rolling door to move a first preset distance in the opposite direction of the resistance and then stop running;

[0011] The prediction control module trains the preset GRU model through the training data to obtain the prediction control model, and can predict the motion trajectory of the object according to the real-time data of the sensor module, and generate the fourth control instruction, so that when the motion trajectory of the object passes under the rolling door, the rolling door reaches the height intersecting with the trajectory within a preset time and stops running;

[0012] A drive control module, used to receive control instructions from the infrared data analysis module, the ultrasonic data analysis module, the resistance analysis protection module and the prediction control module, and start, stop or adjust the movement of the rolling door through the controller of the rolling door;

[0013] The fault feedback module is used to detect the operation status of the sensor module and the rolling door motor, and when an abnormal situation occurs, the abnormal information is sent to the preset terminal device through the wireless communication module.

[0014] Furthermore, the prediction control module includes:

[0015] A historical data storage unit, used for acquiring and storing historical data of multiple sensor modules;

[0016] A data preprocessing unit, used to preprocess historical data to obtain training data, the data preprocessing unit includes a data cleaning module, a data normalization module and a feature extraction module, wherein the data cleaning module is used to remove abnormal data, the data normalization module is used to standardize the data to a preset range, and the feature extraction module is used to extract feature parameters related to the object's motion trajectory;

[0017] A GRU model training unit, used to train a preset GRU model through training data to obtain a predictive control model;

[0018] A real-time prediction unit is used to receive real-time data from multiple sensor modules and input it into a prediction control model to obtain a prediction result;

[0019] The control instruction generating unit is used to generate a fourth control instruction according to the prediction result.

[0020] Furthermore, the preset GRU model is trained through the training data to obtain a predictive control model, including:

[0021] The training goal is to set the prediction error between the actual motion trajectory of the object and the prediction result of the GRU model to be less than a first preset threshold;

[0022] The training data is grouped by batch processing to obtain a training data set, and the training data set is input into the GRU model to obtain the model prediction value;

[0023] Calculate the prediction error in each training iteration according to the loss function;

[0024] The parameters of the GRU model are adjusted until the training target is achieved and then the training is stopped to obtain a predictive control model.

[0025] Furthermore, the loss function is calculated as follows:

[0026] ,

[0027] in, is the prediction error, is the true value, is the model prediction value, is the sample size, For the The predicted distance between the object and the rolling door at a certain moment, is the set safety distance threshold, , and is the weight coefficient.

[0028] Furthermore, the parameters of the GRU model are adjusted, including:

[0029] Calculate the partial derivative of the loss function with respect to each parameter of the GRU model to obtain the gradient value of the current iteration;

[0030] Perform an exponentially weighted moving average on the gradient value to obtain a first-order momentum estimate;

[0031] Perform an exponentially weighted moving average on the squared gradient values ​​to get the second-order momentum estimate;

[0032] Calculate the adaptive learning rate based on the first-order momentum estimate and the second-order momentum estimate;

[0033] Adjust the parameters of the GRU model based on the gradient value and adaptive learning rate.

[0034] Furthermore, the infrared sensor comprises:

[0035] Light sensing module, used to detect the light intensity in the environment in real time;

[0036] The first control module is used to receive the light intensity and automatically adjust the sensitivity of the infrared sensor according to the change of the light intensity;

[0037] The sensitivity of the infrared sensor is adjusted by the following formula:

[0038] ,

[0039] in, is the sensitivity of the infrared sensor, is the current ambient light intensity, For maximum sensitivity, is the light intensity threshold, is the sensitivity adjustment factor.

[0040] Furthermore, the ultrasonic sensor includes a detection range adjustment unit, which is used for dynamic adjustment according to the operating state of the rolling door, specifically including:

[0041] By receiving the operating status signal of the rolling door drive control module, it is determined whether the rolling door is in the rising, falling or stopping state;

[0042] When the rolling door is in the rising state, the control unit sets the detection angle of the ultrasonic sensor to 90 degrees to 180 degrees;

[0043] When the rolling door is in the lowered state, the detection angle of the ultrasonic sensor expands to 0 degrees to 180 degrees.

[0044] Furthermore, the detection range adjustment unit can also be dynamically adjusted according to the moving speed of the object, specifically including:

[0045] Get the real-time speed of the object. When the real-time speed of the object exceeds the preset speed threshold, dynamically expand the detection distance of the ultrasonic sensor.

[0046] The detection distance of the ultrasonic sensor is adjusted by the following formula:

[0047] ,

[0048] in, is the adjusted detection distance, is the initial detection distance, is the object speed, is the adjustment factor;

[0049] When the object speed drops below the preset speed threshold, the ultrasonic sensor's default detection distance is restored.

