CN116647946A - Semiconductor-based heating control system and method thereof - Google Patents
Semiconductor-based heating control system and method thereof Download PDFInfo
- Publication number
- CN116647946A CN116647946A CN202310929535.6A CN202310929535A CN116647946A CN 116647946 A CN116647946 A CN 116647946A CN 202310929535 A CN202310929535 A CN 202310929535A CN 116647946 A CN116647946 A CN 116647946A
- Authority
- CN
- China
- Prior art keywords
- feature
- voltage waveform
- voltage
- local
- characteristic
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000010438 heat treatment Methods 0.000 title claims abstract description 54
- 239000004065 semiconductor Substances 0.000 title claims abstract description 53
- 238000000034 method Methods 0.000 title claims abstract description 39
- 238000001514 detection method Methods 0.000 claims abstract description 82
- 238000004458 analytical method Methods 0.000 claims abstract description 24
- 239000013598 vector Substances 0.000 claims description 68
- 238000010586 diagram Methods 0.000 claims description 23
- 238000005070 sampling Methods 0.000 claims description 20
- 230000000694 effects Effects 0.000 claims description 16
- 239000011159 matrix material Substances 0.000 claims description 16
- 238000000605 extraction Methods 0.000 claims description 14
- 238000013527 convolutional neural network Methods 0.000 claims description 13
- 238000005457 optimization Methods 0.000 claims description 12
- 238000006243 chemical reaction Methods 0.000 claims description 6
- 230000003068 static effect Effects 0.000 claims description 5
- 238000011176 pooling Methods 0.000 claims description 4
- 230000001419 dependent effect Effects 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 230000001105 regulatory effect Effects 0.000 abstract description 2
- 238000012545 processing Methods 0.000 description 7
- 230000033228 biological regulation Effects 0.000 description 4
- 230000008859 change Effects 0.000 description 4
- 238000013136 deep learning model Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 239000000463 material Substances 0.000 description 4
- 230000007246 mechanism Effects 0.000 description 4
- 238000005259 measurement Methods 0.000 description 3
- 238000003909 pattern recognition Methods 0.000 description 3
- 230000000737 periodic effect Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000002123 temporal effect Effects 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 2
- 230000008485 antagonism Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000006399 behavior Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 239000003292 glue Substances 0.000 description 2
- 230000001737 promoting effect Effects 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 238000013519 translation Methods 0.000 description 2
- 239000000853 adhesive Substances 0.000 description 1
- 230000001070 adhesive effect Effects 0.000 description 1
- 230000003042 antagnostic effect Effects 0.000 description 1
- 238000007664 blowing Methods 0.000 description 1
- 238000009529 body temperature measurement Methods 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 230000017525 heat dissipation Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 229910044991 metal oxide Inorganic materials 0.000 description 1
- 150000004706 metal oxides Chemical class 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 238000000547 structure data Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H05—ELECTRIC TECHNIQUES NOT OTHERWISE PROVIDED FOR
- H05B—ELECTRIC HEATING; ELECTRIC LIGHT SOURCES NOT OTHERWISE PROVIDED FOR; CIRCUIT ARRANGEMENTS FOR ELECTRIC LIGHT SOURCES, IN GENERAL
- H05B1/00—Details of electric heating devices
- H05B1/02—Automatic switching arrangements specially adapted to apparatus ; Control of heating devices
Landscapes
- Control Of Resistance Heating (AREA)
Abstract
A semiconductor-based heating control system and method thereof are disclosed. The method comprises the steps of firstly acquiring a voltage detection signal acquired by a semiconductor heating device, then carrying out circuit analysis on the voltage detection signal to obtain an output voltage, and then generating a regulating control signal based on a PID control algorithm and the output voltage so that the actual temperature is consistent with the set temperature. In this way, it is possible to perform feature encoding-feature decoding on the voltage detection signal based on the analysis circuit of the encoder-decoder structure to improve the detection accuracy of the output voltage value, thereby improving the detection accuracy of the real-time temperature detection value, so that the adjustment control signal can be generated based on the PID algorithm to keep the actual temperature consistent with the set temperature.
Description
Technical Field
The present disclosure relates to the field of heating control, and more particularly, to a semiconductor-based heating control system and method thereof.
Background
The semiconductor heating device is a heating device for generating joule heat by utilizing current through semiconductor materials, has the advantages of high response speed, small volume, long service life and the like, and is widely applied to the fields of industry, medical treatment, scientific research and the like.
However, since the temperature of the semiconductor heating device is affected by various factors such as current variation, ambient temperature, heat dissipation conditions, etc., it is difficult to achieve accurate temperature control. Thus, an optimized semiconductor-based heating control scheme is desired.
Disclosure of Invention
In view of this, the present disclosure proposes a semiconductor-based heating control system and method thereof, which can perform feature encoding-feature decoding on the voltage detection signal based on an analysis circuit of an encoder-decoder structure to improve the detection accuracy of the output voltage value, thereby improving the detection accuracy of the real-time temperature detection value, so that an adjustment control signal can be generated based on a PID algorithm to keep the actual temperature consistent with a set temperature.
According to an aspect of the present disclosure, there is provided a semiconductor-based heating control method including: acquiring a voltage detection signal acquired by a semiconductor heating device; performing circuit analysis on the voltage detection signal to obtain an output voltage; and generating a regulation control signal based on a PID control algorithm and the output voltage so that the actual temperature is kept consistent with the set temperature.
