A method and system for obtaining and annotating blackbody temperature data sets

Abstract

The present invention discloses a method and system for obtaining and annotating a blackbody temperature data set, which relates to the technical field of blackbody temperature control systems, and includes building a blackbody temperature acquisition system and a blackbody temperature simulation system. The blackbody temperature simulation system is used to generate a simulated data set to expand the real data set collected by the blackbody temperature acquisition system; perform a single open-loop control on the blackbody temperature acquisition system with a fixed duty cycle to obtain the real data set; to obtain the waste heat data set, under the condition of a fixed duty cycle, perform continuous open-loop control on the blackbody temperature acquisition system; calculate the rate of natural cooling; annotate the collected data set, and distinguish the heating, waste heat, and cooling data sets according to the rate of natural cooling. The present invention can not only comprehensively collect the blackbody temperature change data, but also effectively distinguish the data in the heating, waste heat, and cooling processes, thereby providing support for the learning of the blackbody thermodynamic characteristics by the reinforcement learning algorithm.

Description

Technical Field

[0001] The present invention relates to the technical field of blackbody temperature control systems, and more specifically, to a method and system for obtaining and annotating blackbody temperature data sets. Background Art

[0002] Currently, with the continuous development of temperature control technology, temperature regulation algorithms based on deep learning have become a research hotspot and significant progress has been made. However, current deep learning algorithms still rely on traditional PID control algorithms and have not achieved completely parameterless temperature control. The main reason for this bottleneck is the lack of a scientific method for obtaining and annotating temperature data sets. In the prior art, no mature method for obtaining and annotating temperature data sets has been formed, and this gap has led to application difficulties for parameterless temperature control algorithms based on reinforcement learning.

[0003] Therefore, how to efficiently, comprehensively, and accurately obtain temperature data sets and distinguish temperature data during heating, afterheat, and cooling processes is an urgent problem to be solved by those skilled in the art. Summary of the Invention

[0004] In view of this, taking the controlled object - blackbody as an example, the present invention proposes a method and system for obtaining and annotating blackbody temperature data sets, aiming to solve the problems in the background art and provide reliable data support for parameterless temperature control algorithms based on reinforcement learning.

[0005] To achieve the above object, the present invention adopts the following technical solutions:

[0006] A method for obtaining and annotating blackbody temperature data sets, comprising:

[0007] S1. Build a blackbody temperature acquisition system and a blackbody temperature simulation system. The blackbody temperature simulation system is used to generate a simulated data set to expand the real data set collected by the blackbody temperature acquisition system;

[0008] S2. Perform a single open-loop control on the blackbody temperature acquisition system with a fixed duty cycle to obtain a real data set;

[0009] S3. To obtain an afterheat data set, perform continuous open-loop control on the blackbody temperature acquisition system under a fixed duty cycle;

[0010] S4. Calculate the natural cooling rate in S2;

[0011] S5. Annotate the data sets collected in S2 and S3, and distinguish the heating, afterheat, and cooling data sets according to the natural cooling rate.

[0012] Optionally, the S1 specifically includes the following processes:

[0013] S1.1. Build the hardware system for collecting the blackbody temperature dataset;

[0014] S1.2. Using the step response method, apply an excitation signal with a duty cycle of 50% to the blackbody, conduct a heating experiment on the blackbody system, and heat it in open loop until the temperature remains unchanged;

[0015] S1.3. According to the obtained actual system response curve of the system, perform low-pass filtering on the temperature data;

[0016] S1.4. Solve the system characteristics from the actual system response curve of the system.

[0017] Optionally, the specific process of S2 includes the following:

[0018] Perform single open-loop control for duty cycles of 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, and 100% respectively, and record the temperature data every 0.5 s; under the condition of a fixed duty cycle, collect the temperature data from room temperature to the stable temperature. After the temperature remains stable, continue to collect for 60 s, and then collect the temperature data during the cooling process under room temperature conditions until the temperature drops to room temperature and continue to collect for 60 s.