[0050] Furthermore, the fault feedback module includes:

[0051] A fault diagnosis unit is used to obtain the working status parameters of the infrared sensor, ultrasonic sensor and pressure sensor and the operating parameters of the rolling door motor in real time;

[0052] an intelligent analysis unit, for comparing the working state parameter and the operating parameter with the corresponding rated parameter to obtain a parameter difference, and comparing the parameter difference with a third preset threshold to generate fault feedback information of different levels;

[0053] A wireless communication module is used to receive fault feedback information, generate a fault report according to the fault feedback information, automatically send an alarm signal of the fault report to the user terminal, and trigger the emergency stop mechanism of the rolling door according to the level of the fault feedback information;

[0054] The trend analysis unit is used to obtain a fault trend analysis report based on the historical data of the fault feedback information and send the trend analysis report to the user.

[0055] Furthermore, the fault feedback module also includes:

[0056] A self-repair prompt module, which is used to generate a repair plan based on historical fault feedback information and repair records when the level of the fault feedback information is lower than a preset level, and send the repair steps through a preset terminal device;

[0057] A fault upgrade unit, which is used to automatically upgrade the level of fault feedback information when the repair plan is ineffective, and send a fault report to maintenance personnel through a wireless communication module;

[0058] The fault library update unit is used to automatically update the fault library according to user feedback and repair records.

[0059] The above solution of the present invention includes at least the following beneficial effects:

[0060] By introducing multi-sensor fusion technology, including infrared sensors, ultrasonic sensors and pressure sensors, the defect of the existing electric rolling door being slow to respond after encountering obstacles is effectively solved. The system can accurately detect obstacles or moving objects under and around the rolling door, and make judgments in advance before the objects approach the rolling door, and issue commands to control the rolling door to stop or move in the opposite direction, so as to prevent the objects from being damaged by the closing of the rolling door. Compared with traditional mechanical obstacle protection devices, this system relies on a variety of non-contact sensors, which can identify slight resistance or moving objects approaching the rolling door earlier and more accurately, thus improving safety.

[0061] Infrared sensors are used to detect the movement of objects under the rolling door, ultrasonic sensors are used to measure the distance between the object and the rolling door, and pressure sensors are used to sense the resistance of the rolling door when it encounters an object. Through these sensor modules, the system can not only issue control instructions when an object approaches, but also adjust the movement trajectory of the rolling door in time according to the size of the resistance. This dynamic response capability avoids damage to the motor, prevents deformation of the rolling door, and further improves the service life and reliability of the system. Compared with the mechanical structure's resistance perception method in the prior art, this technical solution can accurately identify a variety of moving objects, such as vehicles, children riding toy cars, pets, etc., so that when they are about to enter under the rolling door, countermeasures can be taken in advance to avoid potential safety accidents.

[0062] In addition, the system also introduces historical data analysis and predictive control modules. By learning and analyzing the historical data of the sensor, a predictive control model is formed, which can pre-judge the movement trajectory of the object and issue control instructions. This intelligent predictive control capability further enhances the safety of the system, ensuring that the rolling door can still be accurately controlled in complex dynamic scenes and timely avoid collisions between objects and the rolling door. Compared with existing technologies, the system can not only respond when encountering obstacles, but also has the ability to predict in advance, effectively improving the level of intelligence. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 It is a flow chart of an intelligent safety control system for rolling shutter doors provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0064] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the 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.

[0065] like Figure 1 As shown, an embodiment of the present invention provides an intelligent safety control system for rolling shutter doors, the system comprising:

[0066] A plurality of sensor modules are installed at the lower edge of the rolling door, wherein the sensor modules include an infrared sensor, an ultrasonic sensor and a pressure sensor. The infrared sensor is used to detect the movement of an object under the rolling door, the ultrasonic sensor is used to measure the distance between the object and the rolling door, and the pressure sensor is used to sense the resistance of the rolling door when it encounters an object;

[0067] The infrared data analysis module is used to receive the detection data of the infrared sensor, determine whether the object enters under the rolling door, and generate a first control instruction;

[0068] The ultrasonic data analysis module is used to receive the detection data of the ultrasonic sensor, determine whether the distance between the rolling door and the object is less than a set safety distance threshold, and generate a second control instruction;

[0069] The resistance analysis protection module is used to receive the resistance data of the pressure sensor and generate a third control instruction when the resistance exceeds a second preset threshold value, so as to control the rolling door to move a first preset distance in the opposite direction of the resistance and then stop running;

