Temperature sensor and intelligent analysis method thereof

Abstract

The invention provides a temperature sensor and an intelligent analysis method thereof, and belongs to the technical field of temperature sensors.The temperature sensor comprises a temperature conducting shell, a temperature sensor body is arranged in the temperature conducting shell, and when the temperature sensor body is used normally, heat is conducted to the surface of the temperature conducting shell through a plurality of heat transfer rods; then cooling liquid is transmitted to two main flow pipes, two main flow shells and two main heat dissipation pipes through a liquid supply mechanism, when the liquid passes through the main heat dissipation pipes, heat existing on the surface of the temperature conduction shell is removed, heat of the temperature sensor body is dissipated, and meanwhile when the heat on the surface of the temperature conduction shell is too much, the heat of the temperature sensor body is dissipated; when the temperature sensor body is overheated, a thermal shrinkage material arranged in the secondary flow shell retracts to pull back a blocking block, at the moment, liquid in the two primary flow shells is conducted to the secondary flow shell and a secondary heat dissipation pipe through two transmission pipes, the heat dissipation effect of the temperature conduction shell is further improved through flowing of the liquid in the secondary heat dissipation pipe, and therefore overheating of the temperature sensor body is reduced, and the service life of the temperature sensor body is prolonged. And the reduction of the temperature detection precision is prevented.

Description

Technical Field

[0001] The present invention belongs to the technical field of temperature sensors, and in particular relates to a temperature sensor and an intelligent analysis method thereof. Background Art

[0002] A temperature sensor is a device that can sense changes in the temperature of the surrounding environment or medium and convert these changes into usable output signals (such as electrical signals). It is the core component of the temperature measurement system and is widely used in many fields such as industry, scientific research, medical treatment, home appliances, and environmental monitoring.

[0003] Temperature sensors are widely used in various fields, among which IC temperature sensors are the most widely used. IC temperature sensors will generate heat when working, which may affect the measurement accuracy, especially in high temperature environments, where the self-heating effect is more significant. The existing IC temperature sensors lack cooling devices, which will further affect the self-heating phenomenon of IC temperature sensors, thereby causing the temperature measurement accuracy to continue to decline. Summary of the invention

[0004] The object of the present invention is to provide a temperature sensor and an intelligent analysis method thereof, aiming to solve the problem that the IC type temperature sensor in the prior art lacks a cooling device, which further affects the self-heating phenomenon of the IC type temperature sensor and leads to a continuous decrease in the temperature measurement accuracy.

[0005] To achieve the above object, the present invention provides the following technical solutions: A temperature sensor, comprising: A temperature conducting shell, wherein a temperature sensor body is arranged in the temperature conducting shell; A plurality of heat transfer rods, wherein the plurality of heat transfer rods are fixedly connected to the outer surface of the temperature sensor body and the inner surface of the temperature conductive shell at the adjacent ends; A plurality of heat dissipation holes, wherein the plurality of heat dissipation holes are opened at both ends of the heat conducting shell, and two main heat dissipation pipes and a secondary heat dissipation pipe are arranged in the plurality of heat dissipation holes; Two main flow shells, both of which are connected to the heat transfer shell, and the two main flow shells are respectively connected and communicated with two main heat dissipation pipes; A secondary flow shell, the secondary flow shell is fixedly connected to one end of the heat transfer shell, and the secondary flow shell is connected and communicated with the secondary heat dissipation pipe; Two transmission pipes, the two transmission pipes are fixedly connected to the adjacent ends of the secondary flow shell and the two primary flow shells; A movable rod, the movable rod movably penetrates one end of the secondary flow shell, one end of the movable rod is fixedly connected to a force-bearing plate, and a spring is sleeved on the circumferential surface of the movable rod; and A blocking block is fixedly connected to one end of the movable rod.

[0006] As a preferred solution of the present invention, it also includes a liquid supply mechanism, which is composed of a mounting rod, a mounting shell, a pump body, a liquid storage shell and two main flow pipes, the two main flow pipes are respectively fixedly connected to the two main flow shells, the mounting rod is fixedly connected to one end of the temperature conduction shell, the liquid storage shell is fixedly connected to one end of multiple mounting rods, the two main flow pipes are respectively connected to the upper and lower ends of the liquid storage shell, the mounting shell is fixedly connected to one end of the liquid storage shell, the pump body is fixedly connected to an inner wall of one side of the mounting shell, and the pump body is connected to the liquid storage shell.

[0007] As a preferred solution of the present invention, it also includes a wiping mechanism, which consists of a connecting block, two L-rods, two double-sided racks of wiping blocks, two half-gears, two rotating rods, two I-shaped wheels, a transmission belt and a driving mechanism. The two half-gears are rotatably connected to the upper end of the temperature conducting shell through two rotating rods respectively, the double-sided rack is slidably arranged on the upper side of the temperature conducting shell, and the double-sided rack is gap-engaged with the two half-gears, the two I-shaped wheels are fixedly connected to the two rotating rods respectively, and the circumferential surfaces of the two I-shaped wheels are transmission-connected with a transmission belt, the connecting block is fixedly connected to the lower end of the double-sided rack, the two L-rods are fixedly connected to the two ends of the connecting block respectively, and the two wiping blocks are fixedly connected to the lower ends of the two L-rods respectively.

[0008] As a preferred solution of the present invention, the driving assembly consists of two support rods, a suspension plate and a motor, the two support rods are fixedly connected to the upper end of the temperature conducting shell, the suspension plate is fixedly connected to the upper ends of the two support rods, the motor is fixedly connected to the lower end of the suspension plate, and the output end of the motor is fixedly connected to one of the rotating rods.

[0009] As a preferred solution of the present invention, one end of the heat conducting shell is fixedly connected to a protective shell, and both ends of the protective shell are provided with a plurality of heat dissipation grooves.

[0010] As a preferred solution of the present invention, a plurality of connection grooves are provided at the lower end of the temperature conducting shell, a plurality of connection wires are fixedly connected to the lower end of the temperature sensor body, and the plurality of connection wires are respectively arranged in the plurality of connection grooves.