[0019] Optionally, the specific process of S3 includes the following:

[0020] S3.1. Record the temperature data every 0.5 s. Under the condition of a fixed duty cycle, start heating from room temperature for 60 s, and then stop inputting the heating signal for 60 s;

[0021] S3.2. Repeat the above process until the blackbody temperature remains unchanged;

[0022] S3.3. Continue heating for 60 s after the temperature remains unchanged;

[0023] S3.4. Disconnect the input of the heating signal and collect the temperature data of the system when it drops to room temperature;

[0024] S3.5. Continue to collect for 60 s after dropping to room temperature;

[0025] S3.6. Perform the above operations for duty cycles of 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, and 100% respectively.

[0026] Optionally, the specific process of S4 includes the following:

[0027] S4.1. Select the time point when the temperature drops to room temperature in S2 and the data of the previous 120 s;

[0028] S4.2. Perform low-pass filtering on the obtained data;

[0029] S4.3. Calculate the natural cooling rate for the temperature drop data within these 120 s.

[0030] Optionally, the S5 specifically includes the following process:

[0031] S5.1. First, label the heating data set. When collecting temperature data, the heating signals are labeled, and the part with heating signal output is the heating data set.

[0032] S5.2. For the remaining data set, label the waste heat characteristic data set by comparing the relationship between the waste heat power and the cooling power.

[0033] S5.3. Excluding the heating data set and the waste heat data set, the other part is the cooling data set.

[0034] Optionally, the steps of the S5.2 are specifically as follows:

[0035] The waste heat characteristic data set is divided into two parts: 1. The waste heat power is greater than the cooling power, which is manifested as the blackbody temperature continuing to rise for a period of time after the heating signal stops; 2. The waste heat power is less than the cooling power, which is manifested as the blackbody starting to cool down after continuing to heat up for a period of time after the heating stops, but the cooling rate is less than the natural cooling rate.

[0036] A blackbody temperature data set acquisition and annotation system includes:

[0037] A system construction module that constructs a blackbody temperature acquisition system and a blackbody temperature simulation system. The blackbody temperature simulation system is used to generate a simulated data set to expand the real data set collected by the blackbody temperature acquisition system.

[0038] A single - time open - loop control module that performs single - time open - loop control on the blackbody temperature acquisition system with a fixed duty cycle to obtain the real data set.

[0039] A continuous open - loop control module that, in order to obtain the waste heat data set, performs continuous open - loop control on the blackbody temperature acquisition system under a fixed duty cycle.

[0040] A rate calculation module that calculates the natural cooling rate in the single - time open - loop control module.

[0041] A data set annotation module that annotates the data sets collected in the single - time open - loop control module and the continuous open - loop control module, and distinguishes the heating, waste heat, and cooling data sets according to the natural cooling rate.

[0042] As can be seen from the above technical solutions, compared with the prior art, the present invention discloses a method and system for obtaining and annotating a blackbody temperature data set, which can accurately collect the temperature data of the blackbody under different duty cycles, heating and afterheat conditions, and annotate the data set for training and optimizing the blackbody temperature control algorithm based on reinforcement learning. It can not only comprehensively collect the blackbody temperature change data, but also effectively distinguish the data in the heating, afterheat and cooling processes, thus providing support for the learning of the blackbody thermodynamics characteristics by the reinforcement learning algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only the embodiments of the present invention, and for those of ordinary skill in the art, other drawings can be obtained according to the provided drawings without creative efforts.

