Intelligent temperature detection device
Through the intelligent temperature detection device, the temperature signal is analyzed using CEEMDAN decomposition and deep learning algorithms, and the problems of low efficiency and insufficient automation of traditional temperature monitoring methods are solved, real-time monitoring and automated alarm of electrical equipment joint temperatures are realized, and the safety of the power grid is improved and power supply stability.
Patent Information
- Application Number
- CN202411424835.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-12
- Publication Date
- 2024-11-08
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Traditional temperature monitoring methods are inefficient and cannot achieve 24-hour uninterrupted monitoring, especially in closed or widely distributed equipment, and the level of automation and intelligence is insufficient, resulting in overheating of electrical connectors and may cause fires, affecting the safety and power supply of the power grid. stability.
An intelligent temperature detection device is designed, including a power extraction module, a temperature measuring sensor, a processor and a communication module. Through CEEMDAN decomposition and deep learning algorithm, analyzing the ambient temperature and temperature measuring signals, extracting signal characteristics, and realizing the characteristics of the self-supervised cluster center. Reverse reinforcement and cascading processing provide accurate temperature measurements and enable real-time monitoring and automated alarms through wireless transmission.
Real-time and remote monitoring of the temperature of electrical equipment connectors is realized, the operation efficiency and safety of the power grid is improved, more accurate temperature measurement values and higher intelligence and automation levels are provided, and the dependence of manual inspections is reduced.
Smart Images

Figure CN118913481A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of temperature detection, and in particular to an intelligent temperature detection device. Background Art
[0002] In modern power systems, the widespread distribution of distribution networks and the extensive use of electrical equipment pose severe challenges to the safe operation of power grids and the quality of power supply. The main circuits of electrical equipment are interconnected through electrical connectors, including incoming cables, outgoing cables, busbar connectors, etc. These electrical connectors will inevitably face aging problems in long-term operation, resulting in increased contact resistance, which in turn causes overheating of the connectors. Overheating of electrical equipment connectors will not only accelerate equipment aging, but may also cause fires, bringing serious safety hazards and economic losses to the power grid. Especially in the peak season of electricity consumption in summer, the heating problem of electrical connectors is particularly prominent. Once the connector overheats, if it is not discovered and handled in time, the continuous rise in temperature may eventually cause the connector to melt, causing power grid accidents and affecting the continuity and stability of power supply. Therefore, it is very important to monitor the temperature of electrical equipment connectors in real time and effectively.
[0003] However, traditional temperature monitoring methods, such as manual inspections and infrared temperature measurement technology, can detect temperature anomalies to a certain extent, but they have obvious shortcomings. Manual inspections are inefficient and cannot achieve 24-hour uninterrupted monitoring; although infrared temperature measurement technology can achieve non-contact measurement, it still requires manual operation in most scenarios, and is limited by the operating environment, making it difficult to apply to closed or widely distributed equipment. In addition, these traditional electrical equipment joint temperature monitoring methods still have a lot of room for improvement in automation and intelligence.
[0004] Therefore, an intelligent temperature detection device is desired. Summary of the invention
[0005] This application is made in view of the above problems. One object of this application is to provide an intelligent temperature detection device.
[0006] An embodiment of the present application provides an intelligent temperature detection device, which includes: a power supply module, a temperature sensor, a processor and a communication module; wherein the power supply module is used to power the temperature sensor, the processor and the communication module; the temperature sensor is used to collect temperature measurement electrical signals; the processor is used to encode and decode the temperature measurement electrical signals based on the time series of the ambient temperature value to obtain a temperature measurement value; and the communication module is used to transmit the temperature measurement value to the outside world.
[0007] For example, according to the intelligent temperature detection device of the embodiment of the present application, the power supply module is a PT power supply module or a CT power supply module.
[0008] For example, according to the intelligent temperature detection device of the embodiment of the present application, the processor includes: an ambient temperature data acquisition module, which is used to obtain the time series of ambient temperature values; a temperature measurement electrical signal decomposition module, which is used to perform CEEMDAN decomposition on the temperature measurement electrical signal to obtain a set of IMF signal components of the temperature measurement electrical signal; a temperature measurement electrical signal component waveform feature extraction module, which is used to perform signal waveform feature extraction on the set of IMF signal components of the temperature measurement electrical signal to obtain a set of waveform semantic feature vectors of the IMF signal components of the temperature measurement electrical signal; an ambient temperature time series encoding module, which is used to perform CEEMDAN decomposition on the temperature measurement electrical signal to obtain a set of IMF signal component waveform semantic feature vectors of the temperature measurement electrical signal; Temperature time series encoding is used to obtain an ambient temperature time series implicit correlation feature vector; a signal component waveform semantic feature selection and enhancement module is used to use the ambient temperature time series implicit correlation feature vector as the self-supervised cluster center representation vector, and to perform feature reverse enhancement based on the self-supervised cluster center on the set of the temperature measurement electrical signal IMF signal component waveform semantic feature vectors to obtain a set of reverse attention selection signal component waveform semantic feature vectors; a temperature measurement module is used to cascade the set of reverse attention selection signal component waveform semantic feature vectors to obtain a cascade feature vector, and then determine the temperature measurement value based on the cascade feature vector.