[0011] As a preferred solution of the present invention, a stabilizing groove is provided at the upper end of the heat conducting shell, a stabilizing rod is fixedly connected to the lower end of the double-sided rack, and the stabilizing rod is slidably connected in the stabilizing groove.

[0012] As a preferred solution of the present invention, an anti-drop rod is fixedly connected between the inner walls of both sides of the stabilizing groove, a limiting hole is opened at one end of the stabilizing rod, and the stabilizing rod is slidably connected to the circumferential surface of the anti-drop rod through the limiting hole.

[0013] As a preferred solution of the present invention, one end of the two mainstream shells is fixedly connected to a connecting rod, one end of the two connecting rods is fixedly connected to a thermal insulation board, and the thermal insulation board is fixedly connected to the thermal conduction shell.

[0014] As a preferred solution of the present invention, a plurality of temperature conducting rods are fixedly connected to one end of the secondary flow shell, a temperature conducting plate is fixedly connected to one end of the plurality of temperature conducting rods, and the temperature conducting plate is fixedly connected to the temperature conducting shell.

[0015] A temperature sensor, characterized in that by integrating support vector machine (SVM) and long short-term memory network (LSTM) machine learning algorithms, we can significantly improve the data analysis and prediction capabilities of the temperature sensor, which enables the sensor to respond to temperature changes in complex environments in real time and provide more accurate and reliable data processing functions: Step 1: Data collection and preprocessing: Collect real-time data of the temperature sensor: data of the temperature sensor body (106), heat transfer rod (103), heat dissipation hole (101), main heat dissipation pipe (102), secondary heat dissipation pipe (109), transmission pipe (204), movable rod (203), force plate (201) and blocking block (205). Ensure data quality and integrity, and perform preprocessing: data cleaning, denoising and normalization; Step 2 Feature Engineering: According to the needs of temperature trend prediction, appropriate features include: Temperature sensor readings : Temperature data measured by the sensor body; Ambient temperature : The temperature of the surrounding environment affects the sensor readings; Heat transfer performance of heat transfer rod : The thermal conductivity of the heat transfer rod material affects the heat transfer rate; Thermal conductivity characteristics of heat dissipation holes : The thermal conductivity of the heat dissipation hole affects the heat dissipation efficiency; The goal of feature engineering is to accurately reflect the relationships and impacts between sensor components; Step 3: Implementation of Support Vector Machine (SVM): Model definition: SVM classifies data by finding an optimal hyperplane; Classification model formula: ; is the classification function, is the symbolic function, is the sum of all support vectors i, is the Lagrange multiplier in the support vector machine, For training samples The category label of is the kernel function used to calculate the similarity or inner product between support vectors. This is the bias term used to adjust the position of the classification hyperplane; Training and Optimization: Using the training set ; is the input feature vector, is the corresponding output value, is a set containing m elements; By minimizing the regularization loss function ; is the weight vector The square of the norm of The regularization parameter controls the penalty strength of the model to balance the maximization of the interval and the penalty of misclassified samples. is the summation operation of the training samples, Hinge Loss function is used to measure the degree of classification error. For sample The real label, is the weight vector, For the sample, is the bias term, The decision function of SVM is used to calculate the sample 's classification score; The goal of SVM is to find a hyperplane that maximizes the margin while satisfying: ; is the weight vector of the SVM model used to define the direction of the hyperplane, The bias term of the SVM model determines the position of the hyperplane. is the weight vector The square of the norm of is a regularization term used to penalize the complexity of the model. The regularization parameter adjusts the complexity of the model, is the sum of the slack variables, ≥0 is a slack variable; The prediction function of the support vector machine: ; is the predicted output value, For the summation symbol, is the Lagrange multiplier of the support vector machine, For training samples The corresponding true output value is is the Gaussian kernel function used to measure the eigenvector and The similarities between is the bandwidth parameter of the Gaussian kernel, For training samples Eigenvector, For training samples The characteristic vector of is the feature vector and The Euclidean distance between is the bias term; Model training and evaluation: Use the preprocessed data set for training, and use cross-validation and other techniques to optimize the parameter C and kernel function selection of the SVM model, evaluate the prediction performance of the model on the test data set, and ensure that it can accurately predict the temperature change trend; Real-time prediction and application: Integrating the trained SVM model into the real-time data processing flow enables it to analyze and predict the changing trends of temperature sensor data in real time, helping to respond to temperature changes in a timely manner and provide accurate data processing capabilities; Step 4 Implementation process of long short-term memory network (LSTM): Model definition: LSTM is a recurrent neural network (RNN) suitable for sequence data. It is particularly good at capturing long-term dependencies in data. The LSTM unit includes a forget gate, an input gate, an output gate, and a hidden state. and cell status The update formula is as follows: Forget Gate: ; For the Gate of Forgetfulness, is the sigmoid function, is the weight of the forget gate, is the hidden state of the previous time step, is the current input, is the bias of the forget gate; Input Gate: ; ; is the input gate, is the sigmoid function, is the weight of the input gate, is the hidden state of the previous time step, is the current input, is the weight bias of the input gate, is the candidate cell state, is the hyperbolic tangent function, is the weight of the candidate cell state, is the bias of the candidate cell state; Cell status update: ; is the cell state, For the Gate of Forgetfulness, is element-wise multiplication, is the cell state at the previous time step, is the input gate, is the candidate cell state; Output Gate: ; ; is the output gate, is the sigmoid function, is the weight of the output gate, is the hidden state of the previous time step, is the current input, is the bias of the output gate, is the hidden state, is the hyperbolic tangent function, is the cell state; Training and Optimization: Training the LSTM model using historical data sequences ; Contains all from t=1 to T A collection of elements, is the element at the tth time step in the sequence; The network parameters are optimized by minimizing the loss function. The training process updates the weights and biases of the model through backpropagation through time (BPTT), thereby improving the model's ability to predict future temperature trends. Step 5 Temperature adjustment device: Design and implement an executive controller to regulate the operation and response of the temperature sensor based on the temperature trend predicted by the machine learning model; 1. Control logic design: Threshold-based control: Set a temperature threshold ,If the real-time temperature exceeds or approaches the threshold, corresponding control measures are taken; Control logic: : Reduce the heat dissipation rate or increase the contact area of ​​the heat transfer rod; is the temperature reading at the current moment; PID Controller: The PID controller adjusts the control output through proportion, integration and differentiation to make the system's feedback signal close to the target value; Controller formula: ; is the output of the controller, is the proportional gain parameter, is the error signal, is the integral gain parameter, From the start time to the current time The integral of the error signal, is the differential gain parameter, is the error signal rate of change over time; 2. Control execution: Implement control logic to monitor real-time temperature data and machine learning predictive outputs; Adjust the sensor’s operating parameters and environmental interactions based on the predicted temperature change trend to achieve stable temperature control and optimize energy efficiency; 3. Feedback mechanism: Introduce feedback mechanisms to adjust controller parameters or strategies in real time to cope with changes and uncertainties in the actual environment; By continuously monitoring and analyzing sensor output data, controller performance is optimized and reliable operation of the sensor is ensured under various operating conditions.