According to another aspect of the present disclosure, there is provided a semiconductor-based heating control system, including: the signal acquisition module is used for acquiring a voltage detection signal acquired by the semiconductor heating device; the circuit analysis module is used for carrying out circuit analysis on the voltage detection signal to obtain output voltage; and a temperature control module for generating a regulation control signal based on a PID control algorithm and the output voltage so that the actual temperature is consistent with the set temperature.
According to an embodiment of the present disclosure, a voltage detection signal acquired by a semiconductor heating device is first acquired, then the voltage detection signal is subjected to circuit analysis to obtain an output voltage, and then an adjustment control signal is generated based on a PID control algorithm and the output voltage so that an actual temperature is kept consistent with a set temperature. In this way, it is possible to perform feature encoding-feature decoding on the voltage detection signal based on the analysis circuit of the encoder-decoder structure to improve the detection accuracy of the output voltage value, thereby improving the detection accuracy of the real-time temperature detection value, so that the adjustment control signal can be generated based on the PID algorithm to keep the actual temperature consistent with the set temperature.
Other features and aspects of the present disclosure will become apparent from the following detailed description of exemplary embodiments, which proceeds with reference to the accompanying drawings.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate exemplary embodiments, features and aspects of the present disclosure and together with the description, serve to explain the principles of the disclosure.
Fig. 1 illustrates a flowchart of a semiconductor-based heating control method according to an embodiment of the present disclosure.
Fig. 2 shows an architectural diagram of a semiconductor-based heating control method according to an embodiment of the present disclosure.
Fig. 3 shows a flowchart of substep S120 of a semiconductor-based heating control method according to an embodiment of the present disclosure.
Fig. 4 shows a flowchart of substep S121 of a semiconductor-based heating control method according to an embodiment of the present disclosure.
Fig. 5 shows a block diagram of a semiconductor-based heating control system according to an embodiment of the present disclosure.
Fig. 6 illustrates an application scenario diagram of a semiconductor-based heating control method according to an embodiment of the present disclosure.
Detailed Description
The following description of the embodiments of the present disclosure will be made clearly and fully with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some, but not all embodiments of the disclosure. All other embodiments, which can be made by one of ordinary skill in the art without undue burden based on the embodiments of the present disclosure, are also within the scope of the present disclosure.
As used in this disclosure and in the claims, the terms "a," "an," "the," and/or "the" are not specific to a singular, but may include a plurality, unless the context clearly dictates otherwise. In general, the terms "comprises" and "comprising" merely indicate that the steps and elements are explicitly identified, and they do not constitute an exclusive list, as other steps or elements may be included in a method or apparatus.
Various exemplary embodiments, features and aspects of the disclosure will be described in detail below with reference to the drawings. In the drawings, like reference numbers indicate identical or functionally similar elements. Although various aspects of the embodiments are illustrated in the accompanying drawings, the drawings are not necessarily drawn to scale unless specifically indicated.
In addition, numerous specific details are set forth in the following detailed description in order to provide a better understanding of the present disclosure. It will be understood by those skilled in the art that the present disclosure may be practiced without some of these specific details. In some instances, methods, means, elements, and circuits well known to those skilled in the art have not been described in detail in order not to obscure the present disclosure.
In view of the above technical problems, the technical concept of the present disclosure is to collect real-time temperature of a heated control device with a thermistor as a temperature sensor, and then generate an output voltage based on a PID control algorithm to generate an adjustment control signal so that actual temperature of the heated control device is consistent with a set temperature. However, when the thermistor is used as a temperature sensor to collect the real-time temperature of the heated control device, the thermistor belongs to a sensitive element, and when the temperature is collected and implemented, the thermistor is attached to the surface of the heated control device, and the collected voltage detection signal is analyzed to obtain a real-time temperature value, but the accuracy of real-time temperature detection is difficult to ensure because the voltage detection signal is weak (meanwhile, other noise may be introduced), and further the control effect of the PID algorithm is difficult to ensure.
It should be understood that a Thermistor (thermal) is a temperature-sensitive resistive element whose resistance value varies with a change in temperature, and is generally made of a metal oxide or semiconductor material, having both a Negative Temperature Coefficient (NTC) and a Positive Temperature Coefficient (PTC) type. The working principle of the thermistor is based on the temperature sensitive property of the material, and when the thermistor is contacted with the heated control device, the thermistor absorbs heat of the heated control device, so that the resistance value is changed. Wherein the resistance value of the NTC thermistor decreases with increasing temperature, and the resistance value of the PTC thermistor increases with increasing temperature. To collect the real-time temperature of the heated control device, a thermistor is typically attached to the surface of the heated control device. By passing a current through the thermistor, measuring the voltage across the resistor, and then calculating the real-time temperature of the heated control device based on the temperature-resistance characteristic of the thermistor. However, since the voltage detection signal of the thermistor may be very weak and may be disturbed by other noise, there may be a problem of accuracy in real-time temperature detection, and in order to improve accuracy, some measures such as amplifying the voltage signal using an amplifier, removing noise by a filter, and the like may be taken. In other words, a thermistor is a temperature sensitive resistive element, and by measuring its voltage signal, the real-time temperature of the heated control device can be obtained. However, in order to ensure accuracy, additional measures may be taken to handle weak signals and noise.