[0009] For example, according to the intelligent temperature detection device of an embodiment of the present application, the temperature measurement electrical signal component waveform feature extraction module is used to: input the set of the temperature measurement electrical signal IMF signal components into a signal waveform feature extractor based on a void convolutional neural network model to obtain a set of waveform semantic feature vectors of the temperature measurement electrical signal IMF signal components.
[0010] For example, according to the intelligent temperature detection device of an embodiment of the present application, the ambient temperature time series encoding module is used to: input the time series of the ambient temperature value into a sequence encoder based on a Bi-LSTM model to obtain an implicitly associated feature vector of the ambient temperature time series.
[0011] For example, according to the intelligent temperature detection device of the embodiment of the present application, the signal component waveform semantic feature selection and enhancement module includes: an anti-correlation unit, which is used to calculate the anti-correlation coefficient between each temperature measurement electrical signal IMF signal component waveform semantic feature vector in the set of the temperature measurement electrical signal IMF signal component waveform semantic feature vector and the self-supervised cluster center representation vector to obtain a sequence of temperature measurement electrical signal IMF signal component waveform semantic anti-correlation coefficients; an activation unit, which is used to input the sequence of temperature measurement electrical signal IMF signal component waveform semantic anti-correlation coefficients into a Sigmoid activation function to obtain a sequence of normalized temperature measurement electrical signal IMF signal component waveform semantic anti-correlation weight coefficients; a weighting unit, which is used to use the sequence of normalized temperature measurement electrical signal IMF signal component waveform semantic anti-correlation weight coefficients to obtain a sequence of normalized temperature measurement electrical signal IMF signal component waveform semantic anti-correlation weight coefficients. The normalized temperature measurement electrical signal IMF signal component waveform semantic anti-correlation weight coefficients in the column are used as weights, and the temperature measurement electrical signal IMF signal component waveform semantic feature vectors in the set of the temperature measurement electrical signal IMF signal component waveform semantic feature vectors are weighted respectively to obtain a set of temperature measurement electrical signal IMF signal component waveform semantic anti-reinforcement feature vectors; a differential unit is used to calculate the positional difference between the set of the temperature measurement electrical signal IMF signal component waveform semantic feature vectors and the set of the temperature measurement electrical signal IMF signal component waveform semantic anti-reinforcement feature vectors of each corresponding group to obtain the set of reverse attention selection signal component waveform semantic feature vectors.
[0012] For example, according to the intelligent temperature detection device of an embodiment of the present application, the anti-correlation unit is used to: calculate the difference vector between the waveform semantic feature vector of the temperature measurement electrical signal IMF signal component and the self-supervised cluster center representation vector to obtain the temperature measurement electrical signal IMF signal component waveform semantic-ambient temperature time series semantic difference vector; calculate the covariance matrix between the waveform semantic feature vector of the temperature measurement electrical signal IMF signal component and the self-supervised cluster center representation vector to obtain the temperature measurement electrical signal IMF signal component waveform semantic-ambient temperature time series semantic covariance matrix; calculate the temperature measurement electrical signal IMF signal component waveform semantic-ambient temperature time series semantic The matrix division between the transposed vector of the ambient temperature time series semantic difference vector and the covariance matrix of the waveform semantics of the temperature measurement electrical signal IMF signal component - the ambient temperature time series semantics is performed to obtain the temperature measurement electrical signal IMF signal component waveform semantics - ambient temperature time series semantics difference feature representation vector; after calculating the vector multiplication between the temperature measurement electrical signal IMF signal component waveform semantics - ambient temperature time series semantics difference feature representation vector and the temperature measurement electrical signal IMF signal component waveform semantics - ambient temperature time series semantics difference vector, the square root of the anti-correlation difference value is calculated to obtain the temperature measurement electrical signal IMF signal component waveform semantic anti-correlation coefficient.
[0013] For example, according to the intelligent temperature detection device of the embodiment of the present application, the temperature measurement module is used to: cascade the set of waveform semantic feature vectors of the reverse attention selection signal component to obtain the cascade feature vector, and then input the cascade feature vector into the temperature decoder to obtain the temperature measurement value.
[0014] According to the intelligent temperature detection device of the embodiment of the present application, it realizes automatic alarm and fault diagnosis, thereby improving the operation efficiency and safety of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings of the embodiments of the present application are briefly introduced below. Obviously, the drawings described below only relate to some embodiments of the present application, and are not intended to limit the present application.
[0016] Figure 1 A schematic diagram of the structure of an intelligent temperature detection device in an embodiment of the present application is shown.
[0017] Figure 2 A schematic diagram of the structure of the processor of the intelligent temperature detection device in an embodiment of the present application is shown.
[0018] Figure 3 A flow chart of an intelligent temperature detection method in an embodiment of the present application is shown.
[0019] Figure 4The application scenario diagram of the intelligent temperature detection device in the embodiment of the present application is shown. DETAILED DESCRIPTION
[0020] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present application.
[0021] The terms used in this specification are those common terms currently widely used in the art in consideration of the functions of the present application, but these terms may vary according to the intentions of those skilled in the art, precedents, or new technologies in the art. In addition, specific terms may be selected, and in this case, their detailed meanings will be described in the detailed description of the present application. Therefore, the terms used in the specification should not be understood as simple names, but rather as a general description based on the meaning of the terms and the present application.
[0022] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.
[0023] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. At the same time, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.