[0016] Compared with the prior art, the present invention has the following beneficial effects: 1. When the temperature sensor body is in normal use, the heat is transferred to the surface of the temperature conducting shell through multiple heat transfer rods, and then the coolant is transferred to the two main flow tubes, two main flow shells and two main heat dissipation tubes through the liquid supply mechanism. When the liquid passes through the main heat dissipation tube, the heat on the surface of the temperature conducting shell is removed, and the heat of the temperature sensor body is also dissipated. At the same time, when there is too much heat on the surface of the temperature conducting shell, the heat shrink material set in the slave flow shell will shrink and pull back the blocking block. At this time, the liquid in the two main flow shells will be transferred to the slave flow shell and the slave heat dissipation tube through the two transmission tubes. Through the flow of liquid in the slave heat dissipation tube, the heat dissipation effect of the temperature conducting shell is further increased, thereby reducing the overheating of the temperature sensor body and preventing the decrease in the accuracy of its temperature detection.

[0017] 2. When the temperature of the thermal conductive shell and the temperature sensor body is still high after heat dissipation, the wiping mechanism arranged on the outside of the thermal conductive shell can be started. By means of physical wiping, the surface temperature of the thermal conductive shell can be continuously reduced, thereby further enhancing the heat dissipation and temperature stability of the thermal conductive shell and the temperature sensor body and preventing a series of problems caused by overheating.

[0018] 3. The connecting wires can be placed in multiple connection grooves to facilitate the connection between the temperature sensor body and the external device. The multiple devices on the left side of the temperature conducting shell can be protected by the protective shell to prevent damage to the devices or components caused by collision.

[0019] 4. Improved the overall heat dissipation efficiency and stability of the temperature sensor: Through the combination of multiple heat transfer rods, main heat dissipation pipes and liquid cooling systems, the present invention significantly improves the heat dissipation efficiency of the temperature conductive shell and the temperature sensor body. Compared with traditional technologies, this structure can effectively reduce the operating temperature of the temperature conductive shell and the sensor body, thereby maintaining its temperature detection accuracy for a long time. Especially in high temperature environments, the heat dissipation effect is further enhanced through the use of precise liquid cooling and heat shrink materials, effectively preventing the performance degradation and damage risks caused by overheating.

[0020] The beneficial effects of the present invention significantly improve the performance and reliability of temperature sensors in various application scenarios. First, through the combination of multiple heat transfer rods and a liquid cooling system, the heat dissipation efficiency of the temperature conductive shell and the temperature sensor body is effectively improved. In normal use, these designs can quickly transfer heat from the sensor body to the surface of the temperature conductive shell, and dissipate the heat through the liquid cooling system to prevent overheating, thereby maintaining the accuracy and stability of the temperature sensor. Secondly, in response to the situation where the temperature is too high, the present invention adopts an external wiping mechanism to reduce the surface temperature of the temperature conductive shell by physical wiping, further enhancing the heat dissipation effect and ensuring the reliable operation of the sensor under extreme working conditions. In addition, the configuration of the connecting groove and the protective shell is also considered in the design, so that the temperature sensor body can be more conveniently connected to the external device, and the key components can be effectively protected from collision damage. In summary, the present invention not only improves the heat dissipation capacity and stability of traditional temperature sensors, but also improves its applicability and long-term reliability in complex environments, providing important technical support and guarantee for modern industrial automation and intelligent applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1 It is the main stereoscopic view in the present invention; Figure 2 It is a top perspective view of the present invention; Figure 3 It is the first main sectional stereogram in the present invention; Figure 4 For the present invention Figure 3 A partial enlarged view of the middle A; Figure 5 It is a second main sectional stereoscopic view of the present invention; Figure 6 It is a top-down perspective view of the present invention; Figure 7 It is a side cutaway stereogram of the present invention; Figure 8 For the present invention Figure 7 A partial enlarged view of point B in the middle; Fig. 9 This is an experimental flow chart of the intelligent temperature control system in the present invention.

[0022] In the figure: 1. temperature conducting shell; 101. heat dissipation hole; 102. main heat dissipation pipe; 103. heat transfer rod; 104. connection groove; 105. connection wire; 106. temperature sensor body; 107. protection shell; 108. heat dissipation groove; 109. secondary heat dissipation pipe; 110. installation shell; 111. liquid storage shell; 112. installation rod; 113. pump body; 114. main flow pipe; 115. main flow shell; 2. secondary flow shell ; 201, force plate; 202, spring; 203, movable rod; 204, transmission tube; 205, blocking block; 3, double-sided rack; 301, support rod; 302, suspension plate; 303, motor; 304, half gear; 305, I-shaped wheel; 306, transmission belt; 307, stabilizing groove; 308, stabilizing rod; 309, anti-drop rod; 310, connecting block; 311, L rod; 312, wiping block. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. Example 1

