CN116560434A - Temperature control method and device, electronic equipment and storage medium - Google Patents
Temperature control method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN116560434A CN116560434A CN202310193486.4A CN202310193486A CN116560434A CN 116560434 A CN116560434 A CN 116560434A CN 202310193486 A CN202310193486 A CN 202310193486A CN 116560434 A CN116560434 A CN 116560434A
- Authority
- CN
- China
- Prior art keywords
- temperature
- neural network
- network
- power supply
- propagation path
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000003860 storage Methods 0.000 title claims abstract description 15
- 238000013528 artificial neural network Methods 0.000 claims abstract description 169
- 239000004065 semiconductor Substances 0.000 claims abstract description 128
- 230000002441 reversible effect Effects 0.000 claims description 93
- 238000005259 measurement Methods 0.000 claims description 63
- 230000007613 environmental effect Effects 0.000 claims description 42
- 238000004590 computer program Methods 0.000 claims description 27
- 230000002457 bidirectional effect Effects 0.000 claims description 22
- 210000002569 neuron Anatomy 0.000 claims description 13
- 238000013473 artificial intelligence Methods 0.000 abstract description 2
- 238000012549 training Methods 0.000 description 18
- 230000008569 process Effects 0.000 description 11
- 230000000694 effects Effects 0.000 description 9
- 230000001105 regulatory effect Effects 0.000 description 9
- 230000006870 function Effects 0.000 description 7
- 230000008878 coupling Effects 0.000 description 6
- 238000010168 coupling process Methods 0.000 description 6
- 238000005859 coupling reaction Methods 0.000 description 6
- 230000001276 controlling effect Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 230000004044 response Effects 0.000 description 5
- 238000001514 detection method Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- XUIMIQQOPSSXEZ-UHFFFAOYSA-N Silicon Chemical compound [Si] XUIMIQQOPSSXEZ-UHFFFAOYSA-N 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 229910052710 silicon Inorganic materials 0.000 description 2
- 239000010703 silicon Substances 0.000 description 2
- 230000003042 antagnostic effect Effects 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D23/00—Control of temperature
- G05D23/19—Control of temperature characterised by the use of electric means
- G05D23/30—Automatic controllers with an auxiliary heating device affecting the sensing element, e.g. for anticipating change of temperature
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Testing Of Individual Semiconductor Devices (AREA)
Abstract
The application is applicable to the technical field of artificial intelligence, and provides a temperature control method, a device, electronic equipment and a storage medium, wherein the method comprises the following steps: inputting the obtained current internal temperature and current environment information of the semiconductor measuring machine to a forward propagation path of a neural network or a forward propagation network to obtain power supply; if the power supply power is obtained through the forward propagation path, inputting the power supply power into the backward propagation path to obtain an internal regulation temperature corresponding to the power supply power, wherein the temperature difference between the internal regulation temperature and a preset internal target temperature is smaller than or equal to a preset temperature difference; if the power supply power is obtained through the forward propagation network, the power supply power is input into the backward propagation network to obtain the built-in regulation and control temperature corresponding to the power supply power, so that the problems that after-sales and manpower of enterprises are consumed because parameters of the PID controller need to be changed again manually when the environment where the semiconductor measuring machine is located is changed are solved.
Description
Technical Field
The application belongs to the technical field of artificial intelligence, and particularly relates to a temperature control method, a temperature control device, electronic equipment and a storage medium.
Background
Semiconductor measurement is an indispensable process for detecting yield and controlling quality in the semiconductor industry, and precise control of temperature is important in semiconductor measurement. Semiconductor silicon wafer products are generally sensitive to temperature, and the physical characteristics of the silicon wafer can be greatly changed under different temperature conditions, so that the measurement result, namely the yield and the yield of the chip are greatly affected. In order to ensure the yield and the rate of finished products of chips, a constant temperature system is required to be arranged in the environment where the semiconductor measurement is needed, namely, the temperature in the semiconductor measurement machine is required to be kept within 0.03 ℃ above and below the constant temperature.
At present, a proportional-integral-differential (PID) controller is generally used at home and abroad to control the temperature in the semiconductor measuring machine.
However, the PID controller requires an engineer to control the temperature of the semiconductor measuring apparatus to fluctuate within a certain interval by changing parameters of the PID controller, such as a proportional term, an integral term, and a differential term. If the environment of the semiconductor measuring machine is changed, the parameters of the PID controller need to be changed again manually, so that the after-sales and manpower of enterprises are very consumed.
Disclosure of Invention
The embodiment of the application provides a temperature control method, a temperature control device, electronic equipment and a storage medium, which can solve the problem that after-sales and manpower of enterprises are consumed because parameters of a PID controller need to be changed again manually when the environment where a semiconductor measuring machine is located is changed.
In a first aspect, an embodiment of the present application provides a method for controlling a temperature, where the method includes:
acquiring current internal temperature and current environment information of a semiconductor measurement machine, wherein the current environment information is used for indicating the current environment of the semiconductor measurement machine;
inputting the current in-machine temperature and the current environment information into a forward propagation path or a forward propagation network of a neural network to obtain power output by the neural network, wherein the power is required to be provided by a power supply when the in-machine temperature of the semiconductor measuring machine is maintained at a preset in-machine target temperature;
if the power supply power is obtained through the forward propagation path, inputting the power supply power into a reverse propagation path of the neural network to obtain an internal regulation temperature which is output by the reverse propagation path and corresponds to the power supply power, wherein the temperature difference between the internal regulation temperature and the preset internal target temperature is smaller than or equal to a preset temperature difference;
And if the power supply power is obtained through the forward propagation network, inputting the power supply power into a backward propagation network of the neural network to obtain the built-in regulation temperature corresponding to the power supply power, which is output by the backward propagation network.