[0070] The prediction control module trains the preset GRU model through the training data to obtain the prediction control model, and can predict the motion trajectory of the object according to the real-time data of the sensor module, and generate the fourth control instruction, so that when the motion trajectory of the object passes under the rolling door, the rolling door reaches the height intersecting with the trajectory within a preset time and stops running;

[0071] A drive control module, used to receive control instructions from the infrared data analysis module, the ultrasonic data analysis module, the resistance analysis protection module and the prediction control module, and start, stop or adjust the movement of the rolling door through the controller of the rolling door;

[0072] A fault feedback module is used to detect the operation of the sensor module and the rolling door motor, and when an abnormal situation occurs, the abnormal information is sent to a preset terminal device through the wireless communication module;

[0073] The power module provides continuous power support for the system.

[0074] In an embodiment of the present invention, the intelligent safety control system for rolling shutter doors uses multiple sensor modules combined with infrared sensors, ultrasonic sensors and pressure sensors to achieve accurate detection and control of objects under the rolling shutter door. Compared with the traditional rolling shutter door system that can only rely on simple mechanical control, the multi-sensor fusion technology of this system effectively improves the safety of the system and avoids the situation where objects are accidentally injured when the rolling shutter door is closed. At the same time, the system comprehensively judges the position and movement state of the object through data from different sensor modules, realizes real-time response and sends control instructions, so that the rolling shutter door can automatically stop or reverse movement when needed to ensure safe operation. Compared with the existing technology, this system provides a more flexible rolling shutter door control mechanism with a higher level of intelligence, and can still work effectively in complex environments, with strong robustness.

[0075] When the infrared sensor detects the movement of an object, it relies on the infrared data analysis module, while the ultrasonic sensor is used to measure the distance between the object and the rolling door. Combined with the pressure sensor, it senses the resistance of the rolling door encountering the object, forming a complete safety protection mechanism. In addition, the drive control module is responsible for receiving control instructions from all analysis modules and protection modules to ensure the timeliness and accuracy of the rolling door movement. The system design also introduces a fault feedback module that can detect system anomalies and send alarm information to the user in a timely manner through the wireless communication module, thereby avoiding safety hazards caused by system failures.

[0076] In a preferred embodiment of the present invention, the prediction control module includes:

[0077] A historical data storage unit, used for acquiring and storing historical data of multiple sensor modules;

[0078] A data preprocessing unit, used to preprocess historical data to obtain training data, the data preprocessing unit includes a data cleaning module, a data normalization module and a feature extraction module, wherein the data cleaning module is used to remove abnormal data, the data normalization module is used to standardize the data to a preset range, and the feature extraction module is used to extract feature parameters related to the object's motion trajectory;

[0079] A GRU model training unit, used to train a preset GRU model through training data to obtain a predictive control model;

[0080] A real-time prediction unit is used to receive real-time data from multiple sensor modules and input it into a prediction control model to obtain a prediction result;

[0081] The control instruction generating unit is used to generate a fourth control instruction according to the prediction result.

[0082] In an embodiment of the present invention, the predictive control module in the rolling door intelligent safety control system can form a predictive control capability for the future trajectory of objects by acquiring and storing historical data of multiple sensor modules. Compared with traditional control systems, this system effectively improves the accuracy of object motion trajectory prediction through the preprocessing of historical data and the training of the GRU model, and can issue control instructions in advance to avoid accidents when the object moves at a fast speed or at a close distance. Compared with the prior art, this system uses deep learning technology to provide an intelligent, data-driven control strategy through the mining of historical data and the combination of real-time data, further improving the safety and predictive capabilities of the system.

[0083] The data preprocessing unit cleans, normalizes and extracts features from the sensor's historical data to ensure that the training data input into the GRU model is consistent and accurate. The model can build a prediction model for the object's motion trajectory by learning the time series features in the sensor data. The real-time prediction unit inputs real-time data into the prediction model, generates control instructions, and drives the adjustment of the rolling door, enabling the system to flexibly respond to complex object motion scenarios.

[0084] In a preferred embodiment of the present invention, the construction process of the GRU model includes:

[0085] Initialize the GRU model structure, including multiple layers of GRU units, each of which includes an update gate, a reset gate, and a candidate hidden state;

[0086] The update gate is used to control the information update ratio between the current input data and the hidden state at the previous moment. The reset gate is used to control the information retention ratio of the hidden state at the previous moment in the current calculation. The candidate hidden state is used to perform nonlinear transformation on the current input data to generate new state information.

[0087] The input layer of the GRU model is used to receive input data, and the output layer is used to output the motion trajectory parameters of the object. Multiple layers of GRU units are connected between the input layer and the output layer, and state information is transmitted through inter-layer connections.