[0024] See also Figure 1-8 , the present invention provides the following technical solutions: A temperature sensor, comprising: A temperature conducting shell 1, in which a temperature sensor body 106 is arranged; A plurality of heat transfer rods 103, wherein the plurality of heat transfer rods 103 are fixedly connected to the outer surface of the temperature sensor body 106 and the inner surface of the temperature conductive shell 1 close to each other; A plurality of heat dissipation holes 101, wherein the plurality of heat dissipation holes 101 are opened at both ends of the heat conducting shell 1, and two main heat dissipation pipes 102 and a secondary heat dissipation pipe 109 are arranged in the plurality of heat dissipation holes 101; Two mainstream shells 115, both of which are connected to the heat conducting shell 1, and the two mainstream shells 115 are respectively connected and communicated with the two main heat dissipation pipes 102; A secondary flow shell 2, which is fixedly connected to one end of the heat transfer shell 1, and is connected and communicated with the secondary heat dissipation pipe 109; Two transmission pipes 204, the two transmission pipes 204 are fixedly connected to the adjacent ends of the secondary flow shell 2 and the two primary flow shells 115; A movable rod 203, the movable rod 203 movably penetrates one end of the secondary flow shell 2, and one end of the movable rod 203 is fixedly connected to the force-bearing plate 201, and a spring 202 is sleeved on the circumferential surface of the movable rod 203; and The blocking block 205 is fixedly connected to one end of the movable rod 203, and also includes a liquid supply mechanism, which is composed of a mounting rod 112, a mounting shell 110, a pump body 113, a liquid storage shell 111 and two mainstream pipes 114. The two mainstream pipes 114 are respectively fixedly connected to the two mainstream shells 115. The mounting rod 112 is fixedly connected to one end of the thermal conductive shell 1, the liquid storage shell 111 is fixedly connected to one end of multiple mounting rods 112, the two mainstream pipes 114 are respectively connected to the upper and lower ends of the liquid storage shell 111, the mounting shell 110 is fixedly connected to one end of the liquid storage shell 111, the pump body 113 is fixedly connected to an inner wall of one side of the mounting shell 110, and the pump body 113 is connected to the liquid storage shell 111.

[0025] In a specific embodiment of the present invention, when the temperature sensor body 106 works normally, the heat it emits will be transferred to the surface of the temperature conducting shell 1 through the multiple heat transfer rods 103, and then the pump body 113 will be started to pump the liquid in the liquid storage shell 111 into the two main flow pipes 114, and the coolant will flow into the main heat dissipation pipe 102 through the main flow pipe 114 and the main flow shell 115. When the coolant passes through the main heat dissipation pipe 102, it will take away the heat on the surface of the temperature conducting shell 1 and remove the heat of the temperature sensor body 106, thereby ensuring the temperature stability of the temperature sensor body 106. When the temperature of the temperature sensor body 106 is too high, its temperature will be transferred to the slave flow shell 2. When the temperature of the slave flow shell 2 is too high, the heat shrink material filled therein will shrink when encountering high temperature, and at the same time, the blocking block 205 will be moved to Pull it to the left, and then the coolant is conducted to the flow shell 2 through the two transmission tubes 204, and the coolant is transmitted to the heat dissipation pipe 109. Through the flow of liquid in the heat dissipation pipe 109, the heat of the temperature conduction shell 1 and the temperature sensor body 106 is removed again. Through the above design, the temperature of the temperature sensor body 106 can be cooled and regulated in real time to prevent the temperature of the temperature sensor body 106 from being too high, which leads to a decrease in the temperature detection accuracy, thereby improving its detection accuracy. The spring 202 can pull the force plate 201 and the movable rod 203 to the right, and reset the blocking block 205 to block the transmission of the transmission tube 204, preventing the coolant from being transmitted from the heat dissipation pipe 109 again, thereby preventing the temperature from being too low, which leads to a low temperature of the temperature sensor body 106, thereby improving the heat dissipation adaptability of the device.

[0026] For details, please refer to Figure 1-Figure 8, and also includes a wiping mechanism, which is composed of a connecting block 310, two L rods 311, two wiping blocks 312, a double-sided rack 3, two half gears 304, two rotating rods, two I-shaped wheels 305, a transmission belt 306 and a driving mechanism. The two half gears 304 are respectively rotatably connected to the upper end of the heat conducting shell 1 through two rotating rods, the double-sided rack 3 is slidably arranged on the upper side of the heat conducting shell 1, and the double-sided rack 3 is gap-engaged with the two half gears 304, the two I-shaped wheels 305 are respectively fixedly connected to the two rotating rods, and the circumferential surfaces of the two I-shaped wheels 305 are connected with the transmission belt 3 06, the connecting block 310 is fixedly connected to the lower end of the double-sided rack 3, the two L rods 311 are respectively fixedly connected to the two ends of the connecting block 310, and the two wiping blocks 312 are respectively fixedly connected to the lower ends of the two L rods 311. The driving assembly consists of two support rods 301, a suspension plate 302 and a motor 303. The two support rods 301 are both fixedly connected to the upper end of the heat conducting shell 1, the suspension plate 302 is fixedly connected to the upper ends of the two support rods 301, the motor 303 is fixedly connected to the lower end of the suspension plate 302, and the output end of the motor 303 is fixedly connected to one of the rotating rods.

[0027] In this embodiment: when the temperature sensor body 106 is cooled and the temperature is still high, the temperature of the temperature conducting shell 1 will also rise. At this time, the motor 303 can be started so that its output end drives one of the rotating rods to rotate, and then the two half gears 304 are driven to rotate synchronously through the transmission belt 306 and the two I-shaped wheels 305, and intermittently mesh with the double-sided rack 3, driving the double-sided rack 3 to move back and forth. At the same time, the double-sided rack 3 will drive the connecting block 310, the two L rods 311 and the two wiping blocks 312 to continuously wipe the upper part of the temperature conducting shell 1, and the wiping block 312 itself is soaked in alcohol or other coolant. After the liquid is wiped to the upper part of the temperature conducting shell 1, it will evaporate quickly, and the temperature of the surface of the temperature conducting shell 1 will be removed by evaporation. Through the above design, the temperature of the temperature conducting shell 1 is further reduced, and then the temperature of the temperature sensor body 106 is reduced, and the problem of insufficient detection accuracy caused by excessive temperature of the temperature sensor body 106 is further prevented.