[0024] In addition, the application architecture diagram in the embodiment of the present application is intended to more clearly illustrate the technical solution in the embodiment of the present application, and does not constitute a limitation on the technical solution provided in the embodiment of the present application. Of course, for other application architectures and business applications, the technical solution provided in the embodiment of the present application is also applicable to similar problems.
[0025] The following is a non-restrictive description of the intelligent temperature detection device provided according to at least one embodiment of the present application through several examples or embodiments. As described below, different features in these specific examples or embodiments can be combined with each other without conflicting with each other to obtain new examples or embodiments, and these new examples or embodiments also fall within the scope of protection of the present application.
[0026] In order to solve the above problems, wireless temperature measurement technology is applied to overcome the limitations of traditional methods, which can realize real-time and remote monitoring of the temperature of electrical equipment joints. Through wireless communication, the temperature monitoring data can be effectively transmitted to the monitoring center to realize automatic alarm and fault diagnosis, thereby improving the operation efficiency and safety of the power grid.
[0027] Specifically, the present application fixes the temperature sensor to the screw at the electrical connector by means of a nut, and at the same time powers the temperature sensor by means of a self-powered method, that is, energy is obtained by using equipotential electric field power extraction technology, and then combined with wireless transmission technology, the temperature data is wirelessly sent to the receiving terminal on the low-voltage side, that is, the processor, to achieve real-time and uninterrupted monitoring of the temperature at the electrical switch connector.
[0028] More specifically, in the technical solution of the present application, an intelligent temperature detection device 800 is proposed, such as Figure 1 As shown, it includes: a power supply module 810, a temperature sensor 820, a processor 830 and a communication module 840; wherein, the power supply module 810 is used to power the temperature sensor 820, the processor 830 and the communication module 840; the temperature sensor 820 is used to collect temperature measurement electrical signals; the processor 830 is used to encode and decode the temperature measurement electrical signals based on the time series of the ambient temperature value to obtain the temperature measurement value; the communication module 840 is used to transmit the temperature measurement value to the outside world. The intelligent temperature detection device 800 can use the PT power supply module or the CT power supply module to self-power the temperature sensor, the processor and the communication module, and is suitable for temperature monitoring in scenarios such as disconnect switch equipment joints and pole-mounted circuit breaker joints, which broadens the temperature monitoring scenarios and improves universality.
[0029] Correspondingly, the power taking module 810 is a PT power taking module or a CT power taking module. Among them, regarding the PT power taking module, the potential transformer (PT for short) is based on the transformer principle to convert the high voltage of the front stage of the transformer into the low voltage of the rear stage, which is mainly used to power the measuring instrument and the relay protection device, to measure the voltage, power and electric energy of the line, or to protect the valuable equipment, motor and transformer in the line when the line fails. The capacity of the voltage transformer in the PT power taking method is very small, generally only a few volt-amperes or tens of volt-amperes, and is generally used in the distribution network, and cannot adapt to the high-voltage transmission line environment. Further, regarding the CT power taking module, the current transformer (CT for short) is based on the electromagnetic induction principle to convert the large AC current on the high-voltage transmission line into a small current of the transformer through electromagnetic coupling, so as to realize the transmission of the power taking, and achieve the purpose of taking power from the transmission line. The advantage of the CT power taking method is that it can provide a stable power output within the normal current range of the conductor, but it is greatly affected by the current change of the transmission line. The current on the line changes greatly, which will cause the output of the current transformer to be unstable.
[0030] In particular, considering that in the actual monitoring and measurement of the temperature of the electrical equipment joints, the fluctuation of the external ambient temperature is a factor that cannot be ignored. Such changes and fluctuations in ambient temperature will introduce irregular noise into the temperature measurement electrical signal, which may mask or blur the actual temperature changes, thereby affecting the accuracy and reliability of the measurement results.
[0031] In response to the above technical problems, the technical concept of the present application is to collect ambient temperature value data through real-time monitoring, and introduce data processing and analysis algorithms based on artificial intelligence and deep learning in the processor to perform time series analysis on the collected ambient temperature value data, and at the same time perform fine-grained signal decomposition analysis on the temperature measurement electrical signal, so as to capture the time series dynamic change characteristics of the ambient temperature and the waveform semantics of each signal component of the temperature measurement electrical signal. Then, the time series characteristics of the ambient temperature are used as the cluster center to select, enhance and aggregate the waveform semantics of each signal component. In this way, the actual device joint temperature component in the temperature measurement electrical signal can be more accurately and effectively identified and understood, thereby providing a more accurate temperature measurement value. In this way, the intelligence level and automation degree of the intelligent temperature detection device can be improved.
[0032] Accordingly, if Figure 2As shown, the processor 830 includes: an ambient temperature data acquisition module 831, which is used to obtain a time series of ambient temperature values; a temperature measurement electrical signal decomposition module 832, which is used to perform CEEMDAN (Complete Ensemble Empirical Mode Decomposition with Adaptive Noise) decomposition on the temperature measurement electrical signal to obtain an IMF (Intrinsic Mode Decomposition) temperature measurement electrical signal. Functions, intrinsic modal components) signal components; a temperature measurement electric signal component waveform feature extraction module 833, used to extract signal waveform features from the set of temperature measurement electric signal IMF signal components to obtain a set of temperature measurement electric signal IMF signal component waveform semantic feature vectors; an ambient temperature time series encoding module 834, used to perform temperature time series encoding on the time series of the ambient temperature value to obtain an ambient temperature time series implicit correlation feature vector; a signal component waveform semantic feature selection and enhancement module 835, used to use the ambient temperature time series implicit correlation feature vector as the self-supervised cluster center representation vector, and perform feature reverse enhancement based on the self-supervised cluster center on the set of temperature measurement electric signal IMF signal component waveform semantic feature vectors to obtain a set of reverse attention selection signal component waveform semantic feature vectors; a temperature measurement module 836, used to cascade the set of reverse attention selection signal component waveform semantic feature vectors to obtain a cascade feature vector, and then determine the temperature measurement value based on the cascade feature vector.