Optionally, the input terminal of the neural network includes at least two neurons, through which the environmental information can be input into the neural network as an influencing factor.
Optionally, the neural network is a reversible neural network, and the reversible neural network includes a forward propagation path and a reverse propagation path.
Optionally, the forward propagation path and the backward propagation path each include a generation type countermeasure network, the generation type countermeasure network includes a generator and a discriminator, and the weight value of the reversible neural network is adjusted by the generator and the discriminator.
Optionally, the neural network is a bidirectional neural network, and the bidirectional neural network includes a forward propagation network, a backward propagation network, and a dynamic network, where the forward propagation network and the backward propagation network are coupled together through the dynamic network.
Optionally, the forward propagation network of the bidirectional neural network is the same as the forward propagation path of the reversible neural network, and the backward propagation network of the bidirectional neural network includes a generative countermeasure network and a normal neural network, and the normal neural network is different from the backward propagation path of the reversible neural network.
Optionally, the current environmental information includes at least one of a temperature of an environment in which the semiconductor measurement machine is located, a humidity of the environment in which the semiconductor measurement machine is located, a hot air port temperature of the semiconductor measurement machine, and a butterfly valve port temperature of the semiconductor measurement machine.
In a second aspect, an embodiment of the present application provides a temperature control device, where the temperature control device includes:
the current information acquisition module is used for acquiring the current internal temperature and the current environment information of the semiconductor measurement machine, wherein the current environment information is used for indicating the current environment of the semiconductor measurement machine;
the forward input module is used for inputting the current internal temperature and the current environmental information into a forward propagation path or a forward propagation network of a neural network to obtain power output by the neural network, wherein the power is required to be provided by a power supply when the internal temperature of the semiconductor measuring machine is maintained at a preset internal target temperature;
the first reverse output module is used for inputting the power supply power into a reverse propagation path of the neural network if the power supply power is obtained through the forward propagation path, so as to obtain an internal regulation temperature which is output by the reverse propagation path and corresponds to the power supply power, wherein the temperature difference between the internal regulation temperature and the preset internal target temperature is smaller than or equal to a preset temperature difference;
And the second reverse output module is used for inputting the power supply power into a reverse propagation network of the neural network if the power supply power is obtained through the forward propagation network, so as to obtain the built-in regulation temperature which is output by the reverse propagation network and corresponds to the power supply power.
Optionally, the input terminal of the neural network includes at least two neurons, through which the environmental information can be input into the neural network as an influencing factor.
Optionally, the neural network is a reversible neural network, and the reversible neural network includes a forward propagation path and a reverse propagation path.
Optionally, the forward propagation path and the backward propagation path each include a generation type countermeasure network, the generation type countermeasure network includes a generator and a discriminator, and the weight value of the reversible neural network is adjusted by the generator and the discriminator.
Optionally, the neural network is a bidirectional neural network, and the bidirectional neural network includes a forward propagation network, a backward propagation network, and a dynamic network, where the forward propagation network and the backward propagation network are coupled together through the dynamic network.
Optionally, the forward propagation network of the bidirectional neural network is the same as the forward propagation path of the reversible neural network, and the backward propagation network of the bidirectional neural network includes a generative countermeasure network and a normal neural network, and the normal neural network is different from the backward propagation path of the reversible neural network.
Optionally, the current environmental information includes at least one of a temperature of an environment in which the semiconductor measurement machine is located, a humidity of the environment in which the semiconductor measurement machine is located, a hot air port temperature of the semiconductor measurement machine, and a butterfly valve port temperature of the semiconductor measurement machine.
In a third aspect, an embodiment of the present application provides an electronic device, including:
a memory, a processor and a computer program stored in the memory and executable on the processor, which when executed by the processor realizes the steps of the temperature control method according to the first aspect.
In a fourth aspect, embodiments of the present application provide a computer-readable storage medium, comprising: the computer readable storage medium stores a computer program which, when executed by a processor, implements the steps of the temperature control method described in the first aspect.
In a fifth aspect, embodiments of the present application provide a computer program product for causing an electronic device to perform the steps of the method for controlling temperature according to the first aspect described above, when the computer program product is run on the electronic device.
Compared with the prior art, the embodiment of the application has the beneficial effects that: because the data input to the neural network comprises the current internal temperature and the current environmental information affecting the semiconductor measuring machine, the forward propagation path or the forward propagation network of the neural network can control the power supply according to the current internal temperature and the current environmental information of the semiconductor measuring machine, wherein the power supply power is the power required to be supplied by the power supply when the internal temperature is maintained at the preset internal target temperature; the back propagation path or the back propagation network of the neural network can control the temperature in the machine according to the power supply output by the forward propagation path or the forward propagation network, so that the temperature in the machine of the semiconductor measuring machine is regulated and controlled, and meanwhile, the temperature difference between the regulated and controlled temperature in the machine and the preset target temperature in the machine is smaller than or equal to the preset temperature difference, so that the effect that the temperature in the machine of the semiconductor measuring machine reaches the constant temperature can be realized. That is, in the embodiment of the present application, when the neural network controls the temperature in the semiconductor measurement machine, the temperature in the semiconductor measurement machine can be automatically controlled in real time according to the current environmental information affecting the temperature in the semiconductor measurement machine, and because the whole temperature control process does not need to be manually involved, the problem that the parameters of the PID controller need to be manually changed again when the environment where the semiconductor measurement machine is located is changed, which results in the consumption of after-sales and manpower of enterprises can be solved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required for the embodiments or the description of the prior art will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of a temperature control method according to an embodiment of the present disclosure;
FIG. 2 is a schematic diagram of a reversible neural network according to an embodiment of the present disclosure;
FIG. 3 is a schematic diagram of a bidirectional neural network according to an embodiment of the present disclosure;
FIG. 4 is a schematic structural diagram of a temperature control device according to an embodiment of the present disclosure;
fig. 5 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
In the following description, for purposes of explanation and not limitation, specific details are set forth, such as particular system configurations, techniques, etc. in order to provide a thorough understanding of the embodiments of the present application. It will be apparent, however, to one skilled in the art that the present application may be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present application with unnecessary detail.