[0028] For details, please refer to Figure 1-Figure 8 One end of the thermal conductive shell 1 is fixedly connected to a protective shell 107, and both ends of the protective shell 107 are provided with a plurality of heat dissipation grooves 108. A plurality of connecting grooves 104 are provided at the lower end of the thermal conductive shell 1, and a plurality of connecting wires 105 are fixedly connected to the lower end of the temperature sensor body 106, and the plurality of connecting wires 105 are respectively arranged in the plurality of connecting grooves 104.

[0029] In this embodiment: the protective shell 107 can be used to protect multiple components on one side of the temperature conducting shell 1, the heat dissipation of the protective shell 107 and internal components can be improved through the heat dissipation grooves 108, and the temperature sensor body 106 can be used to conveniently extend the connecting wires 105 to the outside through the multiple connecting grooves 104, so as to facilitate the connection between the temperature sensor body 106 and external electrical equipment for data transmission, etc.

[0030] For details, please refer to Figure 1-Figure 8 A stabilizing groove 307 is provided at the upper end of the heat conducting shell 1, a stabilizing rod 308 is fixedly connected to the lower end of the double-sided rack 3, the stabilizing rod 308 is slidably connected in the stabilizing groove 307, an anti-drop rod 309 is fixedly connected between the inner walls on both sides of the stabilizing groove 307, a limiting hole is provided at one end of the stabilizing rod 308, and the stabilizing rod 308 is slidably connected to the circumferential surface of the anti-drop rod 309 through the limiting hole.

[0031] In this embodiment: when the double-sided rack 3 slides, it will drive the stabilizing rod 308 to slide in the stabilizing groove 307, thereby improving the horizontal sliding stability of the double-sided rack 3, and when the stabilizing rod 308 moves, it can slide on the circumferential surface of the anti-drop rod 309 through the limiting hole, thereby improving the sliding stability of the stabilizing rod 308 in the stabilizing groove 307 and preventing the stabilizing rod 308 from detaching from the stabilizing groove 307.

[0032] For details, please refer to Figure 1-Figure 8 One end of the two main flow shells 115 is fixedly connected with a connecting rod, one end of the two connecting rods is fixedly connected with a thermal insulation plate, the thermal insulation plate is fixedly connected to the thermal conduction shell 1, and one end of the flow shell 2 is fixedly connected with a plurality of thermal conduction rods, one end of the plurality of thermal conduction rods is fixedly connected with a thermal conduction plate, and the thermal conduction plate is fixedly connected to the thermal conduction shell 1.

[0033] In this embodiment: the mainstream shell 115 is connected to the thermal conductive shell 1 through a connecting rod and a thermal insulation plate, which in turn prevents heat from being transmitted to the mainstream shell 115, and the slave shell 2 is connected to the thermal conductive shell 1 through multiple thermal conductive rods and thermal conductive plates, which can conveniently conduct heat to the slave shell 2 and facilitate the induction of the heat shrinkable material in the slave shell 2.

[0034] The working principle and use process of the present invention are as follows: when the temperature sensor body 106 works normally, the heat it emits will be transferred to the surface of the temperature conducting shell 1 through the multiple heat transfer rods 103, and then the pump body 113 will be started to pump the liquid in the liquid storage shell 111 into the two main flow pipes 114, and the coolant will flow into the main heat dissipation pipe 102 through the main flow pipe 114 and the main flow shell 115. When the coolant passes through the main heat dissipation pipe 102, it will take away the heat on the surface of the temperature conducting shell 1 and remove the heat of the temperature sensor body 106, thereby ensuring the temperature stability of the temperature sensor body 106. When the temperature of the temperature sensor body 106 is too high, its temperature will be transferred to the slave flow shell 2. When the temperature of the slave flow shell 2 is too high, the heat shrink material filled therein will shrink when encountering high temperature, and at the same time, the blocking block 205 will be blocked. Pull it to the left, and then the coolant is conducted to the flow shell 2 through the two transmission tubes 204, and the coolant is transmitted to the heat dissipation pipe 109. The heat of the temperature conduction shell 1 and the temperature sensor body 106 is removed again by the flow of liquid in the heat dissipation pipe 109. Through the above design, the temperature of the temperature sensor body 106 can be cooled and regulated in real time to prevent the temperature of the temperature sensor body 106 from being too high, which may lead to a decrease in the temperature detection accuracy, thereby improving its detection accuracy. The spring 202 can pull the force plate 201 and the movable rod 203 to the right, and reset the blocking block 205 to block the transmission of the transmission tube 204, preventing the coolant from being transmitted from the heat dissipation pipe 109 again, thereby preventing the temperature from being too low, which may lead to a low temperature of the temperature sensor body 106, thereby improving the heat dissipation adaptability of the device.