[0033] Specifically, in the processor, first, the time series of ambient temperature values is obtained, and the time series of ambient temperature values is input into a sequence encoder based on the Bi-LSTM model for encoding to extract the temporal correlation features of the ambient temperature values in the time dimension, thereby obtaining an ambient temperature temporal implicit correlation feature vector.
[0034] Correspondingly, the ambient temperature time series encoding module 834 is used to: input the time series of the ambient temperature values into a sequence encoder based on the Bi-LSTM model to obtain the ambient temperature time series implicit correlation feature vector.
[0035] It should be understood that there is a large amount of information in the temperature measurement electrical signal, which includes the semantics of changes in the temperature of the device joint and the noise semantics of the ambient temperature. Based on this, in order to be able to analyze the temperature measurement electrical signal in a more fine-grained manner, so as to highlight the key semantic information related to the temperature of the device joint, thereby providing a more accurate temperature measurement value, in the technical solution of the present application, the temperature measurement electrical signal is further decomposed by CEEMDAN to obtain a set of IMF signal components of the temperature measurement electrical signal. By performing CEEMDAN decomposition on the temperature measurement electrical signal, the complex temperature measurement electrical signal can be deconstructed into a series of simple intrinsic mode function (IMFs) components. This decomposition helps to identify different components in the signal, including temperature changes and noise, and provides a data basis for subsequent data processing and temperature measurement. Therefore, the IMF components of the temperature measurement electrical signal obtained by CEEMDAN decomposition can better represent the different frequency characteristics of the signal, which is particularly important for processing nonlinear and non-stationary temperature measurement electrical signals. In other words, these components can be more easily feature extracted and analyzed, which helps to identify features related to joint temperature changes in the signal.
[0036] Then, the set of IMF signal components of the temperature measurement electrical signal is input into a signal waveform feature extractor based on a dilated convolutional neural network model for feature mining, so as to respectively extract the waveform semantic feature information in each IMF signal component of the temperature measurement electrical signal, thereby obtaining a set of waveform semantic feature vectors of the IMF signal components of the temperature measurement electrical signal.
[0037] Correspondingly, the temperature measurement electrical signal component waveform feature extraction module 833 is used to: input the set of the temperature measurement electrical signal IMF signal components into the signal waveform feature extractor based on the void convolutional neural network model to obtain the set of the temperature measurement electrical signal IMF signal component waveform semantic feature vectors.
[0038] Furthermore, since the timing changes and volatility of the external temperature will affect the accuracy of the temperature measurement of the equipment joint, it will introduce irregular noise into the signal of the temperature measurement electrical signal. Therefore, it is necessary to filter out the noise information caused by these ambient temperatures during the analysis and aggregation of the waveform semantics of each IMF signal component of the temperature measurement electrical signal, and select and enhance the waveform semantics of the key signal components related to the temperature measurement task. Based on this, in the technical solution of the present application, the ambient temperature timing implicit association feature vector is further used as the self-supervised cluster center representation vector, and the set of the waveform semantic feature vectors of the IMF signal component of the temperature measurement electrical signal is subjected to feature reverse reinforcement based on the self-supervised cluster center to obtain a set of reverse attention selection signal component waveform semantic feature vectors. Through the processing operation of feature reverse reinforcement based on the self-supervised cluster center, the model convergence speed can be accelerated and the robustness of the model to the temperature measurement electrical signal noise and outliers caused by the timing changes of the ambient temperature can be improved. In this way, more attention can be paid to the semantics and useful information of the signal components in the temperature measurement electrical signal that are related to the subsequent device joint temperature measurement task, while ignoring insignificant ambient temperature noise, thereby achieving better performance in the temperature measurement task.
[0039] Specifically, the heterogeneity between them is quantified by calculating the anti-correlation coefficient between the waveform semantic feature vectors of each temperature measurement electrical signal IMF signal component and the self-supervised cluster center representation vector. That is, the degree of semantic irrelevance between the waveform semantics of each temperature measurement electrical signal IMF signal component and the self-supervised cluster center. Then, the anti-correlation coefficient is normalized by the Sigmoid activation function to convert this degree of irrelevance into a standardized weight coefficient, and then the normalized anti-correlation weight coefficient is used to weighted modulate the original feature distribution of the set of waveform semantic feature vectors of the temperature measurement electrical signal IMF signal component to generate a set of waveform semantic anti-enhanced feature vectors of the temperature measurement electrical signal IMF signal component. In this way, the features that are significantly different from the cluster center of the ambient temperature time series dynamic feature are enhanced, while the features that are similar to the cluster center are suppressed, thereby enhancing the distinguishability of the features and improving the richness of the feature expression. Finally, by calculating the positional difference between the waveform semantic feature vectors of each temperature measurement electrical signal IMF signal component and the waveform semantic anti-reinforcement feature vector of the temperature measurement electrical signal IMF signal component, a set of reverse attention selection signal component waveform semantic feature vectors is obtained. That is, after the original waveform semantic feature distribution of the temperature measurement electrical signal IMF signal component is reversely reinforced, the differential operation is used to reversely extract and reversely select the feature components that are positively correlated with the cluster center, that is, those features that are more important in temperature monitoring, thereby improving the model's generalization ability to new data or under different environmental conditions.