Wherein, when attaching the thermistor to the surface of the heated control device, consideration is required to: 1. the position of the thermistor is selected to be a proper position for attaching the thermistor on the surface of the heated control device, and the position can accurately reflect the temperature change of the heated control device and is not influenced by other factors (such as heat conduction, fan blowing and the like); 2. the fixing of the thermistor ensures that the thermistor is firmly attached to the surface of the heated control device so as to prevent the thermistor from moving or falling off in the working process, and the thermistor can be fixed by using modes such as heat conducting glue, heat conducting adhesive sheets or special fixing clamps for the thermistor; 3. the thermistor is in thermal contact with the heated control device, so that the thermistor is in good thermal contact with the heated control device, the temperature change of the heated control device can be accurately sensed, and materials such as heat conducting glue or a heat conducting gasket can be used for improving the thermal contact effect; 4. isolating other heat sources, ensuring that the position to which the thermistor is attached is not interfered by the other heat sources, so as to avoid inaccuracy of temperature measurement, for example, avoiding attaching the thermistor at a position close to a heating element, so as to avoid that the heat of a heated control device directly influences the measurement result of the thermistor; 5. the temperature conduction effect, considering the temperature conduction effect between the thermistor and the heated control device, may cause a certain delay in the measurement result of the thermistor, and in the control system, compensation or adjustment is required according to the actual situation to ensure the accuracy of temperature control. In other words, when attaching the thermistor, it is necessary to select a proper position, firmly fix, ensure good thermal contact, isolate other heat sources, and take into account the temperature conduction effects to ensure accurate measurement of the temperature of the heated control device.
Based on this, the technical idea of the present disclosure is to collect a voltage detection signal using a thermistor as a temperature sensor, and generate an output voltage value through an analysis circuit based on an encoder-decoder structure, and obtain a temperature signal based on the output voltage, and further generate a regulation control signal based on a PID algorithm so that an actual temperature is kept consistent with a set temperature. That is, in the technical scheme of the present disclosure, the voltage detection signal is feature-encoded-feature-decoded based on an analysis circuit of an encoder-decoder structure to improve the detection accuracy of the output voltage value, thereby improving the detection accuracy of the real-time temperature detection value, so that the adjustment control signal can be generated based on a PID algorithm to keep the actual temperature consistent with the set temperature.
Fig. 1 illustrates a flowchart of a semiconductor-based heating control method according to an embodiment of the present disclosure. Fig. 2 shows an architectural diagram of a semiconductor-based heating control method according to an embodiment of the present disclosure. As shown in fig. 1 and 2, a semiconductor-based heating control method according to an embodiment of the present disclosure includes the steps of: s110, acquiring a voltage detection signal acquired by a semiconductor heating device; s120, carrying out circuit analysis on the voltage detection signal to obtain output voltage; and S130, generating a regulating control signal based on a PID control algorithm and the output voltage so that the actual temperature is consistent with the set temperature.
Specifically, in the technical scheme of the present disclosure, first, a voltage detection signal acquired by a semiconductor heating device is acquired. And then, sampling the voltage detection signal and extracting local characteristics to obtain a plurality of voltage waveform local characteristic diagrams. That is, implicit characteristic distribution information about the voltage waveform contained in the voltage detection signal is captured.
In a specific example of the present disclosure, the encoding process for sampling and extracting local features of the voltage detection signal to obtain a plurality of voltage waveform local feature maps includes: firstly, carrying out sliding sampling operation on the voltage detection signals to obtain a plurality of partial images; the plurality of partial images are then passed through a feature extractor based on a convolutional neural network model, respectively, to obtain a plurality of voltage waveform partial feature maps.
Taking into account that there is a temporal correlation between the plurality of partial images obtained after sliding window sampling. Specifically, the sliding window sampling operation is performed on a continuous voltage signal, and each time the sliding window is moved, the sliding window is advanced by a certain step length on the time axis. This temporal correlation means that there is some overlap or continuity between adjacent partial images. That is, a portion of the sampled data may be shared between the partial images, which may ensure that transients or periodic features in the voltage signal are not missed. By this correlation, the trend of the change in the voltage signal and the continuity of the periodic characteristic can be observed. For example, if there is a periodic signal in the circuit, then there should be a similar waveform pattern in adjacent partial images. If there is a transient response in the circuit, abrupt or rapidly changing features may appear in adjacent partial images.
In order to mine and capture the time sequence correlation characteristic, in the technical scheme of the disclosure, the correlation characteristic among the plurality of voltage waveform local characteristic diagrams is extracted to obtain the voltage waveform global semantic characteristic vector. In a specific example of the disclosure, the implementation process of extracting the correlation features between the plurality of voltage waveform local feature graphs to obtain the voltage waveform global semantic feature vector is as follows: firstly, respectively expanding the plurality of voltage waveform local feature images into feature vectors to obtain a plurality of voltage waveform local expansion feature vectors; and then the plurality of voltage waveform local expansion feature vectors pass through a context correlation feature extractor based on the converter module to obtain a voltage waveform global semantic feature vector.
Accordingly, in step S120, as shown in fig. 3, the circuit analysis is performed on the voltage detection signal to obtain an output voltage, including: s121, extracting waveform characteristics of the voltage detection signals to obtain a voltage waveform global semantic feature vector; and S122, determining the output voltage based on the voltage waveform global semantic feature vector. In step S121, as shown in fig. 4, waveform feature extraction is performed on the voltage detection signal to obtain a voltage waveform global semantic feature vector, which includes: s1211, sampling and extracting local characteristics of the voltage detection signals to obtain a plurality of voltage waveform local characteristic diagrams; and S1212, extracting association features among the plurality of voltage waveform local feature graphs to obtain the voltage waveform global semantic feature vector.