[0035] A temperature sensor, characterized in that by integrating support vector machine (SVM) and long short-term memory network (LSTM) machine learning algorithms, we can significantly improve the data analysis and prediction capabilities of the temperature sensor, which enables the sensor to respond to temperature changes in complex environments in real time and provide more accurate and reliable data processing functions: Step 1: Data collection and preprocessing: Collect real-time data of the temperature sensor: data of the temperature sensor body (106), heat transfer rod (103), heat dissipation hole (101), main heat dissipation pipe (102), secondary heat dissipation pipe (109), transmission pipe (204), movable rod (203), force plate (201) and blocking block (205). Ensure data quality and integrity, and perform preprocessing: data cleaning, denoising and normalization; Step 2 Feature Engineering: According to the needs of temperature trend prediction, appropriate features include: Temperature sensor readings : Temperature data measured by the sensor body; Ambient temperature : The temperature of the surrounding environment affects the sensor readings; Heat transfer performance of heat transfer rod : The thermal conductivity of the heat transfer rod material affects the heat transfer rate; Thermal conductivity characteristics of heat dissipation holes : The thermal conductivity of the heat dissipation hole affects the heat dissipation efficiency; The goal of feature engineering is to accurately reflect the relationships and impacts between sensor components; Step 3: Implementation of Support Vector Machine (SVM): Model definition: SVM classifies data by finding an optimal hyperplane; Classification model formula: ; is the classification function, is the symbolic function, is the sum of all support vectors i, is the Lagrange multiplier in the support vector machine, For training samples The category label of is the kernel function used to calculate the similarity or inner product between support vectors. This is the bias term used to adjust the position of the classification hyperplane; Training and Optimization: Using the training set ; is the input feature vector, is the corresponding output value, is a set containing m elements; By minimizing the regularization loss function ; is the weight vector The square of the norm of The regularization parameter controls the penalty strength of the model to balance the maximization of the interval and the penalty of misclassified samples. is the summation operation of the training samples, Hinge Loss function is used to measure the degree of classification error. For sample The real label, is the weight vector, For the sample, is the bias term, The decision function of SVM is used to calculate the sample 's classification score; The goal of SVM is to find a hyperplane that maximizes the margin while satisfying: ; is the weight vector of the SVM model used to define the direction of the hyperplane, The bias term of the SVM model determines the position of the hyperplane. is the weight vector The square of the norm of is a regularization term used to penalize the complexity of the model. The regularization parameter adjusts the complexity of the model, is the sum of the slack variables, ≥0 is a slack variable; The prediction function of the support vector machine: ; is the predicted output value, For the summation symbol, is the Lagrange multiplier of the support vector machine, For training samples The corresponding true output value is is the Gaussian kernel function used to measure the eigenvector and The similarities between is the bandwidth parameter of the Gaussian kernel, For training samples Eigenvector, For training samples The characteristic vector of is the feature vector and The Euclidean distance between is the bias term; Model training and evaluation: Use the preprocessed data set for training, and use cross-validation and other techniques to optimize the parameter C and kernel function selection of the SVM model, evaluate the prediction performance of the model on the test data set, and ensure that it can accurately predict the temperature change trend; Real-time prediction and application: Integrating the trained SVM model into the real-time data processing flow enables it to analyze and predict the changing trends of temperature sensor data in real time, helping to respond to temperature changes in a timely manner and provide accurate data processing capabilities; Step 4 Implementation process of long short-term memory network (LSTM): Model definition: LSTM is a recurrent neural network (RNN) suitable for sequence data. It is particularly good at capturing long-term dependencies in data. The LSTM unit includes a forget gate, an input gate, an output gate, and a hidden state. and cell status The update formula is as follows: Forget Gate: ; For the Gate of Forgetfulness, is the sigmoid function, is the weight of the forget gate, is the hidden state of the previous time step, is the current input, is the bias of the forget gate; Input Gate: ; ; is the input gate, is the sigmoid function, is the weight of the input gate, is the hidden state of the previous time step, is the current input, is the weight bias of the input gate, is the candidate cell state, is the hyperbolic tangent function, is the weight of the candidate cell state, is the bias of the candidate cell state; Cell status update: ; is the cell state, For the Gate of Forgetfulness, is element-wise multiplication, is the cell state at the previous time step, is the input gate, is the candidate cell state; Output Gate: ; ; is the output gate, is the sigmoid function, is the weight of the output gate, is the hidden state of the previous time step, is the current input, is the bias of the output gate, is the hidden state, is the hyperbolic tangent function, is the cell state; Training and Optimization: Training the LSTM model using historical data sequences ; Contains all from t=1 to T A collection of elements, is the element at the tth time step in the sequence; The network parameters are optimized by minimizing the loss function. The training process updates the weights and biases of the model through backpropagation through time (BPTT), thereby improving the model's ability to predict future temperature trends. Step 5 Temperature adjustment device: Design and implement an executive controller to regulate the operation and response of the temperature sensor based on the temperature trend predicted by the machine learning model; 1. Control logic design: Threshold-based control: Set a temperature threshold ,If the real-time temperature exceeds or approaches the threshold, corresponding control measures are taken; Control logic: : Reduce the heat dissipation rate or increase the contact area of ​​the heat transfer rod; is the temperature reading at the current moment; PID Controller: The PID controller adjusts the control output through proportion, integration and differentiation to make the system's feedback signal close to the target value; Controller formula: ; is the output of the controller, is the proportional gain parameter, is the error signal, is the integral gain parameter, From the start time to the current time The integral of the error signal, is the differential gain parameter, is the error signal rate of change over time; 2. Control execution: Implement control logic to monitor real-time temperature data and machine learning predictive outputs; Adjust the sensor’s operating parameters and environmental interactions based on the predicted temperature change trend to achieve stable temperature control and optimize energy efficiency; 3. Feedback mechanism: Introduce feedback mechanisms to adjust controller parameters or strategies in real time to cope with changes and uncertainties in the actual environment; By continuously monitoring and analyzing sensor output data, controller performance is optimized and reliable operation of the sensor is ensured under various operating conditions.

[0036] Step 1: Data collection and preparation 1. Data Collection: Collect real-time data of temperature sensor bodies, heat transfer rods, heat dissipation holes and other components. This data can include sensor readings, ambient temperature and thermal conductivity characteristics of each component.

[0037] 2. Data preparation: The collected data were preprocessed, including outlier removal, data cleaning, and normalization, to ensure data quality and consistency.

[0038] Step 2: Algorithm selection and model building 1. Support vector machine (SVM) model establishment: Feature selection: Select features related to temperature trend prediction, such as sensor readings x t and ambient temperature T ambient。

[0039] Training model: Use the training set to train the model and optimize the SVM model parameters.

[0040] Data example: Group 1: x t =[20,21,22,...,30],T ambient =[25,26,27,...,35] Group 2: x t =[18,20,21,...,28],T ambient =[23,25,26,...,33] Group 3: x t =[22,23,24,...,32],T ambient =[27,28,29,...,37] 2. Long short-term memory network (LSTM) model establishment: Data series preparation: Construct appropriate time series data, including historical temperature sensor readings and ambient temperature data.

[0041] Model training: Use training data to train the LSTM model and optimize network parameters to predict temperature change trends.