[0088] Each GRU unit performs a recursive calculation of the input data at each time step according to the timing characteristics of the time series, and passes the state information to the GRU unit at the next moment;

[0089] The number of nodes, activation function, and gating mechanism parameters of each layer of GRU units are configured to obtain the GRU model.

[0090] In an embodiment of the present invention, the construction process of the GRU model realizes the time series prediction capability based on the recursive neural network. Through the series connection of multiple GRU units, the system can process the time-dependent characteristics in the sensor data and infer the future motion trajectory based on the historical behavior of the object. Compared with the traditional rolling door control method, the GRU model has higher robustness and accuracy in processing complex motion trajectory prediction. Compared with the prior art, after adopting the GRU model, the system can process data with a longer time span, adapt to more complex environmental changes, and improve the system's prediction accuracy of object movement.

[0091] The GRU unit consists of an update gate, a reset gate, and a candidate hidden state. The update gate determines the information update ratio between the current input data and the hidden state at the previous moment, and the reset gate is used to control the degree of information retention of the hidden state at the previous moment. Through these gating mechanisms, the GRU model can dynamically adjust the processing method of time series data, thereby improving the prediction accuracy. With the support of this model, the rolling door control system can make corresponding predictions and judgments based on the different motion characteristics of objects in different time periods.

[0092] Configure the number of nodes in each layer of GRU units: The number of nodes refers to the number of neurons (or units) contained in each layer of GRU units. These nodes are the basic components of the network, responsible for receiving input, processing information and transmitting status.

[0093] The choice of the number of nodes directly affects the computational power and complexity of the GRU layer. The number of nodes is usually configured based on the complexity of the data and the requirements of the task. Simpler tasks can choose fewer nodes, while complex tasks (such as long sequence prediction) require more nodes to capture temporal dependencies.

[0094] Configure activation function: The activation function is a nonlinear transformation function used for each node of the neural network, which determines the output of each node. The GRU unit uses a specific activation function to process information.

[0095] The activation function here can be selected as the Sigmoid function, which is used as the activation function of the gating mechanism (such as the update gate and the reset gate). The output value is between 0 and 1, indicating the switch of the gating state.

[0096] You can also use the Tanh function, which is used to update the activation function of the hidden state. The output value is between 0 and 1, which helps to propagate information in the positive and negative directions and maintain the nonlinear expression ability of GRU.

[0097] The activation function introduces nonlinear properties, enabling the GRU unit to learn and represent complex temporal dependencies rather than just linear patterns.

[0098] Gating mechanism parameter configuration: The gating mechanisms in GRU include update gates and reset gates. These gating mechanisms determine how the current input, previous state, and new candidate state interact with each other.

[0099] The update gate controls the mixing ratio of the current hidden state and the previous hidden state, and determines how much past information is retained. The degree of opening of the update gate is determined by the learned weight and bias parameters.

[0100] Reset gate: controls the influence of the previous hidden state in generating candidate new states, and determines how much past information to clear. The degree of opening of the update gate is determined by the learned weight and bias parameters.

[0101] In a preferred embodiment of the present invention, a preset GRU model is trained by training data to obtain a predictive control model, including:

[0102] The training goal is to set the prediction error between the actual motion trajectory of the object and the prediction result of the GRU model to be less than a first preset threshold;

[0103] The training data is grouped by batch processing to obtain a training data set, and the training data set is input into the GRU model to obtain the model prediction value;

[0104] The prediction error in each training iteration is calculated according to the loss function; the loss function is calculated as follows:

[0105] ,

[0106] in, is the prediction error, is the true value, that is, the actual position of the object in the historical data, is the model prediction value, that is, the object position predicted by the model, is the number of samples, that is, the number of times the object moves in the historical data, For the The predicted distance between the object and the rolling door at a certain moment, is the set safety distance threshold, , and is the weight coefficient;

[0107] Adjust the parameters of the GRU model to reduce the prediction error;

[0108] Repeat the above training steps until the training target is achieved and then stop training to obtain the predictive control model.

[0109] In an embodiment of the present invention, the GRU model is trained through training data, so that the system can autonomously learn the laws of object motion in different scenarios and achieve the effect of accurately predicting the object's motion trajectory. Compared with traditional rule-based control systems, this system can continuously optimize the prediction model through continuous updating of historical data, thereby maintaining efficient performance in actual use. The system groups data through batch processing, uses optimization algorithms to reduce prediction errors, and can achieve training goals in a shorter time. Compared with the prior art, the system can process complex data sets more efficiently and adaptively adjust prediction strategies in different application scenarios, further improving the prediction accuracy of the system.