Accordingly, in a specific example, sampling and local feature extraction are performed on the voltage detection signal in step S1211 to obtain a plurality of voltage waveform local feature graphs, including: performing sliding sampling operation on the voltage detection signals to obtain a plurality of partial images; and respectively passing the plurality of partial images through a characteristic extractor based on a convolutional neural network model to obtain a plurality of voltage waveform partial characteristic diagrams. It will be appreciated that a sliding sampling operation is a method of sampling a signal by sliding a window over the signal, each time by a certain step size, to obtain successive partial signal segments. The sliding sampling operation serves to acquire local features of the signal and to capture temporal variations of the signal. By sliding sampling over the signal, a plurality of partial images can be acquired, each representing a partial feature of the signal over a different period of time. In step S1211, the voltage detection signal is divided into a plurality of partial images by a slide sampling operation, which can be used for subsequent feature extraction. Through a feature extractor based on a convolutional neural network model, feature extraction can be performed on each local image, so as to obtain local feature graphs of a plurality of voltage waveforms, and the local feature graphs can be used for subsequent analysis and processing, such as pattern recognition, anomaly detection and the like. By sliding the sampling operation and feature extraction, the features of the voltage waveform can be more fully understood and analyzed. Further, convolutional neural networks (Convolutional Neural Network, CNN) are a deep learning model, particularly suited for processing tasks with grid structure data, such as image and speech recognition. The convolutional neural network is mainly characterized in that a convolutional layer and a pooling layer are utilized to extract spatial structural features of input data, and tasks such as classification or regression are performed through a full connection layer. The convolution layer uses convolution operations to capture local patterns and features in the input data, while the pooling layer is used to reduce the dimension of the feature map and preserve the most salient features. This hierarchical structure enables convolutional neural networks to automatically learn and extract useful features in the data, thereby enabling efficient feature representation and pattern recognition. Convolutional neural networks have achieved great success in the field of computer vision and are widely applied to tasks such as image classification, target detection, face recognition and the like. The method can learn the abstract features of the image from the original pixel data, and has good translation invariance and local perception performance. In addition, the convolutional neural network can also obtain better effects under the condition of less data through the migration learning and pre-training model.
Accordingly, in a specific example, in step S1212, extracting correlation features between the plurality of voltage waveform local feature graphs to obtain the voltage waveform global semantic feature vector includes: the overall characteristic distribution association effect of the voltage waveform local characteristic graphs is improved in a weighting mode of each characteristic matrix of the voltage waveform local characteristic graphs so as to obtain a plurality of optimized voltage waveform local characteristic graphs; respectively expanding the plurality of optimized voltage waveform local feature graphs into feature vectors to obtain a plurality of voltage waveform local expansion feature vectors; and passing the plurality of voltage waveform local expansion feature vectors through a context-dependent feature extractor based on a converter module to obtain the voltage waveform global semantic feature vector. It will be appreciated that the purpose of developing a plurality of optimized voltage waveform partial feature maps into feature vectors is to represent each partial feature map as a fixed length vector for subsequent feature extraction and analysis. There are various methods for expanding the feature vectors, and a common method includes expanding the feature matrices into one-dimensional vectors by rows or by columns, for example, each row or each column of each feature matrix may be connected to form one feature vector. The advantage of expanding the feature vector is that the information of a plurality of local feature graphs can be compressed and integrated to obtain a global semantic feature vector. This global feature vector may better represent the features and patterns of the overall voltage waveform, facilitating subsequent classification, regression, or processing of other tasks. After the feature vectors are expanded, they can be input into a context-dependent feature extractor based on the converter module, which further extracts the global semantic features of the voltage waveforms. These global features may be used for more advanced analysis and decision making, such as pattern recognition, anomaly detection, etc., to more fully understand and utilize the voltage waveform data.
It is worth mentioning that the converter module (Transformer module) is a deep learning model structure based on the attention mechanism (attention mechanism), originally proposed for natural language processing tasks such as machine translation and language generation, which is excellent in processing sequence data, and has been widely used in various fields. The primary role of the converter module is to capture the dependencies between the different positions in the input sequence and encode these dependencies into meaningful feature representations, which achieve this by means of a self-attention mechanism that can assign different weights according to the interrelationships between the different positions of the input sequence. The converter module is made up of a plurality of attention heads (attention heads), each of which may focus on a different part of the input sequence. It obtains the final output representation by mapping the input sequence to query, key and value spaces, respectively, and calculating the attention weights between them, and then weighting and summing the attention weights and the values. In the sequence modeling task, the converter module may be used to extract global semantic features of the input sequence, capture dependencies between different locations in the sequence, and use these features for subsequent tasks such as classification, generation, etc. It has a strong expression power and parallel computing power and thus tends to be more efficient than a Recurrent Neural Network (RNN) when processing long sequences and large-scale data. The converter module realizes global modeling and feature extraction of the sequence data through an attention mechanism.
Further, the voltage waveform global semantic feature vector is generated by a voltage waveform optimization generator based on an countermeasure generation network to obtain a generated voltage detection signal. And determining an output voltage based on the generated voltage detection signal. Further, a regulation control signal is generated based on a PID control algorithm and the output voltage so that the actual temperature is kept consistent with the set temperature.