[0042] Data example: Group 1: x t =[20,21,22,...,30],T ambient =[25,26,27,...,35] Group 2: x t =[18,19,20,...,28],T ambient =[23,24,25,...,33] Group 3: x t =[22,23,24,...,32],T ambient =[27,28,29,...,37] Step 3: Model evaluation and real-time prediction 1. Model evaluation: The trained SVM and LSTM models are evaluated using the test set to compare their prediction accuracy and generalization ability on unseen data.

[0043] 2. Real-time prediction: Integrate the trained model into the real-time data processing flow to perform real-time prediction and analysis of temperature change trends.

[0044] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A temperature sensor, characterized in that: include: A temperature conducting shell (1), wherein a temperature sensor body (106) is arranged in the temperature conducting shell (1); A plurality of heat transfer rods (103), wherein the plurality of heat transfer rods (103) are fixedly connected to the outer surface of the temperature sensor body (106) and the inner surface of the temperature conductive shell (1) at the adjacent ends; A plurality of heat dissipation holes (101), wherein the plurality of heat dissipation holes (101) are opened at two ends of the heat conducting shell (1), and two main heat dissipation pipes (102) and a secondary heat dissipation pipe (109) are arranged in the plurality of heat dissipation holes (101); Two main flow shells (115), the two main flow shells (115) are both connected to the temperature conducting shell (1), and the two main flow shells (115) are respectively connected to and communicate with two main heat dissipation pipes (102); A secondary flow shell (2), the secondary flow shell (2) being fixedly connected to one end of the heat transfer shell (1), and the secondary flow shell (2) being connected and communicated with a secondary heat dissipation pipe (109); Two transmission pipes (204), the two transmission pipes (204) being fixedly connected to adjacent ends of the secondary flow shell (2) and the two primary flow shells (115); A movable rod (203), the movable rod (203) movably passing through one end of the secondary flow shell (2), one end of the movable rod (203) being fixedly connected to a force-bearing plate (201), a spring (202) being sleeved on the circumferential surface of the movable rod (203); and a blocking block (205), the blocking block (205) being fixedly connected to one end of the movable rod (203).

2. A temperature sensor according to claim 1, characterized in that: The invention also comprises a liquid supply mechanism, the liquid supply mechanism comprising a mounting rod (112), a mounting shell (110), a pump body (113), a liquid storage shell (111) and two main flow pipes (114), the two main flow pipes (114) being respectively fixedly connected to the two main flow shells (115), the mounting rod (112) being fixedly connected to one end of the thermal conductive shell (1), the liquid storage shell (111) being fixedly connected to one end of a plurality of mounting rods (112), the two main flow pipes (114) being respectively connected to the upper and lower ends of the liquid storage shell (111), the mounting shell (110) being fixedly connected to one end of the liquid storage shell (111), the pump body (113) being fixedly connected to an inner wall of one side of the mounting shell (110), and the pump body (113) being in communication with the liquid storage shell (111).

3. A temperature sensor according to claim 2, characterized in that: The wiping mechanism is also included, the wiping mechanism comprising a connecting block (310), two L-rods (311), two wiping blocks (312), a double-sided rack (3), two half gears (304), two rotating rods, two I-shaped wheels (305), a transmission belt (306) and a driving mechanism, the two half gears (304) are respectively rotatably connected to the upper end of the heat conducting shell (1) through the two rotating rods, the double-sided rack (3) is slidably arranged on the upper side of the heat conducting shell (1), and the double-sided rack ( 3) are meshed with the two half gears (304) in a gap, the two I-shaped wheels (305) are respectively fixedly connected to the two rotating rods, and the circumferential surfaces of the two I-shaped wheels (305) are transmission-connected with a transmission belt (306), the connecting block (310) is fixedly connected to the lower end of the double-sided rack (3), the two L-shaped rods (311) are respectively fixedly connected to the two ends of the connecting block (310), and the two wiping blocks (312) are respectively fixedly connected to the lower ends of the two L-shaped rods (311).

4. A temperature sensor according to claim 3, characterized in that: The driving assembly is composed of two supporting rods (301), a suspension plate (302) and a motor (303), wherein the two supporting rods (301) are fixedly connected to the upper end of the thermal conductive shell (1), the suspension plate (302) is fixedly connected to the upper ends of the two supporting rods (301), the motor (303) is fixedly connected to the lower end of the suspension plate (302), and the output end of the motor (303) is fixedly connected to one of the rotating rods.

5. A temperature sensor according to claim 4, characterized in that: One end of the heat-conducting shell (1) is fixedly connected to a protective shell (107), and both ends of the protective shell (107) are provided with a plurality of heat dissipation grooves (108).

6. A temperature sensor according to claim 5, characterized in that: A plurality of connection grooves (104) are provided at the lower end of the temperature conducting shell (1), a plurality of connection wires (105) are fixedly connected to the lower end of the temperature sensor body (106), and the plurality of connection wires (105) are respectively arranged in the plurality of connection grooves (104).

7. A temperature sensor according to claim 6, characterized in that: The upper end of the heat conducting shell (1) is provided with a stabilizing groove (307), the lower end of the double-sided rack (3) is fixedly connected with a stabilizing rod (308), and the stabilizing rod (308) is slidably connected in the stabilizing groove (307).

8. A temperature sensor according to claim 7, characterized in that: An anti-drop rod (309) is fixedly connected between the inner walls on both sides of the stabilizing groove (307), a limiting hole is provided at one end of the stabilizing rod (308), and the stabilizing rod (308) is slidably connected to the circumferential surface of the anti-drop rod (309) through the limiting hole.

9. A temperature sensor according to claim 8, characterized in that: One end of each of the two main flow shells (115) is fixedly connected to a connecting rod, one end of each of the two connecting rods is fixedly connected to a thermal insulation board, and the thermal insulation board is fixedly connected to the thermal conduction shell (1).

10. A temperature sensor according to claim 9, characterized in that: A plurality of temperature conducting rods are fixedly connected to one end of the secondary flow shell (2), a temperature conducting plate is fixedly connected to one end of the plurality of temperature conducting rods, and the temperature conducting plate is fixedly connected to the temperature conducting shell (1).