It should be understood that the terms "comprises" and/or "comprising," when used in this specification and the appended claims, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
It should also be understood that the term "and/or" as used in this specification and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes such combinations.
As used in this specification and the appended claims, the term "if" may be interpreted as "when..once" or "in response to a determination" or "in response to detection" depending on the context. Similarly, the phrase "if a determination" or "if a [ described condition or event ] is detected" may be interpreted in the context of meaning "upon determination" or "in response to determination" or "upon detection of a [ described condition or event ]" or "in response to detection of a [ described condition or event ]".
Reference in the specification to "one embodiment" or "some embodiments" or the like means that a particular feature, structure, or characteristic described in connection with the embodiment is included in one or more embodiments of the application. Thus, appearances of the phrases "in one embodiment," "in some embodiments," "in other embodiments," and the like in the specification are not necessarily all referring to the same embodiment, but mean "one or more but not all embodiments" unless expressly specified otherwise. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless expressly specified otherwise.
The parameters of the PID controller, such as a proportional term, an integral term and a differential term, are changed to control the temperature in the semiconductor measuring machine to fluctuate in a certain interval, so that the semiconductor measuring machine has the advantages of simple control and simple logic.
However, in the above process, the engineer is required to tune parameters by means of MATLAB and formulation, i.e. changing the parameters of the PID controller requires the involvement of the engineer. In addition, in the process of parameter adjustment, the time is long, the working procedure is complicated, and the experience and the manipulation of engineers have great influence on the parameter adjustment result. In addition, when the environment where the semiconductor measuring machine is located is changed, the PID controller cannot learn the corresponding characteristics from the environment, cannot realize the self-adaptive regulation and control of the environment, and needs engineers to manually readjust parameters and restart the control process, so that after-sales and manpower of enterprises are consumed, and the temperature control effect is unstable.
In summary, the method of controlling the temperature in the semiconductor measuring machine by using the PID controller is faced with the problems of difficult parameter adjustment, unstable control effect and the like.
In the embodiment of the application, the temperature in the semiconductor measuring machine is controlled through the neural network. The data input to the neural network comprises the current internal temperature of the semiconductor measuring machine and the current environmental information of the semiconductor measuring machine, the neural network can regulate and control the internal temperature of the semiconductor measuring machine according to the current environmental information of the semiconductor measuring machine, the temperature difference between the regulated and controlled internal regulating temperature and the preset target internal temperature is smaller than or equal to the preset temperature difference, real-time control of the internal temperature of the semiconductor measuring machine according to the current environmental information affecting the internal temperature of the semiconductor measuring machine is achieved, the effect of constant temperature is achieved, manual participation is not needed in the whole temperature control process, and the control effect is stable.
A method for controlling the temperature of the apparatus according to the embodiment of the present application will be described below with reference to the accompanying drawings.
Fig. 1 shows a flow chart of a temperature control method provided in an embodiment of the present application, which may be applied to a temperature control device, including steps S110 to S140, where the specific implementation principle of each step is as follows:
s110, acquiring the current temperature and the current environment information of the semiconductor measuring machine, wherein the current environment information is used for indicating the current environment of the semiconductor measuring machine.
The environmental information in the embodiment of the application may include temperature, humidity, and the like. In consideration of the conduction characteristics of temperature, humidity, etc., when the environmental information of the semiconductor measurement apparatus changes, the temperature inside the semiconductor measurement apparatus is likely to change. In the embodiment of the application, the current environmental information of the semiconductor measuring machine is acquired, so that more accurate in-machine regulation temperature can be obtained later.
In the embodiment of the application, the current environmental information of the semiconductor measuring machine can be collected through the sensor, and the collected current environmental information is sent to the control device of the temperature of the semiconductor measuring machine through the sensor. For example, when the environmental information includes temperature, the temperature in the current environment may be acquired by a temperature sensor; when the environmental information includes humidity, the humidity of the current environment may be collected by a humidity sensor.
S120, inputting the current internal temperature and the current environmental information into a forward propagation path or a forward propagation network of a neural network to obtain power output by the neural network, wherein the power is required to be provided by a power supply when the internal temperature of the semiconductor measuring machine is maintained at a preset internal target temperature.
In this embodiment, the neural network may include a forward propagation path and a backward propagation path, and may also include a forward propagation network and a backward propagation network.
The forward propagation path and the backward propagation path are forward and backward propagation directions in the reversible neural network, and the forward propagation path and the backward propagation path are part of the reversible neural network.
Wherein the forward propagation network and the backward propagation network are two independent neural networks in the bi-directional propagation network.