Accordingly, in step S122, determining the output voltage based on the voltage waveform global semantic feature vector includes: the global semantic feature vector of the voltage waveform is optimized through a voltage waveform optimizing generator based on a countermeasure generating network to obtain a generated voltage detection signal; and determining the output voltage based on the generated voltage detection signal. It should be appreciated that the countermeasure generation network (Generative Adversarial Network, GAN for short) is a deep learning model structure composed of a generator and a arbiter. The method aims at generating a vivid sample through the generator, and judging the generated sample through the judging device, so that the quality of the generated sample is continuously improved by the generator. The basic idea of the antagonism generation network is to constantly optimize the game relationship between the generator and the arbiter by letting the generator and the arbiter learn antagonism to each other, and finally enable the generator to generate samples similar to real data. The goal of the generator is to generate a realistic sample so that the arbiter cannot accurately distinguish between the generated sample and the real sample; the object of the discriminator is to accurately discriminate the generated sample from the real sample, so that the quality of the generated sample is continuously improved by the generator. In step S122, a voltage waveform optimization generator based on the countermeasure generation network is used to generate a voltage detection signal, the generator receives as input a global semantic feature vector of the voltage waveform, and generates a synthesized voltage waveform similar to the true voltage waveform by learning. By continuously optimizing the generator, the generated voltage waveform is more similar to the real voltage waveform, so that a more accurate generated voltage detection signal is obtained. The generated voltage detection signal may be used to determine the output voltage. By processing and analyzing the generated voltage detection signal, a corresponding output voltage value can be obtained. Such a design may enable optimization and control of the voltage waveform by predicting the output voltage through a generator model. The challenge-generating network is a deep learning model for generating realistic samples, which can be generated to resemble real data by challenge learning of a generator and a arbiter. In voltage waveform optimization, a generator based on an countermeasure generation network may generate a voltage detection signal for determining an output voltage.
When the local images of the voltage detection signals are obtained through the feature extractor based on the convolutional neural network model, each feature matrix of the local feature images of the voltage waveforms is used for expressing the image feature semantics of the local voltage waveforms, and certain overall distribution differences exist in the local feature images of the voltage waveforms in consideration of the source semantic distribution differences of the local images and the distribution differences of semantic features based on the convolutional neural network model.
In this way, when the plurality of voltage waveform local feature graphs are respectively expanded into feature vectors to obtain a plurality of voltage waveform local expansion feature vectors, due to the dimension reduction operation of feature expansion, the overall distribution difference among the plurality of voltage waveform local expansion feature vectors is further increased, so that the effect of extracting the context-associated features of the plurality of voltage waveform local expansion feature vectors through the context-associated feature extractor based on the converter module is affected, and the expression effect of the voltage waveform global semantic feature vectors is also affected.
Therefore, the applicant of the present disclosure promotes the global feature distribution association effect of the plurality of voltage waveform local feature graphs by weighting the feature matrices of the plurality of voltage waveform local feature graphs along a channel, wherein the weighted feature vectors are self-tuning structured by directional bias constraints expressed by static scenes of the feature matrices.
Correspondingly, the weighting method for each feature matrix of the plurality of voltage waveform local feature graphs is used for improving the global feature distribution association effect of the plurality of voltage waveform local feature graphs to obtain a plurality of optimized voltage waveform local feature graphs, and the method comprises the following steps: respectively calculating weighted feature vectors of the voltage waveform local feature graphs to obtain a plurality of weighted feature vectors; and weighting each feature matrix of the corresponding plurality of voltage waveform local feature graphs based on the plurality of weighted feature vectors to obtain the plurality of optimized voltage waveform local feature graphs.
More specifically, calculating weighted feature vectors of the plurality of voltage waveform local feature graphs to obtain a plurality of weighted feature vectors, respectively, includes: carrying out channel linear transformation and conversion on each characteristic matrix of the multiple voltage waveform local characteristic diagrams to obtain multiple converted characteristic diagrams; and, based on the plurality of transformed feature maps, self-tuning structuring by directional bias guide constraints of static scene representation of feature matrices along a channel dimension of the plurality of voltage waveform local feature maps to calculate the plurality of weighted feature vectors according to an optimization formula; wherein, the optimization formula is:wherein each eigenvector channel linear transformation of the multiple voltage waveform partial eigenvectors is first converted to +.>Square matrix of>Is the sum of the channel numbers of the plurality of voltage waveform local feature graphs.
wherein ,is the +.th of the multiple post-conversion feature maps along the channel dimension>Characteristic matrix->Is a vector obtained by global pooling of each feature matrix of the plurality of transformed feature maps along the channel dimension,/I>Is the +.th of the multiple post-conversion feature maps along the channel dimension>First->Characteristic value of the location->、/> and />Representing addition, subtraction and multiplication by position, respectively,/->Is the plurality of weighted feature vectors.
That is, when weighting the respective feature matrices of the plurality of voltage waveform partial feature maps with the weighted feature vector, each static scene matrix along the channel dimension of the plurality of voltage waveform partial feature maps may be passed throughRelative to channel control vector->Using directed bias steering for expressing channel dimension associationsThe static feature scene is supported and self-tuned, so that the structuring of the high-dimensional feature manifold is carried out based on a specific convex polygon family (convex polytopes family) of the high-dimensional feature manifold set of the plurality of voltage waveform local feature graphs, which corresponds to the feature scene expressed by each feature matrix, so as to promote the explicit association between the image semantic expression of the scenerization of each feature matrix and the model feature extraction expression of the channel dimension, thereby promoting the global feature distribution association effect of the plurality of voltage waveform local feature graphs, promoting the expression effect of the voltage waveform global semantic feature vector, and improving the signal quality of the generated voltage detection signals, which are obtained by a voltage waveform optimization generator based on an antagonistic generation network, of the voltage waveform global semantic feature vector.