11. The intelligent analysis method of a temperature sensor according to claim 1, characterized in that: By integrating the support vector machine (SVM) and the long short-term memory (LSTM) machine learning algorithms, the data analysis and prediction capabilities of the temperature sensor are improved, enabling the sensor to respond to temperature changes in complex environments in real time and provide more accurate and reliable data processing functions: Step 1: collect real-time data of the temperature sensor: data of the temperature sensor body (106), the heat transfer rod (103), the heat dissipation hole (101), the main heat dissipation pipe (102), the secondary heat dissipation pipe (109), the transmission pipe (204), the movable rod (203), the force plate (201) and the blocking block (205), ensure the data quality and integrity, and perform pre-processing: data cleaning, denoising and normalization; Step 2: According to the needs of temperature change trend prediction, select appropriate features including: Temperature sensor readings : Temperature data measured by the sensor body; Ambient temperature : The temperature of the surrounding environment affects the sensor readings; Heat transfer performance of heat transfer rod : The thermal conductivity of the heat transfer rod material affects the heat transfer rate; Thermal conductivity characteristics of heat dissipation holes : The thermal conductivity of the heat dissipation hole affects the heat dissipation efficiency; The goal of feature engineering is to accurately reflect the relationships and impacts between sensor components; Step 3: Implementation process of support vector machine SVM: Model definition: SVM classifies data by finding an optimal hyperplane; Classification model formula: ； is the classification function, is the symbolic function, is the sum of all support vectors i, is the Lagrange multiplier in the support vector machine, For training samples The category label of is the kernel function used to calculate the similarity or inner product between support vectors. This is the bias term used to adjust the position of the classification hyperplane; Training and Optimization: Using the training set ； is the input feature vector, is the corresponding output value, is a set containing m elements; By minimizing the regularization loss function ； is the weight vector The square of the norm of The regularization parameter controls the penalty strength of the model to balance the maximization of the interval and the penalty of misclassified samples. is the summation operation of the training samples, Hinge Loss function is used to measure the degree of classification error. For sample The real label, is the weight vector, For the sample, is the bias term, The decision function of SVM is used to calculate the sample 's classification score; The goal of SVM is to find a hyperplane that maximizes the margin while satisfying: ； is the weight vector of the SVM model used to define the direction of the hyperplane, The bias term of the SVM model determines the position of the hyperplane. is the weight vector The square of the norm of is a regularization term used to penalize the complexity of the model. The regularization parameter adjusts the complexity of the model, is the sum of the slack variables, ≥0 is a slack variable; The prediction function of the support vector machine: ； is the predicted output value, For the summation symbol, is the Lagrange multiplier of the support vector machine, For training samples The corresponding true output value is is the Gaussian kernel function used to measure the eigenvector and The similarities between is the bandwidth parameter of the Gaussian kernel, For training samples Eigenvector, For training samples The characteristic vector of is the feature vector and The Euclidean distance between is the bias term; Model training and evaluation: Use the preprocessed data set for training, and use cross-validation and other techniques to optimize the parameter C and kernel function selection of the SVM model, evaluate the prediction performance of the model on the test data set, and ensure that it can accurately predict the temperature change trend; Real-time prediction and application: Integrating the trained SVM model into the real-time data processing flow enables it to analyze and predict the changing trends of temperature sensor data in real time, helping to respond to temperature changes in a timely manner and provide accurate data processing capabilities; Step 4 Implementation process of long short-term memory network LSTM: Model definition: LSTM is a recurrent neural network (RNN) suitable for sequence data. It is particularly good at capturing long-term dependencies in data. The LSTM unit includes a forget gate, an input gate, an output gate, and a hidden state. and cell status The update formula is as follows: Forget Gate: ； For the Gate of Forgetfulness, is the sigmoid function, is the weight of the forget gate, is the hidden state of the previous time step, is the current input, is the bias of the forget gate; Input Gate: ； ； is the input gate, is the sigmoid function, is the weight of the input gate, is the hidden state of the previous time step, is the current input, is the weight bias of the input gate, is the candidate cell state, is the hyperbolic tangent function, is the weight of the candidate cell state, is the bias of the candidate cell state; Cell status update: ； is the cell state, For the Gate of Forgetfulness, is element-wise multiplication, is the cell state at the previous time step, is the input gate, is the candidate cell state; Output Gate: ； ； is the output gate, is the sigmoid function, is the weight of the output gate, is the hidden state of the previous time step, is the current input, is the bias of the output gate, is the hidden state, is the hyperbolic tangent function, is the cell state; Training and Optimization: Training the LSTM model using historical data sequences ; Contains all from t=1 to T A collection of elements, is the element at the tth time step in the sequence; The network parameters are optimized by minimizing the loss function. The training process updates the weights and biases of the model through back-propagation BPTT, thereby improving the model's ability to predict future temperature change trends. Step 5 Temperature adjustment device: Design and implement an executive controller to regulate the operation and response of the temperature sensor based on the temperature trend predicted by the machine learning model; 1. Control logic design: Threshold-based control: Set a temperature threshold ,If the real-time temperature exceeds or approaches the threshold, corresponding control measures are taken; Control logic: : Reduce the heat dissipation rate or increase the contact area of ​​the heat transfer rod; is the temperature reading at the current moment; PID Controller: The PID controller adjusts the control output through proportion, integration and differentiation to make the system's feedback signal close to the target value; Controller formula: ; is the output of the controller, is the proportional gain parameter, is the error signal, is the integral gain parameter, From the start time to the current time The integral of the error signal, is the differential gain parameter, is the error signal rate of change over time; 2. Implement control logic to monitor real-time temperature data and machine learning prediction outputs; adjust sensor operating parameters and environmental interactions based on predicted temperature trends to achieve stable temperature control and optimize energy efficiency; 3. Introduce a feedback mechanism to adjust the controller parameters or strategies in real time to cope with changes and uncertainties in the actual environment; optimize the performance of the controller by continuously monitoring and analyzing sensor output data, and ensure the reliable operation of the sensor under various working conditions.

Images