[0040] Correspondingly, the signal component waveform semantic feature selection and enhancement module 835 includes: an anti-correlation unit, which is used to calculate the anti-correlation coefficient between each temperature measurement electrical signal IMF signal component waveform semantic feature vector in the set of the temperature measurement electrical signal IMF signal component waveform semantic feature vector and the self-supervised cluster center representation vector to obtain a sequence of temperature measurement electrical signal IMF signal component waveform semantic anti-correlation coefficients; an activation unit, which is used to input the sequence of temperature measurement electrical signal IMF signal component waveform semantic anti-correlation coefficients into a Sigmoid activation function to obtain a sequence of normalized temperature measurement electrical signal IMF signal component waveform semantic anti-correlation weight coefficients; a weighting unit, which is used to use each normalized temperature measurement electrical signal in the sequence of normalized temperature measurement electrical signal IMF signal component waveform semantic anti-correlation weight coefficients to select and enhance the signal component waveform semantic feature vector. The temperature measurement electric signal IMF signal component waveform semantic anti-correlation weight coefficient is used as a weight, and each temperature measurement electric signal IMF signal component waveform semantic feature vector in the set of the temperature measurement electric signal IMF signal component waveform semantic feature vector is weighted respectively to obtain a set of temperature measurement electric signal IMF signal component waveform semantic anti-reinforcement feature vectors; a differential unit is used to calculate the positional difference between the set of the temperature measurement electric signal IMF signal component waveform semantic feature vectors and the set of the temperature measurement electric signal IMF signal component waveform semantic anti-reinforcement feature vectors of each corresponding group to obtain the set of the reverse attention selection signal component waveform semantic feature vectors.
[0041] Wherein, the anti-correlation unit is used to: calculate the difference vector between the waveform semantic feature vector of the IMF signal component of the temperature measurement electrical signal and the center representation vector of the self-supervised cluster to obtain the waveform semantics of the IMF signal component of the temperature measurement electrical signal - the time series semantics difference vector of the ambient temperature; calculate the covariance matrix between the waveform semantic feature vector of the IMF signal component of the temperature measurement electrical signal and the center representation vector of the self-supervised cluster to obtain the waveform semantics of the IMF signal component of the temperature measurement electrical signal - the time series semantics covariance matrix of the ambient temperature; calculate the waveform semantics of the IMF signal component of the temperature measurement electrical signal - the time series semantics difference vector of the ambient temperature The transposed vector of the temperature measurement electrical signal IMF signal component waveform semantics-ambient temperature time series semantics covariance matrix is used to obtain the temperature measurement electrical signal IMF signal component waveform semantics-ambient temperature time series semantics difference feature representation vector; after calculating the vector multiplication between the temperature measurement electrical signal IMF signal component waveform semantics-ambient temperature time series semantics difference feature representation vector and the temperature measurement electrical signal IMF signal component waveform semantics-ambient temperature time series semantics difference vector, the square root of the anti-correlation difference value is calculated to obtain the temperature measurement electrical signal IMF signal component waveform semantics anti-correlation coefficient.
[0042] In a specific example, the signal component waveform semantic feature selection and enhancement module 835 is used to: use the ambient temperature time series implicit correlation feature vector as the self-supervised cluster center representation vector, and use the following feature selection and enhancement formula to perform feature reverse enhancement based on the self-supervised cluster center on the set of the temperature measurement electrical signal IMF signal component waveform semantic feature vectors to obtain the set of reverse attention selection signal component waveform semantic feature vectors; wherein the feature selection and enhancement formula is: ;in, is the set of waveform semantic feature vectors of the IMF signal component of the temperature measurement electrical signal, is the first in the set of waveform semantic feature vectors of the IMF signal component of the temperature measurement electrical signal The waveform semantic feature vector of the IMF signal component of the temperature measurement electrical signal, is the first in the set of waveform semantic feature vectors of the IMF signal component of the temperature measurement electrical signal The waveform semantic feature vector of the IMF signal component of the temperature measurement electrical signal, is the self-supervised cluster center representation vector, For the said The covariance matrix between the waveform semantic feature vector of the IMF signal component of the temperature measurement electrical signal and the center representation vector of the self-supervised cluster, For the said The anti-correlation coefficient corresponding to the semantic feature vector of the waveform of the IMF signal component of the temperature measurement electrical signal, for function, is the first in the sequence of semantic anti-correlation weight coefficients of the waveform of the normalized temperature measurement electrical signal IMF signal component Normalized temperature measurement electrical signal IMF signal component waveform semantic anti-correlation weight coefficient, is the first in the sequence of semantic anti-correlation weight coefficients of the normalized temperature measurement electrical signal IMF signal component waveform Normalized temperature measurement electrical signal IMF signal component waveform semantic anti-correlation weight coefficient, is the set of semantic anti-enhancement feature vectors of the waveform of the IMF signal component of the temperature measurement electrical signal, For positional subtraction, A set of waveform semantic feature vectors of signal components are selected for the reverse attention.