It should be understood that self-tuning structuring refers to a system or technique that automatically adjusts or optimizes its structure, in which the system is able to adapt to new requirements or optimize performance by sensing changes in environmental or internal conditions and automatically adjusting its structure or parameters based on those changes. Self-tuning architectures include adaptive control systems, optimization algorithms, neural networks, intelligent systems, and the like. The system has stronger adaptability and robustness, and can automatically adjust the behavior of the system under different environments or tasks, thereby improving the performance and efficiency of the system. Self-tuning structuring can be achieved based on feedback control adaptive algorithms, genetic algorithms, fuzzy logic, machine learning and the like, and the methods can automatically adjust the structure, parameters or behaviors of the system according to the requirements and targets of the system so as to achieve self-adaption and optimization.
In summary, according to the semiconductor-based heating control method of the embodiment of the present disclosure, the voltage detection signal may be feature-encoded-feature-decoded based on an analysis circuit of an encoder-decoder structure to improve the detection accuracy of the output voltage value, thereby improving the detection accuracy of the real-time temperature detection value, so that the adjustment control signal may be generated based on a PID algorithm to keep the actual temperature consistent with the set temperature.
Fig. 5 shows a block diagram of a semiconductor-based heating control system 100 according to an embodiment of the present disclosure. As shown in fig. 5, a semiconductor-based heating control system 100 according to an embodiment of the present disclosure includes: a signal acquisition module 110 for acquiring a voltage detection signal acquired by the semiconductor heating device; the circuit analysis module 120 is configured to perform circuit analysis on the voltage detection signal to obtain an output voltage; and a temperature control module 130 for generating an adjustment control signal based on the PID control algorithm and the output voltage so that the actual temperature is consistent with the set temperature.
In one possible implementation, the circuit analysis module 120 includes: the waveform feature extraction unit is used for extracting waveform features of the voltage detection signals to obtain a voltage waveform global semantic feature vector; and an output voltage control unit for determining the output voltage based on the voltage waveform global semantic feature vector.
In one possible implementation manner, the waveform feature extraction unit includes: the sampling and local feature extraction subunit is used for sampling and extracting local features of the voltage detection signals to obtain a plurality of voltage waveform local feature graphs; and a correlation feature extraction subunit, configured to extract correlation features between the plurality of voltage waveform local feature graphs to obtain the voltage waveform global semantic feature vector.
Here, it will be understood by those skilled in the art that the specific functions and operations of the respective units and modules in the above-described semiconductor-based heating control system 100 have been described in detail in the above description of the semiconductor-based heating control method with reference to fig. 1 to 4, and thus, repetitive descriptions thereof will be omitted.
As described above, the semiconductor-based heating control system 100 according to the embodiment of the present disclosure may be implemented in various wireless terminals, such as a server or the like having a semiconductor-based heating control algorithm. In one possible implementation, the semiconductor-based heating control system 100 according to embodiments of the present disclosure may be integrated into a wireless terminal as one software module and/or hardware module. For example, the semiconductor-based heating control system 100 may be a software module in the operating system of the wireless terminal, or may be an application developed for the wireless terminal; of course, the semiconductor-based heating control system 100 may also be one of a number of hardware modules of the wireless terminal.
Alternatively, in another example, the semiconductor-based heating control system 100 and the wireless terminal may be separate devices, and the semiconductor-based heating control system 100 may be connected to the wireless terminal through a wired and/or wireless network and transmit interactive information in a agreed data format.
Fig. 6 illustrates an application scenario diagram of a semiconductor-based heating control method according to an embodiment of the present disclosure. As shown in fig. 6, in this application scenario, first, a voltage detection signal (e.g., D illustrated in fig. 6) acquired by a semiconductor heating device (e.g., N illustrated in fig. 6) is acquired, and then the voltage detection signal is input to a server (e.g., S illustrated in fig. 6) in which a semiconductor-based heating control algorithm is disposed, wherein the server is capable of processing the voltage detection signal using the semiconductor-based heating control algorithm to obtain a generated voltage detection signal, and then the output voltage is determined based on the generated voltage detection signal.
The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The foregoing description of the embodiments of the present disclosure has been presented for purposes of illustration and description, and is not intended to be exhaustive or limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the various embodiments described. The terminology used herein was chosen in order to best explain the principles of the embodiments, the practical application, or the improvement of technology in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.
Claims (6)
1. A semiconductor-based heating control method, comprising: acquiring a voltage detection signal acquired by a semiconductor heating device; performing circuit analysis on the voltage detection signal to obtain an output voltage; and generating an adjustment control signal based on a PID control algorithm and the output voltage to maintain the actual temperature consistent with the set temperature; wherein, carry out circuit analysis to the voltage detection signal in order to obtain output voltage, include: sampling the voltage detection signals and extracting local characteristics to obtain a plurality of voltage waveform local characteristic diagrams; the overall characteristic distribution association effect of the voltage waveform local characteristic graphs is improved in a weighting mode of each characteristic matrix of the voltage waveform local characteristic graphs so as to obtain a plurality of optimized voltage waveform local characteristic graphs; respectively expanding the plurality of optimized voltage waveform local feature graphs into feature vectors to obtain a plurality of voltage waveform local expansion feature vectors; passing the plurality of voltage waveform local expansion feature vectors through a context-dependent feature extractor based on a converter module to obtain a voltage waveform global semantic feature vector; and determining the output voltage based on the voltage waveform global semantic feature vector.
2. The semiconductor-based heating control method according to claim 1, wherein sampling and local feature extraction of the voltage detection signal to obtain a plurality of voltage waveform local feature maps comprises: performing sliding sampling operation on the voltage detection signals to obtain a plurality of partial images; and respectively passing the plurality of partial images through a characteristic extractor based on a convolutional neural network model to obtain a plurality of voltage waveform partial characteristic diagrams.