After receiving the current environmental information sent by the sensor, the temperature control device can input the current environmental information into a forward propagation path or a forward propagation network of the neural network, and the forward propagation path or the forward propagation network adjusts and outputs the power of the power supply according to the input current temperature and the current environmental information.
The power supply power is required to supply power when the internal temperature of the semiconductor measuring machine is maintained at a preset internal target temperature. When the temperature in the semiconductor measuring machine is maintained at the preset target temperature in the machine, the temperature in the semiconductor measuring machine is maintained in a stable state, namely the inside of the semiconductor measuring machine is basically in a constant temperature state.
The preset target temperature in the machine is an artificially set temperature in the machine, and when the temperature difference between the current temperature in the machine and the preset target temperature in the machine is within a preset temperature difference range, the machine can work normally.
And S130, inputting the power supply power into a reverse propagation path of the neural network to obtain an in-machine regulation temperature corresponding to the power supply power output by the reverse propagation path, wherein the temperature difference between the in-machine regulation temperature and the preset in-machine target temperature is smaller than or equal to a preset temperature difference if the power supply power is obtained through the forward propagation path.
In this embodiment, the power supply power output from the forward propagation path of the neural network may be input to the corresponding reverse propagation path, and the in-machine regulation temperature corresponding to the power supply power may be output through the reverse propagation path.
The temperature difference between the finally obtained built-in regulation temperature and the preset built-in target temperature is smaller than or equal to the preset temperature difference.
The preset temperature difference may be not greater than 0.03 degrees, and may be specifically set according to the operating temperature requirement of the semiconductor measuring apparatus, which is not limited herein.
When the temperature difference between the built-in regulation temperature and the preset built-in target temperature is smaller than or equal to the preset temperature difference, the built-in regulation temperature of the semiconductor measuring machine is basically the same as the built-in temperature of the semiconductor measuring machine when the semiconductor measuring machine works normally, and the semiconductor measuring machine basically keeps a constant temperature state after the built-in temperature of the semiconductor measuring machine is regulated to the built-in regulation temperature.
And S140, inputting the power supply power into a back propagation network of the neural network if the power supply power is obtained through the forward propagation network, and obtaining the built-in regulation temperature corresponding to the power supply power output by the back propagation network.
In this embodiment, the power output by the forward propagation network of the neural network may be input to the corresponding reverse propagation network, and the in-machine regulation temperature corresponding to the power may be output through the reverse propagation network.
It should be understood that, in the steps S110 to S140, since the data input to the neural network includes, in addition to the current in-plane temperature, the current environmental information affecting the semiconductor measurement device, the forward propagation path or the forward propagation network of the neural network may control the power of the power supply according to the current in-plane temperature and the current environmental information of the semiconductor measurement device, where the power of the power supply is the power that the power supply needs to provide when the in-plane temperature is maintained at the preset in-plane target temperature; the back propagation path or the back propagation network of the neural network can control the temperature in the machine according to the power supply output by the forward propagation path or the forward propagation network, so that the temperature in the machine of the semiconductor measuring machine is regulated and controlled, and meanwhile, the temperature difference between the regulated and controlled temperature in the machine and the preset target temperature in the machine is smaller than or equal to the preset temperature difference, so that the effect that the temperature in the machine of the semiconductor measuring machine reaches the constant temperature can be realized. That is, in the embodiment of the present application, when the neural network controls the temperature in the semiconductor measurement machine, the temperature in the semiconductor measurement machine can be automatically controlled in real time according to the current environmental information affecting the temperature in the semiconductor measurement machine, and because the whole temperature control process does not need to be manually involved, the problem that the parameters of the PID controller need to be manually changed again when the environment where the semiconductor measurement machine is located is changed, which results in the consumption of after-sales and manpower of enterprises can be solved.
In some embodiments, the input terminal of the neural network includes at least two neurons, through which the environmental information can be input into the neural network as an influence factor.
In this embodiment, the environmental information of the semiconductor measurement apparatus may include at least one of a temperature of an environment in which the semiconductor measurement apparatus is located, a humidity of the environment in which the semiconductor measurement apparatus is located, a hot air port temperature of the semiconductor measurement apparatus, and a butterfly valve port temperature of the semiconductor measurement apparatus.
The input end of the neural network comprises at least two neurons, and at least one of the current internal temperature and the environment information of the semiconductor measuring machine can be input into the at least two neurons.
For example, when the input end of the neural network includes three neurons, the current internal temperature of the semiconductor measurement machine and the temperature of the environment where the semiconductor measurement machine is located can be respectively input into the two neurons; when the input end of the neural network comprises three neurons, the current built-in temperature of the semiconductor measuring machine, the temperature of the environment where the semiconductor measuring machine is positioned and the humidity of the environment where the semiconductor measuring machine is positioned can be respectively input into the three neurons.
The input end of the neural network comprises at least two neurons, at least one of the environmental information of the semiconductor measuring machine can be used as an influence factor to be input into the neural network, and the neural network controls the temperature in the semiconductor measuring machine according to the environmental information.
It should be understood that, because of factors such as heat conduction, environmental information of the external environment affects the temperature in the semiconductor measurement machine, and in the embodiment of the present application, the environmental information is input into the neural network as an affecting factor to control the temperature in the semiconductor measurement machine, so that the regulated temperature is more accurate.
In some embodiments, the neural network is a reversible neural network that includes a forward propagation path and a reverse propagation path.