[0043] Then, after the set of the reverse attention selection signal component waveform semantic feature vectors is cascaded to obtain the cascade feature vector, the cascade feature vector is input into the temperature decoder to obtain the temperature measurement value. In other words, the aggregation features of the waveform semantic features of the key signal components after reverse attention selection are used for decoding regression to generate the temperature measurement value. In this way, a more accurate temperature measurement value can be provided, and the intelligence level and automation degree of the intelligent temperature detection device can be improved.
[0044] Correspondingly, the temperature measurement module 836 is used to: cascade the set of the reverse attention selection signal component waveform semantic feature vectors to obtain the cascade feature vector, and then input the cascade feature vector into the temperature decoder to obtain the temperature measurement value.
[0045] Preferably, the present application takes into account that the ambient temperature time series implicit correlation feature vector represents the local time series semantic feature of the ambient temperature, and the set of the temperature measurement electrical signal IMF signal component waveform semantic feature vectors respectively represents the waveform semantic features of each temperature measurement electrical signal IMF signal component obtained by CEEMDAN decomposition of the temperature measurement electrical signal. In this way, when the ambient temperature time series implicit correlation feature vector is used as the self-supervised cluster center representation vector, and the set of the temperature measurement electrical signal IMF signal component waveform semantic feature vectors is subjected to feature reverse attention selection based on the self-supervised cluster center to obtain a set of reverse attention selected signal component waveform semantic feature vectors, the ambient temperature time series implicit correlation feature vector is used as the feature vector screening anchor center and the anti-correlation coefficient between the ambient temperature time series implicit correlation feature vector and each temperature measurement electrical signal IMF signal component waveform semantic feature vector in the set of the temperature measurement electrical signal IMF signal component waveform semantic feature vector is combined to realize feature reverse selection. However, considering that the feature modal differences will result in significant structural heterogeneity in the feature distribution between the individual reverse attention selection signal component waveform semantic feature vectors in the set of the reverse attention selection signal component waveform semantic feature vectors, the cascaded feature vector obtained by cascading the set of the reverse attention selection signal component waveform semantic feature vectors will also cause outlier decoding regression inference mapping deviation due to the difference in constraint coverage anti-correlation coefficients, thereby affecting the accuracy of the temperature measurement value obtained by the cascaded feature vector through the temperature decoder.
[0046] Therefore, in a preferred example, after the set of semantic feature vectors of the waveform of the reverse attention selection signal component is cascaded to obtain a cascade feature vector, the cascade feature vector is input into a temperature decoder to obtain the temperature measurement value, including: calculating the sum of the absolute values of the eigenvalues of the cascade feature vector to obtain a first temperature measurement cascade and modulation value, and calculating the square root of the sum of the squares of the eigenvalues of the cascade feature vector to obtain a second temperature measurement cascade and modulation value; performing point subtraction on the cascade feature vector and the second temperature measurement cascade and modulation value, and respectively adding the eigenvalue number of the cascade feature vector and the inverse of the first temperature measurement cascade and modulation value to obtain the temperature measurement value. The method comprises the steps of: performing a dot multiplication on the cascade eigenvalues and taking the inverse of each eigenvalue to obtain a first temperature measurement cascade phase transition vector; performing a dot subtraction on the cascade eigenvector and the first temperature measurement cascade and modulation value, respectively performing a dot multiplication on the cascade eigenvector and the inverse of the second temperature measurement cascade and modulation value, and taking the inverse of each eigenvalue to obtain a second temperature measurement cascade phase transition vector; subtracting the dot product vector of the second temperature measurement cascade phase transition vector and the weighted hyperparameter from the first temperature measurement cascade phase transition vector to obtain an optimized cascade eigenvector; and passing the optimized cascade eigenvector through a temperature decoder to obtain a temperature measurement value.
[0047] Here, the concatenated feature vector The optimization expression is: ;in, is the concatenated feature vector, represents a set of real numbers, representing the concatenated eigenvector The eigenvalues at the positions, represents the number of eigenvalues of the concatenated eigenvector, represents the first temperature measurement cascade and the modulation value, represents the second temperature measurement cascade and the modulation value, It means point multiplication by position. It means subtracting by position. It means to calculate the inverse of each eigenvalue of the eigenvector. represents the first temperature measurement cascade phase transition vector, represents the second temperature measurement cascade phase transition vector, represents the weighted hyperparameter, Represents the optimized concatenated feature vector.
[0048] In the preferred example, the difference of the feature value of the cascade feature vector relative to the feature set of the whole vector of the cascade feature vector and the modulation representation is used as the semantic change intensity information, and the phase conversion corresponding to the position-based intensity modulation is performed through the different and modulation representation forms, so that the aggregation enhancement of the semantic change phase perception can improve the axial aggregation receptive field along the feature aggregation direction by performing the space translation operation based on alternating stacking under the scale balance of the vector set of the cascade feature vector, thereby improving the perception effect of the aggregated semantics of the cascade feature vector on the detail semantic change, so as to improve the accuracy of the temperature measurement value obtained by the optimized cascade feature vector through the temperature decoder by avoiding the counterfactual reasoning mapping of the outlier feature distribution of the cascade feature vector to the class probability. In this way, the actual device joint temperature component in the temperature measurement electrical signal can be more accurately and effectively identified and understood, thereby providing a more accurate temperature measurement value, and improving the intelligence level and automation level of the intelligent temperature detection device.