3. The semiconductor-based heating control method according to claim 2, wherein weighting each of the feature matrices of the plurality of voltage waveform local feature maps to promote global feature distribution correlation effects of the plurality of voltage waveform local feature maps to obtain a plurality of optimized voltage waveform local feature maps, comprises: respectively calculating weighted feature vectors of the voltage waveform local feature graphs to obtain a plurality of weighted feature vectors; and weighting each feature matrix of the corresponding plurality of voltage waveform local feature graphs based on the plurality of weighted feature vectors to obtain the plurality of optimized voltage waveform local feature graphs.
4. The semiconductor-based heating control method according to claim 3, wherein calculating weighted feature vectors of the plurality of voltage waveform local feature maps to obtain a plurality of weighted feature vectors, respectively, comprises: carrying out channel linear transformation and conversion on each characteristic matrix of the multiple voltage waveform local characteristic diagrams to obtain multiple converted characteristic diagrams; and based on the plurality of converted feature maps, calculating the plurality of weighted feature vectors by self-tuning structuring through directional bias constraints of static scene representation of feature matrices along channel dimensions of the plurality of voltage waveform local feature maps according to the following optimization formula; wherein, the optimization formula is: wherein ,/>Is the +.th of the multiple post-conversion feature maps along the channel dimension>Characteristic matrix->Is a vector obtained by global pooling of each feature matrix of the plurality of transformed feature maps along the channel dimension,/I>Is the +.th of the multiple post-conversion feature maps along the channel dimension>First->Characteristic value of the location->、/> and />Representing addition, subtraction and multiplication by position, respectively,/->Is the plurality of weighted feature vectors.
5. The semiconductor-based heating control method of claim 4, wherein determining the output voltage based on the voltage waveform global semantic feature vector comprises: the global semantic feature vector of the voltage waveform is optimized through a voltage waveform optimizing generator based on a countermeasure generating network to obtain a generated voltage detection signal; and determining the output voltage based on the generated voltage detection signal.
6. A semiconductor-based heating control system using the semiconductor-based heating control method according to claim 1, comprising: the signal acquisition module is used for acquiring a voltage detection signal acquired by the semiconductor heating device; the circuit analysis module is used for carrying out circuit analysis on the voltage detection signal to obtain output voltage; and a temperature control module for generating an adjustment control signal based on a PID control algorithm and the output voltage so that an actual temperature is maintained consistent with a set temperature; wherein, the circuit analysis module includes: the characteristic extraction unit is used for sampling the voltage detection signals and extracting local characteristics to obtain a plurality of voltage waveform local characteristic diagrams; the weighting optimization unit is used for improving the global feature distribution association effect of the voltage waveform local feature graphs in a weighting mode of each feature matrix of the voltage waveform local feature graphs so as to obtain a plurality of optimized voltage waveform local feature graphs; the characteristic diagram unfolding unit is used for respectively unfolding the plurality of optimized voltage waveform local characteristic diagrams into characteristic vectors to obtain a plurality of voltage waveform local unfolding characteristic vectors; the context correlation coding unit is used for enabling the plurality of voltage waveform local expansion feature vectors to pass through the context correlation feature extractor based on the converter module to obtain a voltage waveform global semantic feature vector; and an output voltage control unit for determining the output voltage based on the voltage waveform global semantic feature vector.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310929535.6A CN116647946B (en) | 2023-07-27 | 2023-07-27 | Semiconductor-based heating control system and method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310929535.6A CN116647946B (en) | 2023-07-27 | 2023-07-27 | Semiconductor-based heating control system and method thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116647946A true CN116647946A (en) | 2023-08-25 |
| CN116647946B CN116647946B (en) | 2023-10-13 |
Family
ID=87643835
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310929535.6A Active CN116647946B (en) | 2023-07-27 | 2023-07-27 | Semiconductor-based heating control system and method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116647946B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117288705A (en) * | 2023-09-25 | 2023-12-26 | 武汉怡特环保科技有限公司 | Miniature ultraviolet absorption method multipath absorption tank ozone gas sensor based on Internet of things |
| CN117310351A (en) * | 2023-11-30 | 2023-12-29 | 广州思林杰科技股份有限公司 | Detection equipment for PD protocol of factory charging wire |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101093166A (en) * | 2007-08-10 | 2007-12-26 | 北京航空航天大学 | Device for controlling average wavelength of light source in broad spectrum based on monitoring detector |
| CN112036320A (en) * | 2020-09-07 | 2020-12-04 | 华北电力大学（保定） | Partial discharge state identification method and device, discharge simulator and identification equipment |
| WO2021035807A1 (en) * | 2019-08-23 | 2021-03-04 | 深圳大学 | Target tracking method and device fusing optical flow information and siamese framework |