[0110] The calculation process of the loss function ensures the effective measurement of the model error during the training process. The optimization algorithm adaptively adjusts the GRU model parameters through first-order and second-order momentum estimation to reduce the model prediction error, and through repeated training, the model has a strong generalization ability. The batch processing method improves the training efficiency and effectively prevents the model from overfitting by processing the data in groups.

[0111] In a preferred embodiment of the present invention, the parameters of the GRU model are adjusted, including:

[0112] Calculate the partial derivative of the loss function with respect to each parameter of the GRU model to obtain the gradient value of the current iteration; the calculation formula of the gradient value is:

[0113] ,

[0114] in, For the The loss function at the iteration Parameters The gradient value of For the parameters Gradient calculation, that is, calculation of loss function Partial derivatives with respect to each parameter;

[0115] The gradient value is exponentially weighted moving averaged to obtain the first-order momentum estimate; the calculation formula for the first-order momentum estimate is:

[0116] ,in, For the The first-order momentum estimate for the iteration, is the momentum attenuation coefficient;

[0117] The squared values ​​of the gradients are exponentially weighted moving averaged to obtain the second-order momentum estimate. The calculation formula for the second-order momentum estimate is:

[0118] ,in, For the The second-order momentum estimate for the iteration, is the momentum attenuation coefficient;

[0119] Calculate the adaptive learning rate based on the first-order momentum estimate and the second-order momentum estimate;

[0120] According to the gradient value and adaptive learning rate, adjust the parameters of the GRU model; the parameter adjustment formula is:

[0121] ,in, For the The parameters of the iteration (the parameters include the weights and parameters of the GRU model), is the initial global learning rate, To prevent the minimum value from dividing by zero, take , is the deviation correction value of the first-order momentum, is the deviation correction value of the second-order momentum, is the index.

[0122] In the embodiment of the present invention, the system further improves the training efficiency and accuracy of the GRU model through the optimization algorithm, combines the advantages of momentum and adaptive learning rate adjustment, and can automatically adjust the parameters of the GRU model in each training iteration, so that the model can maintain a low prediction error in complex environments. Compared with the traditional gradient descent algorithm. Compared with the prior art, this system reduces the possibility of overfitting and optimizes the training speed by dynamically adjusting the model parameters.

[0123] The construction of the loss function further helps the system control errors during the training process. By providing layer-by-layer feedback on the model prediction error, the system can quickly identify and correct problems in the model, allowing the model to more accurately predict the trajectory of an object.

[0124] In a preferred embodiment of the present invention, the infrared sensor comprises:

[0125] Light sensing module, used to detect the light intensity in the environment in real time;

[0126] The first control module is used to receive the light intensity and automatically adjust the sensitivity of the infrared sensor according to the change of the light intensity;

[0127] The sensitivity of the infrared sensor is adjusted by the following formula:

[0128] ,

[0129] in, is the sensitivity of the infrared sensor, is the current ambient light intensity, For maximum sensitivity, is the light intensity threshold, is the sensitivity adjustment factor.

[0130] In the embodiment of the present invention, the system can accurately detect the existence or movement trajectory of objects under different ambient light conditions through the automatic sensitivity adjustment function of the infrared sensor. Compared with the traditional rolling door system, which is prone to false detection when the light changes greatly, this system automatically adjusts the sensitivity of the infrared sensor according to the real-time change of light intensity to ensure that excessive detection errors are avoided when the light is strong, and improves the detection sensitivity to ensure safety when the light is weak. This function significantly improves the system's adaptability under various lighting conditions and avoids system instability or safety hazards caused by environmental changes.

[0131] For example, in a large warehouse, an automated object recognition system is used to track and manage inventory. Since the lighting conditions in the warehouse change frequently, especially with the transition from day to night, and some areas are poorly lit (such as deep in the warehouse or in the corners), traditional sensors are prone to false alarms or inaccurate detection.

[0132] When the light in the warehouse is strong, the light sensing module will automatically reduce the sensitivity of the sensor to avoid false alarms caused by over-sensitivity; in areas with low light (such as corners or when operating at night), the system will increase the sensitivity of the sensor to ensure that objects can be accurately detected.

[0133] This sensitivity adjustment mechanism based on light intensity ensures that the warehouse management system can accurately detect and track objects in inventory under different light conditions, reducing manual intervention and improving management efficiency.