[0049] Based on the above embodiments, see Figure 3 As shown in FIG. 1 , it is a flow chart of an intelligent temperature detection method in an embodiment of the present application. Figure 3 As shown, the intelligent temperature detection method according to the embodiment of the present application includes the following steps: S510, obtaining the time series of ambient temperature values; S520, performing CEEMDAN decomposition on the temperature measuring electrical signal collected by the temperature measuring sensor to obtain a set of IMF signal components of the temperature measuring electrical signal; S530, performing signal waveform feature extraction on the set of IMF signal components of the temperature measuring electrical signal to obtain a set of waveform semantic feature vectors of the IMF signal components of the temperature measuring electrical signal; S540, performing temperature time series encoding on the time series of the ambient temperature values to obtain an ambient temperature time series implicit correlation feature vector; S550, using the ambient temperature time series implicit correlation feature vector as the self-supervised cluster center representation vector, performing feature reverse enhancement based on the self-supervised cluster center on the set of waveform semantic feature vectors of the IMF signal components of the temperature measuring electrical signal to obtain a set of waveform semantic feature vectors of reverse attention selection signal components; S560, after cascading the set of waveform semantic feature vectors of the reverse attention selection signal components to obtain a cascade feature vector, the temperature measurement value is determined based on the cascade feature vector.
[0050] Here, those skilled in the art can understand that the specific operations of each step in the above-mentioned intelligent temperature detection method have been referred to above. Figure 1 to Figure 2 The intelligent temperature detection device 800 is described in detail, and therefore, its repeated description will be omitted.
[0051] Figure 4 FIG. 1 is an application scenario diagram of the intelligent temperature detection device according to an embodiment of the present application. Figure 4As shown, in this application scenario, first, the time series of ambient temperature values and the temperature measurement electrical signals are obtained, and then the time series of ambient temperature values and the temperature measurement electrical signals are input into a server deployed with an intelligent temperature detection algorithm, wherein the server can use the intelligent temperature detection algorithm to process the time series of ambient temperature values and the temperature measurement electrical signals to obtain temperature measurement values.
[0052] Based on the above embodiments, another exemplary embodiment of an electronic device is also provided in the embodiments of the present application. In some possible implementations, the electronic device in the embodiments of the present application may include a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor can implement the steps of the intelligent temperature detection method in the above embodiments when executing the program.
[0053] The embodiment of the present application also provides a computer-readable storage medium, on which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, the intelligent temperature detection method according to the embodiment of the present application described with reference to the above figures can be executed. The computer-readable storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. The volatile memory may, for example, include a random access memory (RAM) and / or a cache memory (cache), etc. The non-volatile memory may, for example, include a read-only memory (ROM), a hard disk, a flash memory, etc.
[0054] The embodiment of the present application also provides a computer program product or a computer program, which includes computer executable instructions, and the computer executable instructions are stored in a computer readable storage medium. The processor of the computer device reads the computer executable instructions from the computer readable storage medium, and the processor executes the computer executable instructions, so that the computer device executes the intelligent temperature detection method according to the embodiment of the present application.
[0055] Those skilled in the art will appreciate that the contents disclosed in this application may be subject to various variations and improvements. For example, the various devices or components described above may be implemented by hardware, or by software, firmware, or a combination of some or all of the three.
[0056] In addition, although the present application makes various references to certain units in the system according to embodiments of the present application, any number of different units can be used and run on the client and / or server. The units are only illustrative, and different aspects of the system and method can use different units.
[0057] Those skilled in the art will appreciate that all or part of the steps in the above method can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, such as a read-only memory, a disk or an optical disk. Optionally, all or part of the steps in the above embodiment can also be implemented using one or more integrated circuits. Accordingly, each module / unit in the above embodiment can be implemented in the form of hardware or in the form of software function modules. The present application is not limited to any particular form of combination of hardware and software.
[0058] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art to which this application belongs. It should also be understood that terms such as those defined in common dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an idealized or extremely formal sense, unless explicitly defined as such herein.
[0059] The above is an explanation of the present application and should not be considered as limiting thereof. Although several exemplary embodiments of the present application have been described, those skilled in the art will readily appreciate that many modifications may be made to the exemplary embodiments without departing from the novel teachings and advantages of the present application.
Claims
1. An intelligent temperature detection device, characterized in that: include: A power supply module, a temperature measuring sensor, a processor and a communication module; wherein the power supply module is used to supply power to the temperature measuring sensor, the processor and the communication module; the temperature measuring sensor is used to collect temperature measuring electrical signals; the processor is used to encode and decode the temperature measuring electrical signals based on the time series of the ambient temperature values to obtain the temperature measurement values; the communication module is used to transmit the temperature measurement values to the outside world; wherein the processor comprises: an ambient temperature data acquisition module, used to obtain the time series of the ambient temperature values; a temperature measuring electrical signal decomposition module, used to perform CEEMDAN decomposition on the temperature measuring electrical signals to obtain a set of IMF signal components of the temperature measuring electrical signals; a temperature measuring electrical signal component waveform feature extraction module, used to perform signal waveform feature extraction on the set of IMF signal components of the temperature measuring electrical signals; The invention discloses a signal component waveform semantic feature extraction module, which is used to obtain a set of waveform semantic feature vectors of the IMF signal component of the temperature measurement electrical signal; an ambient temperature time series encoding module, which is used to perform temperature time series encoding on the time series of the ambient temperature value to obtain an ambient temperature time series implicit correlation feature vector; a signal component waveform semantic feature selection and enhancement module, which is used to use the ambient temperature time series implicit correlation feature vector as the self-supervised cluster center representation vector, and perform feature reverse enhancement based on the self-supervised cluster center on the set of waveform semantic feature vectors of the IMF signal component of the temperature measurement electrical signal to obtain a set of waveform semantic feature vectors of reverse attention selection signal component; a temperature measurement module, which is used to cascade the set of waveform semantic feature vectors of reverse attention selection signal component to obtain a cascade feature vector, and then determine the temperature measurement value based on the cascade feature vector.
2. The intelligent temperature detection device according to claim 1, characterized in that: The power taking module is a PT power taking module or a CT power taking module.
3. The intelligent temperature detection device according to claim 2, characterized in that: The temperature measurement electrical signal component waveform feature extraction module is used to: input the set of the temperature measurement electrical signal IMF signal components into a signal waveform feature extractor based on a void convolutional neural network model to obtain a set of waveform semantic feature vectors of the temperature measurement electrical signal IMF signal components.
4. The intelligent temperature detection device according to claim 3, characterized in that: The ambient temperature time series encoding module is used to: input the time series of the ambient temperature values into a sequence encoder based on a Bi-LSTM model to obtain an implicitly associated feature vector of the ambient temperature time series.
5. The intelligent temperature detection device according to claim 4, characterized in that: The signal component waveform semantic feature selection and enhancement module includes: an anti-correlation unit, which is used to calculate the anti-correlation coefficient between each temperature measurement electrical signal IMF signal component waveform semantic feature vector in the set of the temperature measurement electrical signal IMF signal component waveform semantic feature vector and the self-supervised cluster center representation vector to obtain a sequence of temperature measurement electrical signal IMF signal component waveform semantic anti-correlation coefficients; an activation unit, which is used to input the sequence of temperature measurement electrical signal IMF signal component waveform semantic anti-correlation coefficients into a Sigmoid activation function to obtain a sequence of normalized temperature measurement electrical signal IMF signal component waveform semantic anti-correlation weight coefficients; a weighting unit, which is used to use each normalized temperature measurement electrical signal in the sequence of normalized temperature measurement electrical signal IMF signal component waveform semantic anti-correlation weight coefficients to select and enhance the signal component waveform semantic feature, and the self-supervised cluster center representation vector to obtain a sequence of temperature measurement electrical signal IMF signal component waveform semantic anti-correlation weight coefficients. The IMF signal component waveform semantic anti-correlation weight coefficient is used as the weight, and each temperature measurement electrical signal IMF signal component waveform semantic feature vector in the set of the temperature measurement electrical signal IMF signal component waveform semantic feature vector is weighted respectively to obtain a set of temperature measurement electrical signal IMF signal component waveform semantic anti-reinforcement feature vectors; a differential unit is used to calculate the positional difference between the set of the temperature measurement electrical signal IMF signal component waveform semantic feature vectors and the set of the temperature measurement electrical signal IMF signal component waveform semantic anti-reinforcement feature vectors of each corresponding group to obtain the set of the reverse attention selection signal component waveform semantic feature vectors.
6. The intelligent temperature detection device according to claim 5, characterized in that: The anti-correlation unit is used to: calculate the difference vector between the waveform semantic feature vector of the IMF signal component of the temperature measurement electrical signal and the center representation vector of the self-supervised cluster to obtain the waveform semantics of the IMF signal component of the temperature measurement electrical signal-ambient temperature time series semantics difference vector; calculate the covariance matrix between the waveform semantic feature vector of the IMF signal component of the temperature measurement electrical signal and the center representation vector of the self-supervised cluster to obtain the waveform semantics of the IMF signal component of the temperature measurement electrical signal-ambient temperature time series semantics covariance matrix; calculate the transformation matrix of the waveform semantics of the IMF signal component of the temperature measurement electrical signal-ambient temperature time series semantics difference vector The method comprises the following steps: performing matrix division between the temperature measurement electrical signal IMF signal component waveform semantics-ambient temperature time series semantics covariance matrix and the temperature measurement electrical signal IMF signal component waveform semantics-ambient temperature time series semantics difference feature representation vector; calculating the temperature measurement electrical signal IMF signal component waveform semantics-ambient temperature time series semantics difference feature representation vector and the temperature measurement electrical signal IMF signal component waveform semantics-ambient temperature time series semantics difference vector, and then calculating the square root of the anti-correlation difference value to obtain the temperature measurement electrical signal IMF signal component waveform semantics anti-correlation coefficient.
7. The intelligent temperature detection device according to claim 6, characterized in that: The temperature measurement module is used to: cascade the set of waveform semantic feature vectors of the reverse attention selection signal component to obtain the cascade feature vector, and then input the cascade feature vector into a temperature decoder to obtain the temperature measurement value.