[0044] Figure 1 It is a flowchart of the method provided by the present invention;

[0045] Figure 2a It is a real system diagram of the surface source blackbody provided by the present invention, Figure 2b It is a simplified temperature control model of the large surface source blackbody provided by the present invention;

[0046] Figure 3 It is a diagram for obtaining the system characteristic parameters provided by the present invention;

[0047] Figure 4 It is a schematic diagram of the data set annotation process provided by the present invention;

[0048] Figure 5 It is an analysis diagram of the afterheat and heat dissipation characteristics provided by the present invention;

[0049] In the figure, ① - conical blackbody source; ② - temperature equalizing plate; ③ - heating device; ④ - ceramic terminal; ⑤ - electric heating wire installation groove; ⑥ - PT1000; ⑦ - PT100 installation hole. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0050] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0051] The embodiments of the present invention disclose a method for obtaining and annotating a blackbody temperature data set, as Figure 1 shown, including:

[0052] S1. Build a blackbody temperature acquisition system and a blackbody temperature simulation system. The blackbody temperature simulation system is used to generate a simulated data set to expand the real data set collected by the blackbody temperature acquisition system.

[0053] S2. Perform a single open-loop control on the blackbody temperature acquisition system with a fixed duty cycle to obtain a real data set.

[0054] S3. To obtain a waste heat data set, perform continuous open-loop control on the blackbody temperature acquisition system under a fixed duty cycle.

[0055] S4. Calculate the natural cooling rate in S2.

[0056] S5. Label the data sets collected in S2 and S3, and distinguish the heating, waste heat, and cooling data sets according to the natural cooling rate.

[0057] In a specific embodiment, S1 specifically includes the following processes:

[0058] S1.1. Build a hardware system for collecting the blackbody temperature data set.

[0059] S1.2. Use the step response method to give the blackbody an excitation signal with a duty cycle of 50%, conduct a temperature rise experiment on the blackbody system, and perform open-loop temperature rise until the temperature remains unchanged.

[0060] S1.3. According to the obtained real system response curve of the system, perform low-pass filtering on the temperature data.

[0061] S1.4. Solve the system characteristics from the real system response curve of the system.

[0062] The derivation process of the system characteristics for S1.4 is as follows:

[0063] The real system model of the large area source blackbody is as Figure 2a shown. The composition of each small area source blackbody is as shown in B. The conical blackbody source ① is beneficial to improving the surface emissivity of the blackbody. The surface emissivity of the large area source blackbody designed in the laboratory can reach 99.99%. The area source blackbody is installed on the heat dissipation plate ② to more evenly transfer the heat generated by the heating device ③. The heating wire is installed on the heating wire installation groove ⑤ using the ceramic terminal ④. The PT1000 ⑥ is installed in the PT100 installation hole ⑦, and the temperature of the back surface of the blackbody is fed back to the upper computer C in real time and displayed using RS485 communication. The upper computer integrates functions such as temperature control, water-cooling and air-cooling control, system self-check, and PID self-tuning. There are water-cooling and air-cooling modules installed on the back of the large area source blackbody surface to dissipate heat from the area source blackbody and accelerate the cooling performance of the area source blackbody.

[0064] From the above real system model of the large area source blackbody, after simplifying it, we get asFigure 2b The blackbody temperature control model shown. Where P N is the heating power, P R is the actual power, P s is the heat dissipation power, K is the heat dissipation coefficient, ΔT is the value of the temperature rise after heating, T E is the ambient temperature, T is the temperature of the blackbody after heating, T B is the current temperature of the blackbody, Q n is the moment t n The heat generated.

[0065] P is the total power of the heating wire, N is the duty cycle ratio, then the heating power P N is:

[0066] P N = PN

[0067] For simplicity, the model assumes that the heat dissipation power is linearly related to the difference between the blackbody temperature and the ambient temperature, i.e.:

[0068] P S = K(T - T E ) = KΔT

[0069] The actual power P R is the difference between the heating power of the heater and the heat dissipation power:

[0070] P R = P N - P S

[0071] In the time t, the total heat Q received by the blackbody is:

[0072]

[0073] C is the specific heat capacity of the blackbody material, m is the mass of the blackbody, and the value of the blackbody temperature rise ΔT is:

[0074]

[0075] Assuming that the blackbody temperature is the same as the ambient temperature before heating, then the temperature of the blackbody after heating is:

[0076] T = T E + ΔT

[0077] The relationship between the moment t and the temperature T is calculated as shown in Table 1.

[0078] Table 1 Relationship between moment and temperature

[0079]

[0080] ΔT -1Denote the interval between the temperature change Δt before heating and the moment. For simplicity of the model, set Δt to 1:

[0081]

[0082] Using mathematical induction, we get:

[0083]

[0084] Denote as the heating coefficient, and n as the time coefficient. Then the temperature rise value ΔT

[0085] ΔT n = α[1 - (1 - β) n+1 .

[0086] It can be deduced from the above model that as long as the heating coefficient α and the time coefficient β of a heating system are obtained, the temperature rise characteristic curve of the system can be obtained.

[0087] Using the step response method, an excitation signal with a duty cycle of 50% is given to the controlled object, and a temperature rise experiment is carried out on the temperature control system. When the temperature rises in an open loop until it remains unchanged, the obtained true system response curve of the system is as Figure 3 shown. The ordinate of this curve is the main coordinate axis. The system characteristics are solved from this figure for subsequent algorithm simulation and acquisition of the simulation data set. First, the temperature rise rate of each point is obtained from the response curve. As shown by the yellow curve, the ordinate of this curve is on the secondary coordinate axis. At the highest point C of the temperature rise rate, the pure lag time τ and the time coefficient β are obtained by using the tangent method of the response curve: draw a straight line with the starting temperature value and the stable temperature value. Denote the intersection point of the response curve and the initial temperature as A, the intersection point of the tangent line and the initial temperature as B, and the intersection point with the steady-state temperature as D. The difference between the abscissas of point A and point B is the pure lag time τ, and the difference between the abscissas of point A and point D is the time coefficient β. The output N is the difference between the initial temperature and the steady-state temperature. The heating coefficient α is the ratio of the output N to the input M.

[0088] The open-loop temperature response curve of the black body was collected using the hardware system. The system characteristics of the black body temperature control system were obtained according to S1.4 as Figure 3 shown. By solving, the time coefficient β of the black body temperature control is 118, the heating coefficient α is 1.7884, and the pure lag time τ is 7.5. Based on these two system characteristic values, Figure 3The simulation curve in has a correlation coefficient of 0.9985 with the actual temperature curve, which can also prove the accuracy of the physical model established above. Since it takes too long to obtain the real temperature dataset, the above blackbody model can be used to generate a simulated dataset to supplement the blackbody heating response curves under various different duty cycles to expand the number of datasets.

[0089] In a specific embodiment, S2 specifically includes the following process:

[0090] Perform single open-loop control on duty cycles of 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, and 100% respectively, and record the temperature data every 0.5 s; under the condition of a fixed duty cycle, collect the temperature data from room temperature to stable temperature. After the temperature remains stable, continue to collect the cooling data after 60 s and then collect the room temperature data until the temperature drops to room temperature and continue to collect for 60 s.

[0091] In a specific embodiment, S3 specifically includes the following process:

[0092] S3.1: Record the temperature data every 0.5 s. Under the condition of a fixed duty cycle, start heating from room temperature for 60 s and then stop inputting the heating signal for 60 s;

[0093] S3.2: Repeat the above process until the blackbody temperature remains unchanged;

[0094] S3.3: Continue heating for 60 s after the temperature remains unchanged;

[0095] S3.4: Disconnect the heating signal input and collect the temperature data when the system drops to room temperature;

[0096] S3.5: Continue to collect for 60 s after dropping to room temperature;

[0097] S3.6: Perform the above operations on duty cycles of 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, and 100% respectively.

[0098] In a specific embodiment, S4 specifically includes the following process:

[0099] S4.1: Take the time point when the temperature drops to room temperature in S2 and the data of the previous 120 s;

[0100] S4.2: Perform low-pass filtering on the obtained data;

[0101] S4.3: Calculate the natural cooling rate for the cooling data within these 120 s:

[0102]

[0103] In a specific embodiment, as Figure 4 shown, S5 specifically includes the following processes:

[0104] S5.1. First, label the heating-up data set. When collecting temperature data, the heating signal is labeled. The part with the heating signal output is the heating data set.

[0105] S5.2. For the remaining data set, label the waste heat characteristic data set by comparing the relationship between the waste heat power and the cooling power.

[0106] S5.3. Excluding the heating data set and the waste heat data set, the other part is the cooling data set.

[0107] The steps of S5.2 are specifically as follows:

[0108] The waste heat characteristic data set is divided into two parts: as Figure 5 shown, 1. The waste heat power is greater than the cooling power, which is manifested as the blackbody temperature continuing to rise for a period of time after the heating signal stops; 2. The waste heat power is less than the cooling power, which is manifested as the blackbody starting to cool down after continuing to heat up for a period of time after the heating stops, but the cooling rate is less than the natural cooling rate.

[0109] After the heating stops for the first part, the blackbody temperature will still continue to rise for a period of time. The temperature change rate at this time can be expressed as:

[0110]

[0111] where P 余热 is the waste heat power, T 环境 is the ambient temperature, k is the heat dissipation coefficient, and C is the specific heat capacity of the blackbody material;

[0112] At this time the temperature rises;

[0113] For the second part where the waste heat power is less than the cooling power, it is manifested as the blackbody temperature starting to drop after continuing to rise for a period of time after the heating stops, but the drop rate is less than the natural cooling rate. At this time the temperature starts to drop. Compare the at this time with ΔV 降温 . If then it still belongs to the waste heat characteristic data set until the waste heat characteristic data set ends, and a column of eigenvalue 0.5 is added to the column after the duty cycle to label the waste heat characteristic data set.

[0114] Data and annotation diagrams of three characteristics are shown in Tables 2, 3, and 4. The first column in the three diagrams is temperature data; the second column is the PWM duty cycle of the current temperature, where 0 represents no PWM input; the third column is the annotation of the waste heat characteristic.

[0115] Table 2 Annotation Table of Heating Characteristic Data Set

[0116]

[0117] Table 3 Annotation Table of Waste Heat Characteristic Data Set

[0118]

[0119] Table 4 Annotation Table of Cooling Characteristic Data Set

[0120]

[0121] A blackbody temperature data set acquisition and annotation system includes:

[0122] A system construction module that constructs a blackbody temperature acquisition system and a blackbody temperature simulation system. The blackbody temperature simulation system is used to generate a simulated data set to expand the real data set collected by the blackbody temperature acquisition system.

[0123] A single - time open - loop control module that performs single - time open - loop control on the blackbody temperature acquisition system with a fixed duty cycle to obtain a real data set.

[0124] A continuous open - loop control module that, to obtain a waste heat data set, performs continuous open - loop control on the blackbody temperature acquisition system under a fixed duty cycle.

[0125] A rate calculation module that calculates the rate of natural temperature drop in the single - time open - loop control module.

[0126] A data set annotation module that annotates the data sets collected in the single - time open - loop control module and the continuous open - loop control module, and distinguishes heating, waste heat, and cooling data sets according to the rate of natural temperature drop.

[0127] In this specification, each embodiment is described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The same or similar parts among the embodiments can be referred to each other. For the devices disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple, and the relevant parts can be referred to the description of the method part.

[0128] The foregoing description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for obtaining and annotating a blackbody temperature data set, characterized in that Including: S1. Build a blackbody temperature acquisition system and a blackbody temperature simulation system. The blackbody temperature simulation system is used to generate a simulated data set to augment the real data set collected by the blackbody temperature acquisition system; S2. Perform a single open-loop control on the blackbody temperature acquisition system with a fixed duty cycle to obtain a real data set; S3. To obtain a waste heat data set, under the condition of a fixed duty cycle, perform continuous open-loop control on the blackbody temperature acquisition system; S4. Calculate the natural cooling rate in S2; S5. Label the data sets collected in S2 and S3, and distinguish the heating, waste heat, and cooling data sets according to the natural cooling rate; The specific content of S2 includes the following processes: Perform single open-loop control on duty cycles of 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, and 100% respectively, and record the temperature data every 0.5 s; under the state of a fixed duty cycle, collect the temperature data from room temperature to stable temperature. After the temperature remains stable, continue to collect for 60 s, and then collect the cooling data under room temperature conditions until the temperature drops to room temperature and continue to collect for 60 s; The specific content of S3 includes the following processes: S3.

1. Record the temperature data every 0.5 s. Under the condition of a fixed duty cycle, start heating from room temperature for 60 s, and then stop the input of the heating signal for 60 s; S3.

2. Repeat this process until the blackbody temperature remains unchanged; S3.

3. Continue to heat for 60 s after the temperature remains unchanged; S3.

4. Disconnect the input of the heating signal and collect the temperature data when the acquisition system drops to room temperature; S3.

5. Continue to collect for 60 s after dropping to room temperature; S3.

6. Perform the above operations on duty cycles of 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, and 100% respectively; The specific content of S5 includes the following processes: S5.

1. First, label the heating data set. When collecting temperature data, label the heating signal. The part with the heating signal output is the heating data set; S5.

2. For the remaining data sets, label the waste heat characteristic data set by comparing the relationship between the waste heat power and the cooling power; S5.

3. The other part except the heating data set and the waste heat data set is the cooling data set; The steps of S5.2 are specifically as follows: The waste heat characteristic data set is divided into two parts:

1. The waste heat power is greater than the cooling power, which is manifested as the blackbody temperature will continue to rise for a period of time after the heating signal stops; 2. The waste heat power is less than the cooling power, which is manifested as the blackbody starts to cool after continuing to rise for a period of time after the heating stops, but the cooling rate is less than the natural cooling rate.

2. The method for obtaining and annotating a blackbody temperature data set according to claim 1, wherein, The specific content of S1 Includes the following processes: S1.

1. Build a hardware system for collecting blackbody temperature data sets; S1.

2. Use the step response method to give the blackbody an excitation signal with a duty cycle of 50%, perform a heating experiment on the blackbody system, and perform open-loop heating until the temperature remains unchanged; S1.

3. According to the obtained real system response curve of the system, perform low-pass filtering on the temperature data; S1.

4. Solve the system characteristics from the real system response curve of the system.

3. A method for obtaining and annotating a blackbody temperature data set according to claim 1, characterized in that, The specific content of S4 Includes the following processes: S4.

1. Obtain the time point when the temperature in S2 drops to room temperature and the data in the previous 120 s; S4.

2. Perform low-pass filtering on the obtained data; S4.

3. Calculate the natural cooling rate for the cooling data within these 120 s.

4. A blackbody temperature data set acquisition and annotation system, characterized in that, Applying the method for obtaining and annotating a blackbody temperature data set according to any one of claims 1-3, comprising: A system construction module that constructs a blackbody temperature acquisition system and a blackbody temperature simulation system. The blackbody temperature simulation system is used to generate a simulated data set to expand the real data set collected by the blackbody temperature acquisition system; A single open-loop control module that performs single open-loop control on the blackbody temperature acquisition system with a fixed duty cycle to obtain a real data set; A continuous open-loop control module that performs continuous open-loop control on the blackbody temperature acquisition system under a fixed duty cycle to obtain a waste heat data set; A rate calculation module that calculates the natural cooling rate in the single open-loop control module; A data set annotation module that annotates the data sets collected in the single open-loop control module and the continuous open-loop control module, and differentiates the heating, waste heat, and cooling data sets according to the natural cooling rate.

Images