[0134] In a preferred embodiment of the present invention, the ultrasonic sensor includes a detection range adjustment unit, which is used for dynamic adjustment according to the operating state of the rolling door, specifically including:

[0135] By receiving the operating status signal of the rolling door drive control module, it is determined whether the rolling door is in the rising, falling or stopping state;

[0136] When the rolling door is in the rising state, the control unit sets the detection angle of the ultrasonic sensor to 90 degrees to 180 degrees;

[0137] When the rolling door is in the lowered state, the detection angle of the ultrasonic sensor expands to 0 degrees to 180 degrees.

[0138] In an embodiment of the present invention, the detection range of the ultrasonic sensor can be dynamically adjusted according to the operating state of the rolling door, thereby improving the detection accuracy and response speed of the system. When the rolling door is in an ascending state, the detection angle of the ultrasonic sensor will be limited to between 90 and 180 degrees, so as to focus on detecting objects in the ascending path of the rolling door. When the rolling door is descending, the system will expand the detection angle to between 0 and 180 degrees, ensuring that the area under the rolling door can be fully monitored to prevent the rolling door from accidentally colliding with objects. Compared with the prior art, the present system can automatically adjust the working mode of the sensor according to the real-time status of the rolling door, which significantly improves the flexibility and safety of the system.

[0139] The system works with the drive module to obtain the operating status signal of the rolling door, thereby dynamically adjusting the detection angle of the ultrasonic sensor. When the rolling door is rising, the detection angle is limited to between 90 and 180 degrees, so as to focus on detecting the situation of objects above the rolling door; when the rolling door is descending, the detection angle is expanded to 0 to 180 degrees, ensuring full coverage of the entire area below, effectively improving the safety of the system.

[0140] In a preferred embodiment of the present invention, the detection range adjustment unit can also be dynamically adjusted according to the moving speed of the object, specifically including:

[0141] Get the real-time speed of the object. When the real-time speed of the object exceeds the preset speed threshold, dynamically expand the detection distance of the ultrasonic sensor.

[0142] The detection distance of the ultrasonic sensor is adjusted by the following formula:

[0143] ,

[0144] in, is the adjusted detection distance, is the initial detection distance, is the object speed, is the adjustment factor;

[0145] When the object speed drops below the preset speed threshold, the ultrasonic sensor's default detection distance is restored.

[0146] In an embodiment of the present invention, the ultrasonic sensor can adjust its detection distance according to the real-time speed of the object, further enhancing the system's ability to respond to fast-moving objects. Compared with the fixed detection distance in traditional systems, this system monitors the speed of the object in real time. When the speed of the object exceeds the preset speed threshold, it automatically expands the detection distance to ensure that the system can respond to fast-moving objects in a timely manner to avoid accidents. This function makes the system more flexible, able to adapt to various dynamically changing scenarios, and improves the safety performance of the system. Through this dynamic adjustment method, the system can effectively expand the detection range to cope with objects at different speeds, achieving higher detection accuracy and safety assurance.

[0147] In a preferred embodiment of the present invention, the fault feedback module includes:

[0148] A fault diagnosis unit is used to obtain the working status parameters of the infrared sensor, ultrasonic sensor and pressure sensor and the operating parameters of the rolling door motor in real time;

[0149] an intelligent analysis unit, for comparing the working state parameter and the operating parameter with the corresponding rated parameter to obtain a parameter difference, and comparing the parameter difference with a third preset threshold to generate fault feedback information of different levels;

[0150] A wireless communication module is used to receive fault feedback information, generate a fault report according to the fault feedback information, automatically send an alarm signal of the fault report to the user terminal, and trigger the emergency stop mechanism of the rolling door according to the level of the fault feedback information;

[0151] The trend analysis unit is used to obtain a fault trend analysis report based on the historical data of the fault feedback information and send the trend analysis report to the user.

[0152] In the embodiment of the present invention, by acquiring the operating parameters of devices such as infrared sensors, ultrasonic sensors and pressure sensors in real time and combining them with the operating parameters of the rolling door motor, the operating status of the system can be monitored quickly and accurately. The intelligent analysis unit generates fault feedback information of different levels in a timely manner by comparing the difference between the current working status parameters and the predetermined threshold.

[0153] By further equipped with a wireless communication module, it can transmit real-time generated fault information to the user terminal and perform emergency shutdown operations when necessary to ensure the safe operation of the system. At the same time, the trend analysis unit can generate trend analysis reports based on historical data to help users predict potential faults and perform preventive maintenance. This multi-level fault feedback mechanism significantly improves the system's fault detection and maintenance capabilities, reduces equipment downtime, and improves overall work efficiency.

[0154] The third preset threshold is set according to adverse consequences caused when the parameter difference reaches a certain value during actual use.

[0155] Suppose, in a large automated warehouse, rolling shutters are frequently used equipment, and the long-term operation of rolling shutter motors is prone to wear, overload and other problems. Suppose at a certain moment, the working current of the motor exceeds the rated current, and the intelligent analysis unit determines that the motor is in an overload state by comparing it with the rated current.

[0156] At this point, the system will generate a level 1 warning and send a fault report to the operator’s terminal device via the wireless communication module. If the fault worsens, the motor temperature continues to rise or the current continues to exceed the standard, the system will trigger the emergency shutdown mechanism of the rolling door according to the set rules to prevent equipment damage or safety accidents.

[0157] In addition, based on the historical data of fault feedback information, the trend analysis unit found that the temperature change of the rolling door motor when running under high load has a certain regularity, and generated a trend analysis report, recommending regular maintenance under specific loads. Through these preventive measures, warehouse managers can arrange maintenance in advance, thereby reducing the failure rate of equipment and ensuring efficient operation of the warehouse.

[0158] In a preferred embodiment of the present invention, the fault feedback module further includes:

[0159] A self-repair prompt module, which is used to generate a repair plan based on historical fault feedback information and repair records when the level of the fault feedback information is lower than a preset level, and send the repair steps through a preset terminal device;

[0160] A fault upgrade unit, which is used to automatically upgrade the level of fault feedback information when the repair plan is ineffective, and send a fault report to maintenance personnel through a wireless communication module;

[0161] The fault library update unit is used to automatically update the fault library according to user feedback and repair records.

[0162] In the embodiment of the present invention, the fault feedback module in the system realizes the self-repair function of minor faults through the self-repair prompt module, further improving the maintenance convenience and intelligence level of the system. Compared with the traditional system that can only wait for manual maintenance, this system can provide automatic repair solutions according to the type of minor faults, push specific repair steps through the user terminal, help users quickly solve problems, and reduce system downtime. This function significantly improves the self-maintenance capability of the system and extends the service life of the system.

[0163] When the fault feedback information shows the lowest level, the self-repair prompt module generates a repair plan based on the historical fault records and repair plans and sends it to the user terminal. If the user executes the repair plan and the fault is resolved, the system automatically updates the fault library and records the repair operation. If the repair is ineffective, the fault upgrade unit automatically increases the fault level, sends a fault report to the maintenance personnel, and recommends further repair plans. In this way, the reliance on manual maintenance is effectively reduced.

[0164] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. An intelligent safety control system for rolling shutter doors, characterized in that: The system comprises: A plurality of sensor modules are installed at the lower edge of the rolling door, wherein the sensor modules include an infrared sensor, an ultrasonic sensor and a pressure sensor. The infrared sensor is used to detect the movement of an object under the rolling door, the ultrasonic sensor is used to measure the distance between the object and the rolling door, and the pressure sensor is used to sense the resistance of the rolling door when it encounters an object; The infrared data analysis module is used to receive the detection data of the infrared sensor, determine whether the object enters under the rolling door, and generate a first control instruction; The ultrasonic data analysis module is used to receive the detection data of the ultrasonic sensor, determine whether the distance between the rolling door and the object is less than a set safety distance threshold, and generate a second control instruction; The resistance analysis protection module is used to receive the resistance data of the pressure sensor and generate a third control instruction when the resistance exceeds a second preset threshold value, so as to control the rolling door to move a first preset distance in the opposite direction of the resistance and then stop running; The prediction control module is used to train the preset GRU model through the training data to obtain the prediction control model, and can predict the motion trajectory of the object according to the real-time data of the sensor module, and generate a fourth control instruction, so that when the motion trajectory of the object passes under the rolling door, the rolling door reaches a height that intersects with the trajectory within a preset time and stops running; A drive control module, used to receive control instructions from the infrared data analysis module, the ultrasonic data analysis module, the resistance analysis protection module and the prediction control module, and start, stop or adjust the movement of the rolling door through the controller of the rolling door; A fault feedback module is used to detect the operation of the sensor module and the rolling door motor, and when an abnormal situation occurs, the abnormal information is sent to a preset terminal device through the wireless communication module; The ultrasonic sensor includes a detection range adjustment unit, which is used for dynamic adjustment according to the operating state of the rolling door, specifically including: By receiving the operating status signal of the rolling door drive control module, it is determined whether the rolling door is in the rising, falling or stopping state; When the rolling door is in the rising state, the control unit sets the detection angle of the ultrasonic sensor to 90 degrees to 180 degrees; When the rolling door is in the lowered state, the detection angle of the ultrasonic sensor is extended to 0 to 180 degrees; The detection range adjustment unit can also be dynamically adjusted according to the moving speed of the object, specifically including: Get the real-time speed of the object. When the real-time speed of the object exceeds the preset speed threshold, dynamically expand the detection distance of the ultrasonic sensor. The detection distance of the ultrasonic sensor is adjusted by the following formula: ， in, is the adjusted detection distance, is the initial detection distance, is the object speed, is the adjustment factor; When the object speed drops below the preset speed threshold, the ultrasonic sensor's default detection distance is restored.

2. The intelligent safety control system for rolling shutter doors according to claim 1, characterized in that: The prediction control module comprises: A historical data storage unit, used for acquiring and storing historical data of multiple sensor modules; A data preprocessing unit, used to preprocess historical data to obtain training data, the data preprocessing unit includes a data cleaning module, a data normalization module and a feature extraction module, wherein the data cleaning module is used to remove abnormal data, the data normalization module is used to standardize the data to a preset range, and the feature extraction module is used to extract feature parameters related to the object's motion trajectory; A GRU model training unit, used to train a preset GRU model through training data to obtain a predictive control model; A real-time prediction unit is used to receive real-time data from multiple sensor modules and input it into a prediction control model to obtain a prediction result; The control instruction generating unit is used to generate a fourth control instruction according to the prediction result.

3. The intelligent safety control system for rolling shutter doors according to claim 2, characterized in that: The preset GRU model is trained through training data to obtain a predictive control model, including: The training goal is to set the prediction error between the actual motion trajectory of the object and the prediction result of the GRU model to be less than a first preset threshold; The training data is grouped by batch processing to obtain a training data set, and the training data set is input into the GRU model to obtain the model prediction value; Calculate the prediction error in each training iteration according to the loss function; The parameters of the GRU model are adjusted until the training target is achieved and then the training is stopped to obtain a predictive control model.

4. The intelligent safety control system for rolling shutter doors according to claim 3 is characterized in that: The loss function is calculated as follows: ， in, is the prediction error, is the true value, is the model prediction value, is the sample size, For the The predicted distance between the object and the rolling door at a certain moment, is the set safety distance threshold, , and is the weight coefficient.

5. The intelligent safety control system for rolling shutter doors according to claim 4, characterized in that: Adjust the parameters of the GRU model, including: Calculate the partial derivative of the loss function with respect to each parameter of the GRU model to obtain the gradient value of the current iteration; Perform an exponentially weighted moving average on the gradient value to obtain a first-order momentum estimate; Perform an exponentially weighted moving average on the squared gradient values ​​to get the second-order momentum estimate; Calculate the adaptive learning rate based on the first-order momentum estimate and the second-order momentum estimate; Adjust the parameters of the GRU model based on the gradient value and adaptive learning rate.

6. The intelligent safety control system for rolling shutter doors according to claim 5, characterized in that: The infrared sensor comprises: Light sensing module, used to detect the light intensity in the environment in real time; The first control module is used to receive the light intensity and automatically adjust the sensitivity of the infrared sensor according to the change of the light intensity; The sensitivity of the infrared sensor is adjusted by the following formula: ， in, is the sensitivity of the infrared sensor, is the current ambient light intensity, For maximum sensitivity, is the light intensity threshold, is the sensitivity adjustment factor.

7. The intelligent safety control system for rolling shutter doors according to claim 6, characterized in that: The fault feedback module comprises: A fault diagnosis unit is used to obtain the working status parameters of the infrared sensor, ultrasonic sensor and pressure sensor and the operating parameters of the rolling door motor in real time; an intelligent analysis unit, for comparing the working state parameter and the operating parameter with the corresponding rated parameter to obtain a parameter difference, and comparing the parameter difference with a third preset threshold to generate fault feedback information of different levels; A wireless communication module is used to receive fault feedback information, generate a fault report according to the fault feedback information, automatically send an alarm signal of the fault report to the user terminal, and trigger the emergency stop mechanism of the rolling door according to the level of the fault feedback information; The trend analysis unit is used to obtain a fault trend analysis report based on the historical data of the fault feedback information and send the trend analysis report to the user.

8. The intelligent safety control system for rolling shutter doors according to claim 7, characterized in that: The fault feedback module also includes: A self-repair prompt module, which is used to generate a repair plan based on historical fault feedback information and repair records when the level of the fault feedback information is lower than a preset level, and send the repair steps through a preset terminal device; A fault upgrade unit, which is used to automatically upgrade the level of fault feedback information when the repair plan is ineffective, and send a fault report to maintenance personnel through a wireless communication module; The fault library update unit is used to automatically update the fault library according to user feedback and repair records.

Images