In the reversible neural network, the forward propagation path and the backward propagation path are integrated and are part of the reversible neural network, so that the two directions can be trained simultaneously to obtain the same training model. In addition, as both ends of the reversible neural network can be used as input and output, the mapping formed by the input and output at both ends can be simultaneously the optimal solution, and the problem of non-uniqueness of the solution in the conventional neural network can be solved.
Fig. 2 shows the structure of the reversible neural network provided by the present embodiment, which has one Forward-propagating (Forward-propagating) path and one Backward-propagating (Backward-propagating) path. In the forward propagation path, input data (input) is input from the left side, sequentially passes through each hidden layer of the forward propagation path, and finally output data (output) is obtained. The output data (output) of the forward propagation path may be input into the reverse propagation path from the right side as the input data (input) of the reverse propagation path, sequentially pass through each hidden layer of the reverse propagation path, and finally obtain the output data (output).
All layers in the middle of the reversible neural network are called as hidden layers, the hidden layers can convert input data features into another latitude space to serve as more abstract characterization, the features can be better linearized, and the hidden layers do not directly receive external signals or directly send signals to the outside.
In this embodiment, the characteristic that the reversible neural network can autonomously plan its own learning and task is utilized, that is, the temperature in the semiconductor measurement machine reaches a virtuous circle through the forward and reverse propagation paths of the reversible neural network, and the corresponding optimal response is continuously calculated according to the change of the external environment of the semiconductor measurement machine.
Because the reversible neural network almost reacts in real time and the calculation of the reaction almost does not need time, the reversible neural network can be very sensitive and can adjust the temperature in the semiconductor measuring machine in real time with high precision.
Specifically, the current environmental information of the semiconductor measurement machine may be input at the left end of the Forward-locating path, and the right end outputs the power that the power supply needs to provide when the temperature of the semiconductor measurement machine is maintained at the preset target temperature in the machine, i.e. the power supply power. The right end of the backsard-positioning path inputs power supply power, and the left end outputs the temperature in the semiconductor measuring machine in the next second under the current power supply power, namely the temperature in the semiconductor measuring machine is regulated and controlled.
When the external environment of the semiconductor measuring machine is changed, the reversible neural network obtains the optimal reaction for keeping the internal temperature of the semiconductor measuring machine stable according to the acquired current environment information and by means of deep learning training (for example, given parameters of the neural network and the neural network, including learning rate, training times and training batch number, and in each training time, the weight value of the neural network is continuously adjusted by one forward transmission and one backward transmission), so that the internal temperature of the semiconductor measuring machine is kept near the preset target internal temperature. When the input of the forward propagation path deviates from the preset target in-machine temperature, the reverse neural path can calculate the in-machine temperature at the next moment, and if the calculated in-machine temperature at the next moment deviates from the preset target in-machine temperature, the reversible neural network can adjust the power of the power supply in advance. For example, when the external temperature of the semiconductor measurement machine is increased, the reversible neural network controls the power supply to reduce the power so as to reduce the production of heat and keep the internal temperature of the semiconductor measurement machine stable; when the external temperature of the semiconductor measuring machine is reduced, the reversible neural network can control the power supply to increase the power so as to improve the production of heat and keep the internal temperature of the semiconductor measuring machine stable.
It should be understood that, in the reversible neural network, since the forward propagation path and the backward propagation path are integrated, the same training model is obtained in the same training, so that the training result is relatively accurate, and the problem of multiple solutions is avoided. Therefore, the temperature inside the semiconductor measuring machine is controlled through the forward propagation path and the backward propagation path of the reversible neural network, and the temperature regulation and control effect is more accurate.
In some embodiments, the forward propagation path and the backward propagation path each include a generative countermeasure network, the generative countermeasure network including a generator and a discriminator, and the weight value of the reversible neural network is adjusted by the generator and the discriminator.
The generator and the discriminator are constantly opposed to each other, so that the parameters of the reversible neural network can be continuously adjusted, namely the weight value of the reversible neural network can be adjusted, and finally the discriminator can not judge whether the output result of the generator is real or not, so that the reversibility of the reversible neural network can be realized by spurious.
In some embodiments, the neural network is a bidirectional neural network, the bidirectional neural network including a forward propagation network and a backward propagation network, the forward propagation network, the backward propagation network, and a dynamic network, the forward propagation network and the backward propagation network being coupled together by the dynamic network.
The structure of the bidirectional neural network is shown in fig. 3, and includes a forward propagation network, a backward propagation network and a dynamic network, where the forward propagation network and the backward propagation network have opposite directions and each include at least one node, and the node may be a convolution bidirectional node (convolutional invertible block) or a fully connected bidirectional node (fully connected invertible block). In fig. 3, x represents an input, and z represents an output; at the same time z can also be input and x can also be output.
The front and back propagation networks are independent conventional neural networks, each node of the front and back propagation networks is coupled together through a Dynamic network (Dynamic network), each node of the front and back propagation networks can be programmed by using the Dynamic network to perform certain customized tasks through information of the node, the tasks include, but are not limited to, size cutting, size changing, data adjustment, data normalization and the like, and basically any task can be realized through one Dynamic network. Therefore, by using the dynamic network, in the same training, the front training model and the back training model are obtained, but the front training model and the back training model have coupling.
In some embodiments, the forward propagation network of the bi-directional neural network is the same as the forward propagation path of the reversible neural network, which includes a generative antagonizing network and a common neural network, which is different from the reverse propagation path of the reversible neural network. The propagation direction of the normal neural network is from left to right, and the propagation direction of the reverse propagation path of the reversible neural network is from right to left.
The front and back propagation networks are coupled together through the dynamic network, so that the front and back models can be obtained in one training process, but the two models have a coupling relationship.
The coupling relation is that the objective functions of the forward and reverse propagation networks are added together to form a large objective function, and a global optimum is obtained in iteration at the same time.
Specifically, the training model can be obtained by training through setting parameters such as learning rate, learning batch number, training times and the like.
The forward propagation network is consistent with the forward propagation path of the reversible neural network, and the backward propagation network can replace the backward propagation path of the reversible neural network by a generated countermeasure network GAN plus a conventional neural network.
Wherein x can be the current in-machine temperature and the current environment information of the semiconductor measuring machine, and z can be the power supply.
It should be understood that by training the forward and reverse neural networks, the power of the power supply and the internal temperature of the semiconductor measurement machine can be continuously predicted to adjust the internal temperature of the semiconductor measurement machine.
Fig. 4 shows a temperature control device M400 according to an embodiment of the present application, corresponding to the temperature control method shown in fig. 1, including:
the current information acquisition module M410 is configured to acquire current internal temperature and current environmental information of a semiconductor measurement apparatus, where the current environmental information is used to indicate an environment where the semiconductor measurement apparatus is currently located;
the forward input module M420 is configured to input the current in-machine temperature and the current environmental information into a forward propagation path or a forward propagation network of a neural network, so as to obtain power output by the neural network, where the power is required to be provided by a power supply when the in-machine temperature of the semiconductor measurement machine is maintained at a preset in-machine target temperature;
a first reverse output module M430, configured to, if the power supply power is obtained through the forward propagation path, input the power supply power into a reverse propagation path of the neural network, and obtain an in-machine regulation temperature output by the reverse propagation path and corresponding to the power supply power, where a temperature difference between the in-machine regulation temperature and the preset in-machine target temperature is less than or equal to a preset temperature difference;
And the second back output module M440 is configured to input the power supply power into a back propagation network of the neural network if the power supply power is obtained through the forward propagation network, so as to obtain an in-machine regulation temperature corresponding to the power supply power, which is output by the back propagation network.
Optionally, the input terminal of the neural network includes at least two neurons, through which the environmental information can be input into the neural network as an influencing factor.
Optionally, the neural network is a reversible neural network, and the reversible neural network includes a forward propagation path and a reverse propagation path.
Optionally, the forward propagation path and the backward propagation path each include a generation type countermeasure network, the generation type countermeasure network includes a generator and a discriminator, and the weight value of the reversible neural network is adjusted by the generator and the discriminator.
Optionally, the neural network is a bidirectional neural network, and the bidirectional neural network includes a forward propagation network, a backward propagation network, and a dynamic network, where the forward propagation network and the backward propagation network are coupled together through the dynamic network.
Optionally, the forward propagation network of the bidirectional neural network is the same as the forward propagation path of the reversible neural network, and the backward propagation network of the bidirectional neural network includes a generative countermeasure network and a normal neural network, and the normal neural network is different from the backward propagation path of the reversible neural network.
Optionally, the current environmental information includes at least one of a temperature of an environment in which the semiconductor measurement machine is located, a humidity of the environment in which the semiconductor measurement machine is located, a hot air port temperature of the semiconductor measurement machine, and a butterfly valve port temperature of the semiconductor measurement machine.
It will be appreciated that various implementations and combinations of implementations and advantageous effects thereof in the above embodiments are equally applicable to this embodiment, and will not be described here again.
Fig. 5 is a schematic structural diagram of an electronic device according to an embodiment of the present application. As shown in fig. 5, the electronic device D10 of this embodiment includes: at least one processor D100 (only one is shown in fig. 5), a memory D101 and a computer program D102 stored in the memory D101 and executable on the at least one processor D100, the processor D100 implementing the steps in any of the various method embodiments described above when executing the computer program D102. Alternatively, the processor D100 may perform the functions of the modules/units of the apparatus embodiments described above, such as the functions of the modules M410 to M440 shown in fig. 4, when executing the computer program D102.
In some embodiments, the processor D100, when executing the computer program D102, implements the following steps:
acquiring current internal temperature and current environment information of a semiconductor measurement machine, wherein the current environment information is used for indicating the current environment of the semiconductor measurement machine;
inputting the current in-machine temperature and the current environment information into a forward propagation path or a forward propagation network of a neural network to obtain power output by the neural network, wherein the power is required to be provided by a power supply when the in-machine temperature of the semiconductor measuring machine is maintained at a preset in-machine target temperature;
if the power supply power is obtained through the forward propagation path, inputting the power supply power into a reverse propagation path of the neural network to obtain an internal regulation temperature which is output by the reverse propagation path and corresponds to the power supply power, wherein the temperature difference between the internal regulation temperature and the preset internal target temperature is smaller than or equal to a preset temperature difference;
and if the power supply power is obtained through the forward propagation network, inputting the power supply power into a backward propagation network of the neural network to obtain the built-in regulation temperature corresponding to the power supply power, which is output by the backward propagation network.
In some embodiments, the processor D100, when executing the computer program D102, the input of the neural network comprises at least two neurons through which the environmental information can be input into the neural network as an influencing factor.
In some embodiments, the neural network is a reversible neural network that includes a forward propagation path and a reverse propagation path when the processor D100 executes the computer program D102.
In some embodiments, when the processor D100 executes the computer program D102, a generative countermeasure network is included in both the forward propagation path and the backward propagation path, the generative countermeasure network including a generator and a arbiter by which weight values of the reversible neural network are adjusted.
In some embodiments, when the processor D100 executes the computer program D102, the neural network is a bidirectional neural network comprising a forward propagation network, a backward propagation network, and a dynamic network through which the forward propagation network and the backward propagation network are coupled together.
In some embodiments, when the processor D100 executes the computer program D102, the forward propagation network of the bi-directional neural network is the same as the forward propagation path of the reversible neural network, and the backward propagation network of the bi-directional neural network includes a generative countermeasure network and a normal neural network, the normal neural network being different from the backward propagation path of the reversible neural network.
In some embodiments, the current environmental information includes at least one of a temperature of an environment in which the semiconductor metrology tool is located, a humidity of the environment in which the semiconductor metrology tool is located, a hot air port temperature of the semiconductor metrology tool, and a butterfly valve port temperature of the semiconductor metrology tool when the processor D100 executes the computer program D102.
The electronic device D10 may be a computing device such as a desktop computer, a notebook computer, a palm computer, a cloud server, etc. The electronic device may include, but is not limited to, a processor D100, a memory D101. It will be appreciated by those skilled in the art that fig. 5 is merely an example of the electronic device D10 and is not meant to be limiting of the electronic device D10, and may include more or fewer components than shown, or may combine certain components, or different components, such as may also include input-output devices, network access devices, etc.
The processor D100 may be a central processing unit (Central Processing Unit, CPU), the processor D100 may also be other general purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), off-the-shelf programmable gate arrays (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory D101 may in some embodiments be an internal storage unit of the electronic device D10, such as a hard disk or a memory of the electronic device D10. The memory D101 may also be an external storage device of the electronic device D10 in other embodiments, for example, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash Card (Flash Card) or the like, which are provided on the electronic device D10. Further, the memory D101 may also include both an internal storage unit and an external storage device of the electronic device D10. The memory D101 is used for storing an operating system, an application program, a boot loader (BootLoader), data, other programs, etc., such as program codes of the computer program. The memory D101 may also be used to temporarily store data that has been output or is to be output.
It should be noted that, because the content of information interaction and execution process between the above devices/units is based on the same concept as the method embodiment of the present application, specific functions and technical effects thereof may be referred to in the method embodiment section, and will not be described herein again.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-described division of the functional units and modules is illustrated, and in practical application, the above-described functional distribution may be performed by different functional units and modules according to needs, i.e. the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-described functions. The functional units and modules in the embodiment may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit, where the integrated units may be implemented in a form of hardware or a form of a software functional unit. In addition, specific names of the functional units and modules are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present application. The specific working process of the units and modules in the above system may refer to the corresponding process in the foregoing method embodiment, which is not described herein again.
Embodiments of the present application also provide a computer readable storage medium storing a computer program, which when executed by a processor, may implement the steps in the above-described method embodiments.
Embodiments of the present application provide a computer program product which, when run on an electronic device, causes the electronic device to perform the steps of the method embodiments described above.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the present application implements all or part of the flow of the method of the above embodiments, and may be implemented by a computer program to instruct related hardware, where the computer program may be stored in a computer readable storage medium, where the computer program, when executed by a processor, may implement the steps of each of the method embodiments described above. Wherein the computer program comprises computer program code which may be in source code form, object code form, executable file or some intermediate form etc. The computer readable medium may include at least: any entity or device capable of carrying computer program code to a photographing device/terminal apparatus, recording medium, computer Memory, read-Only Memory (ROM), random access Memory (Random Access Memory, RAM), electrical carrier signals, telecommunications signals, and software distribution media. Such as a U-disk, removable hard disk, magnetic or optical disk, etc. In some jurisdictions, computer readable media may not be electrical carrier signals and telecommunications signals in accordance with legislation and patent practice.
In the foregoing embodiments, the descriptions of the embodiments are emphasized, and in part, not described or illustrated in any particular embodiment, reference is made to the related descriptions of other embodiments.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus/electronic device and method may be implemented in other manners. For example, the apparatus/electronic device embodiments described above are merely illustrative, e.g., the division of the modules or units is merely a logical function division, and there may be additional divisions in actual implementation, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection via interfaces, devices or units, which may be in electrical, mechanical or other forms.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
The above embodiments are only for illustrating the technical solution of the present application, and are not limiting; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present application, and are intended to be included in the scope of the present application.
Claims (10)
1. A method for controlling a temperature, the method comprising:
acquiring current internal temperature and current environment information of a semiconductor measurement machine, wherein the current environment information is used for indicating the current environment of the semiconductor measurement machine;
Inputting the current in-machine temperature and the current environment information into a forward propagation path or a forward propagation network of a neural network to obtain power output by the neural network, wherein the power is required to be provided by a power supply when the in-machine temperature of the semiconductor measuring machine is maintained at a preset in-machine target temperature;
if the power supply power is obtained through the forward propagation path, inputting the power supply power into a reverse propagation path of the neural network to obtain an internal regulation temperature which is output by the reverse propagation path and corresponds to the power supply power, wherein the temperature difference between the internal regulation temperature and the preset internal target temperature is smaller than or equal to a preset temperature difference;
and if the power supply power is obtained through the forward propagation network, inputting the power supply power into a backward propagation network of the neural network to obtain the built-in regulation temperature corresponding to the power supply power, which is output by the backward propagation network.
2. The method of controlling temperature according to claim 1, wherein the input terminal of the neural network includes at least two neurons through which the environmental information can be input into the neural network as an influence factor.
3. The temperature control method according to claim 1 or 2, wherein the neural network is a reversible neural network including a forward propagation path and a reverse propagation path.
4. The method of controlling temperature according to claim 3, wherein the forward propagation path and the backward propagation path each include a generation type countermeasure network, the generation type countermeasure network including a generator and a discriminator, and the weight value of the reversible neural network is adjusted by the generator and the discriminator.
5. The method of controlling temperature according to claim 1 or 2, wherein the neural network is a bidirectional neural network including a forward propagation network, a backward propagation network, and a dynamic network, the forward propagation network and the backward propagation network being coupled together through the dynamic network.
6. The method of controlling temperature according to claim 5, wherein a forward propagation network of the bidirectional neural network is identical to a forward propagation path of the reversible neural network, and a backward propagation network of the bidirectional neural network includes a generation type countermeasure network and a normal neural network, the normal neural network being different from the backward propagation path of the reversible neural network.
7. The method of claim 2, wherein the current environmental information includes at least one of a temperature of an environment in which the semiconductor measurement tool is located, a humidity of the environment in which the semiconductor measurement tool is located, a hot air outlet temperature of the semiconductor measurement tool, and a butterfly air outlet temperature of the semiconductor measurement tool.
8. A temperature control device, characterized in that the temperature control device comprises:
the current information acquisition module is used for acquiring the current internal temperature and the current environment information of the semiconductor measurement machine, wherein the current environment information is used for indicating the current environment of the semiconductor measurement machine;
the forward input module is used for inputting the current internal temperature and the current environmental information into a forward propagation path or a forward propagation network of a neural network to obtain power output by the neural network, wherein the power is required to be provided by a power supply when the internal temperature of the semiconductor measuring machine is maintained at a preset internal target temperature;
the first reverse output module is used for inputting the power supply power into a reverse propagation path of the neural network if the power supply power is obtained through the forward propagation path, so as to obtain an internal regulation temperature which is output by the reverse propagation path and corresponds to the power supply power, wherein the temperature difference between the internal regulation temperature and the preset internal target temperature is smaller than or equal to a preset temperature difference;
And the second reverse output module is used for inputting the power supply power into a reverse propagation network of the neural network if the power supply power is obtained through the forward propagation network, so as to obtain the built-in regulation temperature which is output by the reverse propagation network and corresponds to the power supply power.
9. An electronic device comprising a memory, a processor and a computer program stored in the memory and capable of running on the processor, characterized in that the processor implements the temperature control method according to any one of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium storing a computer program, characterized in that the computer program, when executed by a processor, implements the temperature control method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310193486.4A CN116560434A (en) | 2023-02-21 | 2023-02-21 | Temperature control method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310193486.4A CN116560434A (en) | 2023-02-21 | 2023-02-21 | Temperature control method and device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116560434A true CN116560434A (en) | 2023-08-08 |
Family
ID=87492211
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310193486.4A Pending CN116560434A (en) | 2023-02-21 | 2023-02-21 | Temperature control method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116560434A (en) |
-
2023
- 2023-02-21 CN CN202310193486.4A patent/CN116560434A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20170060104A1 (en) | Numerical controller with machining condition adjustment function which reduces chatter or tool wear/breakage occurrence | |
| US20220326748A1 (en) | Advanced thermal control for ssd | |
| US11604472B2 (en) | Method for controlling vehicle speed for autonomous driving, electronic device, and computer-readable storage medium | |
| CN105528000B (en) | A kind of intelligent temperature control table for aircraft | |
| CN106292550B (en) | The Servocontrol device of function with vehicle air-conditioning gain | |
| CN112947088B (en) | Modeling and control method of temperature and humidity system based on closed space | |
| CN102509152A (en) | Switched reluctance motor on-line modeling method based RBF neural network | |
| CN106773654A (en) | A kind of Fractional Order PID Controller parameter optimization setting method | |
| CN110703693A (en) | Iterative learning feedforward control method and system for machine tool feeding system | |
| KR102382047B1 (en) | Automatic learning tuning system of motor controller using PSO | |
| CN110094838A (en) | A kind of variable element MFA control method based on air-conditioning system | |
| CN102156407A (en) | Predictive control method and device for ARM (advanced RISC machines)-based embedded type model | |
| CN116560434A (en) | Temperature control method and device, electronic equipment and storage medium | |
| CN1573624B (en) | Apparatus and method for batch property estimation | |
| CN111624978B (en) | Method and device for evaluating performance of water supply flow controller | |
| US8190536B2 (en) | Method of performing parallel search optimization | |
| CN116954334A (en) | Heat dissipation control method, device, server, computer equipment and storage medium | |
| CN104632416A (en) | Control method for rotating speed of gas turbine | |
| CN105122158B (en) | Programmable control unit with short delaing time | |
| CN114411165B (en) | PEM water electrolysis hydrogen production water body temperature control method, system, equipment and medium | |
| CN212302318U (en) | CT data acquisition system constant temperature control device | |
| CN109343680A (en) | A kind of fan regulation and controlling method and system based on machine learning algorithm | |
| CN111722636A (en) | Self-balancing vehicle design scheme based on multi-innovation Kalman filtering | |
| CN110134095A (en) | The method and terminal device of Power Plant Thermal analog control system optimization | |
| CN103558762A (en) | Method for implementing immune genetic PID controller based on graphic configuration technology |
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 |