| KR102437350B1 (en) * | 2021-12-21 | 2022-08-29 | (주)씨어스테크놀로지 | Method and Apparatus for Detecting Multiple beat by using Electrocardiogram Global Feature Vector |
| CN115047925A (en) * | 2022-02-28 | 2022-09-13 | 中国科学院国家空间科学中心 | Passive radiation type constant temperature control system and control method based on PID controller |
| CN115375691A (en) * | 2022-10-26 | 2022-11-22 | 济宁九德半导体科技有限公司 | Image-based semiconductor diffusion paper source defect detection system and method thereof |
| CN115393779A (en) * | 2022-10-31 | 2022-11-25 | 济宁九德半导体科技有限公司 | Control system and control method for laser cladding metal ball manufacturing |
| US20230095182A1 (en) * | 2021-03-08 | 2023-03-30 | Tencent Technology (Shenzhen) Company Limited | Method and apparatus for extracting biological features, device, medium, and program product |
| CN116420941A (en) * | 2023-05-11 | 2023-07-14 | 深圳中科云信息技术有限公司 | Microwave heating control system and method for electronic atomization equipment |
| CN116470885A (en) * | 2023-06-17 | 2023-07-21 | 浙江佳环电子有限公司 | High-voltage pulse circuit system and control method thereof |
-
2023
- 2023-07-27 CN CN202310929535.6A patent/CN116647946B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101093166A (en) * | 2007-08-10 | 2007-12-26 | 北京航空航天大学 | Device for controlling average wavelength of light source in broad spectrum based on monitoring detector |
| WO2021035807A1 (en) * | 2019-08-23 | 2021-03-04 | 深圳大学 | Target tracking method and device fusing optical flow information and siamese framework |
| CN112036320A (en) * | 2020-09-07 | 2020-12-04 | 华北电力大学（保定） | Partial discharge state identification method and device, discharge simulator and identification equipment |
| US20230095182A1 (en) * | 2021-03-08 | 2023-03-30 | Tencent Technology (Shenzhen) Company Limited | Method and apparatus for extracting biological features, device, medium, and program product |
| KR102437350B1 (en) * | 2021-12-21 | 2022-08-29 | (주)씨어스테크놀로지 | Method and Apparatus for Detecting Multiple beat by using Electrocardiogram Global Feature Vector |
| CN115047925A (en) * | 2022-02-28 | 2022-09-13 | 中国科学院国家空间科学中心 | Passive radiation type constant temperature control system and control method based on PID controller |
| CN115375691A (en) * | 2022-10-26 | 2022-11-22 | 济宁九德半导体科技有限公司 | Image-based semiconductor diffusion paper source defect detection system and method thereof |
| CN115393779A (en) * | 2022-10-31 | 2022-11-25 | 济宁九德半导体科技有限公司 | Control system and control method for laser cladding metal ball manufacturing |
| CN116420941A (en) * | 2023-05-11 | 2023-07-14 | 深圳中科云信息技术有限公司 | Microwave heating control system and method for electronic atomization equipment |
| CN116470885A (en) * | 2023-06-17 | 2023-07-21 | 浙江佳环电子有限公司 | High-voltage pulse circuit system and control method thereof |
Non-Patent Citations (1)
| Title |
|---|
| 高崧淋;侯立梅;: "基于PID控制的半导体激光器温度检测电路", 信息与电脑(理论版), no. 09 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117288705A (en) * | 2023-09-25 | 2023-12-26 | 武汉怡特环保科技有限公司 | Miniature ultraviolet absorption method multipath absorption tank ozone gas sensor based on Internet of things |
| CN117310351A (en) * | 2023-11-30 | 2023-12-29 | 广州思林杰科技股份有限公司 | Detection equipment for PD protocol of factory charging wire |
| CN117310351B (en) * | 2023-11-30 | 2024-03-19 | 广州思林杰科技股份有限公司 | Detection equipment for PD protocol of factory charging wire |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116647946B (en) | 2023-10-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116647946B (en) | Semiconductor-based heating control system and method thereof | |
| Afif et al. | Deep learning based application for indoor scene recognition | |
| CN107369166B (en) | Target tracking method and system based on multi-resolution neural network | |
| CN106355195B (en) | System and method for measuring image definition value | |
| CN108548539B (en) | Navigation method and device based on image recognition, terminal and readable storage medium | |
| CN111709295A (en) | SSD-MobileNet-based real-time gesture detection and recognition method and system | |
| CN116272363A (en) | Ship exhaust gas mixed desulfurization system and method thereof | |
| Papa et al. | METER: a mobile vision transformer architecture for monocular depth estimation | |
| CN116894210B (en) | Electronic device comprising force sensor and data processing method | |
| CN117150272A (en) | Unmanned aerial vehicle real-time track prediction method and system based on artificial intelligence | |
| CN117190657A (en) | Tunnel type drying production line and method thereof | |
| CN117072740A (en) | Automatic control system and method for pneumatic regulating valve | |
| Mikhaylenko et al. | Analysis of the predicting neural network person recognition system by picture image | |
| CN116468894A (en) | Distance self-adaptive mask generation method for supervised learning of lithium battery pole piece | |
| CN113221671A (en) | Environment-independent action identification method and system based on gradient and wireless signal | |
| CN116266249A (en) | IGBT module residual life span working condition prediction method | |
| CN108696722B (en) | Target monitoring method, system and device and storage medium | |
| Li et al. | Production evaluation of citrus fruits based on the yolov5 compressed by knowledge distillation | |
| Kumar et al. | Implementation of sign language recognition with TinyML using smart gloves | |
| CN116777183B (en) | Unmanned ship cluster intelligent scheduling method and system | |
| Chowdhury et al. | A novel fusion method for intense heat detection and localization solely based on surface temperature with multi-mode DEA feedback | |
| Raksha et al. | Home Automation through Hand Gestures Using ResNet50 and 3D-CNN | |
| Hu et al. | Theory of spatiotemporal deep analogs and their application to solar forecasting | |
| CN116880198B (en) | Power equipment self-adaptive control system and method for supersonic unmanned aerial vehicle | |
| CN116393188B (en) | Microfluidic chip and method suitable for capturing circulating tumor cells |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |