LCD temperature detection method and system based on 3D printer

Abstract

The application discloses an LCD temperature detection method based on a 3D printer, and draws two temperature curves of an LCD under the same power light source when a cooling mechanism is not started to determine a normal temperature range; according to an actually collected LCD temperature value and slice data graphic data in a 3D printing process, a time change model of the LCD temperature value and the slice data is established, a time change correlation model representing the correlation change of the LCD temperature value and the slice data graphic data is established, slice data of a 3D printing model to be printed is acquired, temperature prediction is performed on any layer of slice data through the trained time change correlation model, and a predicted temperature change curve is drawn; according to the obtained predicted temperature change curve L3(t), a time point of temperature monitoring is set for a current printing task, and when actual printing is performed, temperature monitoring is performed according to the set time point; whether the temperature is within the temperature range of the drawn two temperature curves is judged, and if not, a cooling strategy is directly executed.

Description

A method and system for LCD temperature detection based on 3D printer Technical Field

[0001] This invention relates to the field of 3D printing technology, and in particular to an LCD temperature detection method and system based on a 3D printer. Background Technology

[0002] An LCD 3D printer is a type of 3D printer that utilizes the principle of liquid resin photopolymerization. It employs LCD imaging technology, driven by a computer and display driver circuitry. First, the computer program discretizes the 3D model into layers of planar graphics. Then, the display driver circuitry controls the LCD screen, converting designated areas into transparent regions corresponding to the planar graphics (other areas remain opaque black). This allows ultraviolet light from below the LCD screen to pass through and illuminate the liquid resin above the transparent areas, causing the resin to solidify. The opaque areas of the LCD screen block ultraviolet light from the non-cured areas, keeping the resin in other locations liquid. This layering process ultimately produces the desired product. LCDs typically use high-frequency ultraviolet light with high radiation energy. Under continuous exposure, the LCD screen absorbs a large amount of heat, causing its temperature to rise. The permissible temperature range for LCD screens is generally between 0 and 70°C. Exceeding this temperature range will significantly shorten the lifespan of the LCD screen, or even render it unusable.

[0003] To effectively address the issue of high temperatures in LCD screens and extend their lifespan, LCD printers are equipped with various cooling systems. Currently, two common types are air cooling, which uses a fan to accelerate airflow around the screen and remove some of the heat; and water cooling, which uses a cooling plate with liquid channels installed beneath the screen to absorb heat from the LCD through a cooling medium. However, both methods have the following problems: air cooling utilizes airflow, but air itself has a very low thermal conductivity, meaning the heat absorbed by the screen cannot be quickly transferred to the air for removal; water cooling offers much faster heat conduction, but the complex internal flow channel structure of the cooling plate and the flowing liquid medium inevitably cause the projection angle of ultraviolet light to deflect when it passes through the cooling plate, thus affecting the dimensional accuracy of the cured image.

[0004] Moreover, both of the above methods are essentially passive heat dissipation, meaning that the screen absorbs a large amount of heat and then attempts to reduce that heat through various means. Therefore, how to change this passive remedial approach and shift the focus from heat dissipation after the LCD screen absorbs and heats up to the stage where the LCD screen absorbs and heats up, directly reducing the heat absorbed by the LCD screen at its source, and thus maintaining a lower screen temperature, has become a crucial technical problem for those skilled in the art. Therefore, some existing technologies start from the LCD material or structure, reducing the high-frequency ultraviolet light irradiating the LCD screen during the printing process to reduce the heat absorbed by the LCD screen at its source, and then using heat dissipation devices for further cooling. Summary of the Invention

[0005] This invention aims to at least solve one of the technical problems existing in the prior art. To this end, this invention discloses an LCD temperature detection method based on a 3D printer, the method comprising the following steps:

[0006] Step 1: Plot two temperature curves of the LCD under the same power light source without turning on the cooling mechanism. One of the two temperature curves is the temperature change curve L1(t) of the LCD when it is completely unblocked, and the other is the temperature change curve L2(t) of the LCD when it is fully blocked.

[0007] Step 2: Based on the actual LCD temperature values ​​and slice data graphic data collected during the 3D printing process, establish a time variation model of LCD temperature values ​​and slice data, then establish a time variation correlation model representing the correlation between LCD temperature values ​​and slice data graphic data, and finally obtain a trained time variation correlation model by training the time variation correlation model.

[0008] Step 3: Obtain the slice data of the 3D printed model to be printed, and use the trained time change correlation model to predict the temperature of any slice data and plot the predicted temperature change curve L3(t).

[0009] Step 4: Based on the predicted temperature change curve L3(t), set the temperature monitoring time point for the current printing task, and monitor the temperature according to the set time point during actual printing.

[0010] Step 5: Determine whether the temperature is within the temperature range of the two temperature curves drawn in Step 1. If it is not within the temperature range, execute the cooling strategy directly.

[0011] Preferably, step 2 further includes: inputting the correlated LCD temperature values ​​and sliced ​​image data into a convolutional neural network for spatiotemporal relationship learning, wherein the sliced ​​image data includes extractable image features:

[0012] OUT Δt =F(θ(IMAGE(Δt),T)+μ)

[0013] Among them, OUT Δt θ represents the output data after learning by the convolutional network; F is the activation function of the time-varying correlation model; IMAGE(Δt) represents the image features of the slice data as it changes over time; T represents the temperature value of the LCD; θ and μ are the adjustable parameters of the convolutional network to be learned.

[0014] Preferably, the weight values ​​of the adjustable parameters in the convolutional neural network are determined by gradient descent.

[0015] Preferably, step 4 further includes the method for determining the time point as follows:

[0016] L3(t) = α(L2(t) - L1(t)) + β

[0017] Wherein, α and β represent the bias coefficients set at the time point, which are pre-set by the 3D printing equipment manufacturer based on the heat resistance of the LCD, and the time point t calculated according to the equation represents the set time point.

[0018] Preferably, the cooling strategy selection method is as follows: first, the dataset for a single cooling method is initialized, and the collected cooling data set is normalized to output a data set in a unified format. Then, based on the changing trend of L3(t) after the set time point, the remaining data information is used to form a data association matrix and a judgment matrix of cooling method processing influencing factors, and the corresponding strategy is executed based on the judgment matrix.

[0019] Preferably, the method for determining whether the temperature is within the temperature range of the two temperature curves drawn in step 1 is to use image comparison. A temperature risk warning curve L4(t) is drawn in advance based on the two temperature curves. When the predicted temperature change curve L3(t) intersects with the temperature risk warning curve L4(t), it is determined that the temperature exceeds the range and a cooling strategy needs to be implemented.

[0020] This invention also discloses an LCD temperature detection system based on a 3D printer, the system comprising:

[0021] The LCD temperature range determination unit plots two temperature curves of the LCD under the same power light source without the cooling mechanism activated. One of the temperature curves is L1(t) when the LCD is completely unobstructed, and the other is L2(t) when the LCD is fully obstructed. The LCD temperature prediction unit establishes a time-varying model of the LCD temperature values ​​and slice data based on the actual LCD temperature values ​​and slice data collected during the 3D printing process. This model then establishes a time-varying correlation model representing the correlation between the LCD temperature values ​​and slice data, and trains the time-varying correlation model to obtain... The system includes: a trained time-varying correlation model; a prediction curve generation unit, which acquires slice data of the 3D printing model to be printed, performs temperature prediction on any slice data using the trained time-varying correlation model, and plots the predicted temperature change curve L3(t); a risk point setting unit, which sets the temperature monitoring time point for the current printing task based on the obtained predicted temperature change curve L3(t), and performs temperature monitoring based on the set time point during actual printing; and a cooling strategy execution unit, which determines whether the temperature is within the temperature range of the two temperature curves plotted by the LCD temperature range determination unit. If the temperature is not within the temperature range, the cooling strategy is executed directly.

[0022] Preferably, the LCD temperature prediction unit further includes: inputting the correlated LCD temperature value and sliced ​​data graphic data into a convolutional neural network for spatiotemporal relationship learning, wherein the sliced ​​data graphic data includes extractable image features:

[0023] OUT Δt =F(θ(IMAGE(Δt),T)+μ)

[0024] Among them, OUT Δt θ represents the output data after learning by the convolutional network; F is the activation function of the time-varying correlation model; IMAGE(Δt) represents the image features of the slice data as it changes over time; T represents the temperature value of the LCD; θ and μ are the adjustable parameters of the convolutional network to be learned, and the weights of the adjustable parameters in the convolutional neural network are determined by gradient descent.

[0025] Preferably, the method for determining the time point is as follows:

[0026] L3(t) = α(L2(t) - L1(t)) + β

[0027] Wherein, α and β represent the bias coefficients set at the time point, which are pre-set by the 3D printing equipment manufacturer based on the heat resistance of the LCD, and the time point t calculated according to the equation represents the set time point.

[0028] Preferably, the cooling strategy selection method is as follows: first, the dataset for a single cooling method is initialized, and the collected cooling data set is normalized to output a data set in a unified format. Then, based on the changing trend of L3(t) after the set time point, the remaining data information is used to form a data association matrix and a judgment matrix of cooling method processing influencing factors, and the corresponding strategy is executed based on the judgment matrix.

[0029] Compared with the prior art, the beneficial effects of this invention are: it eliminates the need for frequent temperature measurements of the LCD; it uses temperature prediction to set monitoring time points within the temperature change range of the LCD under both shading and unshading conditions; and it flexibly changes the monitoring time points based on different slice data. Furthermore, since general advanced LCD printing equipment has multiple cooling methods, this invention introduces a cooling decision-making approach to correlate the temperature prediction trend after the monitoring point, linking the cooling strategy with the temperature prediction trend. This is an innovative design not found in the 3D printing field. Finally, this invention also specifically designs a method to determine whether the temperature will exceed the range using images, and the plotted curves provide a clear overview for the user. Attached Figure Description

[0030] The invention will be further understood from the following description taken in conjunction with the accompanying drawings. The components in the drawings are not necessarily drawn to scale, but rather the emphasis is on illustrating the principles of the embodiments. In the drawings, the same reference numerals designate corresponding parts in different views.

[0031] Figure 1 is a flowchart of an LCD temperature detection method based on a 3D printer according to the present invention. Detailed Implementation

[0032] The technical solution of the present invention will now be described in more detail with reference to the accompanying drawings and embodiments.

[0033] Mobile terminals implementing various embodiments of the present invention will now be described with reference to the accompanying drawings. In the following description, suffixes such as "module," "part," or "unit" used to denote elements are used only for the convenience of the description of the present invention and have no specific meaning in themselves. Therefore, "module" and "part" can be used interchangeably.

[0034] Mobile terminals can be implemented in various forms. For example, the terminals described in this invention may include mobile terminals such as mobile phones, smartphones, laptops, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), navigation devices, etc., as well as fixed terminals such as digital TVs, desktop computers, etc. Hereinafter, it will be assumed that the terminal is a mobile terminal. However, those skilled in the art will understand that, in addition to elements specifically designed for mobile purposes, the construction according to embodiments of the present invention can also be applied to fixed-type terminals.

[0035] Figure 1 illustrates a method for LCD temperature detection based on a 3D printer. The method includes the following steps:

[0036] Step 1: Plot two temperature curves of the LCD under the same power light source without turning on the cooling mechanism. One of the two temperature curves is the temperature change curve L1(t) of the LCD when it is completely unblocked, and the other is the temperature change curve L2(t) of the LCD when it is fully blocked.

[0037] Step 2: Based on the actual LCD temperature values ​​and slice data graphic data collected during the 3D printing process, establish a time variation model of LCD temperature values ​​and slice data, then establish a time variation correlation model representing the correlation between LCD temperature values ​​and slice data graphic data, and finally obtain a trained time variation correlation model by training the time variation correlation model.

[0038] Step 3: Obtain the slice data of the 3D printed model to be printed, and use the trained time change correlation model to predict the temperature of any slice data and plot the predicted temperature change curve.

[0039] Step 4: Based on the predicted temperature change curve L3(t), set the temperature monitoring time point for the current printing task, and monitor the temperature according to the set time point during actual printing.

[0040] Step 5: Determine whether the temperature is within the temperature range of the two temperature curves drawn in Step 1. If it is not within the temperature range, execute the cooling strategy directly.

[0041] This embodiment describes a temperature detection method for an LCD screen. However, for a complete LCD-based 3D printing device, heat dissipation is required not only for the LCD screen but also for the corresponding light source. Traditional light source heat dissipation methods correlate the power of the light source with the heat generated, and then link the generated heat to the light source cooling module. That is, the higher the power of the light source, the higher the power of the cooling module. When using air cooling, the higher the power of the light source, the higher the speed of the air-cooling fan; while when using liquid cooling, the higher the power of the light source, the higher the power of the liquid cooling compressor. Similarly, the temperature detection and cooling strategy for the LCD screen in this embodiment can also be used to manage the temperature of the light source.

[0042] Furthermore, step 2 further includes:

[0043] The correlated LCD temperature values ​​and sliced ​​image data are input into a convolutional neural network for spatiotemporal relationship learning. The sliced ​​image data includes extractable image features.

[0044] OUT Δt =F(θ(IMAGE(Δt),T)+μ)

[0045] Among them, OUT Δt θ represents the output data after learning by the convolutional network; F is the activation function of the time-varying correlation model; IMAGE(Δt) represents the image features of the slice data as it changes over time; T represents the temperature value of the LCD; θ and μ are the adjustable parameters of the convolutional network to be learned.

[0046] In this embodiment, optionally, the activation function can simultaneously use ReLU (Rectifier Linear Unit) and NReLU (Negative ReLU) functions as activation functions, where the specific formulas are as follows: ReLU(x) = max(x,0), NReLU(x) = max(-x,0). In this embodiment, the activation function used in the activation function layer is A(x) = (ReLU(x), NReLU(x)).

[0047] Furthermore, the weights of the adjustable parameters in the convolutional neural network are determined using gradient descent.

[0048] Furthermore, step 4 further includes the method for determining the time point as follows:

[0049] L3(t) = α(L2(t) - L1(t)) + β

[0050] Wherein, α and β represent the bias coefficients set at the time point, which are pre-set by the 3D printing equipment manufacturer based on the heat resistance of the LCD, and the time point t calculated according to the equation represents the set time point.

[0051] Furthermore, the cooling strategy selection method is as follows: first, the dataset for a single cooling method is initialized, and the collected cooling data set is normalized to output a data set in a unified format. Then, based on the changing trend of L3(t) after the set time point, the remaining data information is used to form a data association matrix and a judgment matrix of cooling method processing influencing factors, and the corresponding strategy is executed based on the judgment matrix.

[0052] For example, in this embodiment, assuming there are 4 strategies, the corresponding set of optional cooling methods is C1: air cooling for LCD, C2: liquid cooling for cooling, and C3: turning off the heat source, etc.

[0053]

[0054] From a hardware perspective, the present invention also discloses an LCD temperature detection system based on a 3D printer. The system includes: an LCD temperature range determination unit, which plots two temperature curves of the LCD under the same power light source without the cooling mechanism activated. One of the two temperature curves is L1(t) representing the temperature change over time when the LCD is completely unobstructed, and the other is L2(t) representing the temperature change when the LCD is fully obstructed; and an LCD temperature prediction unit, which establishes a time-varying model of the LCD temperature value and slice data based on the actual LCD temperature values ​​and slice data graphic data collected during the 3D printing process. This model then establishes a time-varying correlation model representing the correlation between the LCD temperature value and the slice data graphic data, and trains the time-varying correlation model to obtain a trained time-varying correlation model.

[0055] The prediction curve generation unit acquires slice data of the 3D printing model to be printed, predicts the temperature of any slice data layer using the trained time-varying correlation model, and plots the predicted temperature change curve L3(t). The risk point setting unit sets the temperature monitoring time point for the current printing task based on the obtained predicted temperature change curve L3(t), and monitors the temperature according to the set time point during actual printing. The cooling strategy execution unit determines whether the temperature is within the temperature range of the two temperature curves plotted by the LCD temperature range determination unit. If it is not within the temperature range, the cooling strategy is executed directly.

[0056] Preferably, the LCD temperature prediction unit further includes: inputting the correlated LCD temperature value and sliced ​​data graphic data into a convolutional neural network for spatiotemporal relationship learning, wherein the sliced ​​data graphic data includes extractable image features:

[0057] OUT Δt =F(θ(IMAGE(Δt),T)+μ)

[0058] Among them, OUT Δt θ represents the output data after learning by the convolutional network; F is the activation function of the time-varying correlation model; IMAGE(Δt) represents the image features of the slice data as it changes over time; T represents the temperature value of the LCD; θ and μ are the adjustable parameters of the convolutional network to be learned, and the weights of the adjustable parameters in the convolutional neural network are determined by gradient descent.

[0059] Preferably, the method for determining the time point is as follows:

[0060] L3(t) = α(L2(t) - L1(t)) + β

[0061] Wherein, α and β represent the bias coefficients set at the time point, which are pre-set by the 3D printing equipment manufacturer based on the heat resistance of the LCD, and the time point t calculated according to the equation represents the set time point.

[0062] Preferably, the selection method for the cooling strategy is as follows: first, initialize the dataset for a single cooling method, and normalize the collected cooling data set to output a data set in a uniform format. Then, based on the changing trend of L3(t) after the set time point, form a data association matrix and a judgment matrix of the influencing factors of the cooling method processing according to the remaining data information, and execute the corresponding strategy based on the judgment matrix.

[0063] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0064] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0065] While the invention has been described above with reference to various embodiments, it should be understood that many changes and modifications can be made without departing from the scope of the invention. Therefore, the detailed description above is intended to be illustrative rather than restrictive, and it should be understood that the following claims (including all equivalents) are intended to define the spirit and scope of the invention. These embodiments should be understood as illustrative only and not as limiting the scope of protection of the invention. After reading the description of this invention, those skilled in the art can make various alterations or modifications to the invention, and these equivalent changes and modifications also fall within the scope defined by the claims of this invention.

Claims

1. A method for LCD temperature detection based on a 3D printer, characterized in that, The method includes the following steps: Step 1, plotting two temperature curves of the LCD under the same power light source without turning on the cooling mechanism, wherein one of the two temperature curves is the temperature change curve L1(t) of the LCD when it is completely unobstructed, and the other is the temperature change curve L2(t) of the LCD when it is fully obstructed; Step 2, establishing a time change model of LCD temperature value and slice data based on the actual LCD temperature value and slice data graphic data collected during the 3D printing process, and then establishing a time change correlation model representing the correlation between LCD temperature value and slice data graphic data, and then training the time change correlation model to obtain a trained time change correlation model; Step 3, acquiring the slice data of the 3D printing model to be printed, predicting the temperature of any layer of slice data through the trained time change correlation model and plotting the predicted temperature change curve L3(t); Step 4, setting the temperature monitoring time point for the current printing task based on the obtained predicted temperature change curve L3(t), and monitoring the temperature according to the set time point during actual printing; Step 5, determining whether it is within the temperature range of the two temperature curves plotted in Step 1, and if it is not within the temperature range, directly executing the cooling strategy.

2. The LCD temperature detection method based on a 3D printer as described in claim 1, characterized in that, Step 2 further includes: inputting the correlated LCD temperature values ​​and sliced ​​image data into a convolutional neural network for spatiotemporal relationship learning, wherein the sliced ​​image data includes extractable image features: OUT Δt =F(θ(IMAGE(Δt),T)+μ)where, OUT Δt θ represents the output data after learning by the convolutional network; F is the activation function of the time-varying correlation model; IMAGE(Δt) represents the image features of the slice data as it changes over time; T represents the temperature value of the LCD; θ and μ are the adjustable parameters of the convolutional network to be learned.

3. The LCD temperature detection method based on a 3D printer as described in claim 2, characterized in that, The weights of adjustable parameters in a convolutional neural network are determined using gradient descent.

4. The LCD temperature detection method based on a 3D printer as described in claim 3, characterized in that, Step 4 further includes the following method for determining the time point: L3(t)=α(L2(t)-L1(t))+β, where α and β represent the bias coefficients for setting the time point, which are preset by the 3D printing equipment manufacturer according to the heat resistance of the LCD, and the time point t calculated according to the equation represents the set time point.

5. The LCD temperature detection method based on a 3D printer as described in claim 4, characterized in that, The cooling strategy selection method is as follows: first, the dataset for a single cooling method is initialized, and the collected cooling data set is normalized to output a data set in a unified format. Then, based on the changing trend of L3(t) after the set time point, the remaining data information is used to form a data association matrix and a judgment matrix of the influencing factors of the cooling method, and the corresponding strategy is executed based on the judgment matrix.

6. The LCD temperature detection method based on a 3D printer as described in claim 5, characterized in that, The method for determining whether the temperature is within the temperature range of the two temperature curves drawn in step 1 is to use image comparison. A temperature risk warning curve L4(t) is drawn in advance based on the two temperature curves. When the predicted temperature change curve L3(t) intersects with the temperature risk warning curve L4(t), it is determined that the temperature exceeds the range and a cooling strategy needs to be implemented.

7. An LCD temperature detection system based on a 3D printer, characterized in that, The system includes: an LCD temperature range determination unit, which plots two temperature curves of the LCD under the same power light source without the cooling mechanism activated, wherein one of the two temperature curves is the temperature change curve L1(t) of the LCD when it is completely unobstructed, and the other is the temperature change curve L2(t) of the LCD when it is fully obstructed; and an LCD temperature prediction unit, which establishes a time-varying model of LCD temperature values ​​and slice data based on the actual LCD temperature values ​​and slice data graphic data collected during the 3D printing process, and further establishes a time-varying correlation model representing the correlation between LCD temperature values ​​and slice data graphic data, and then performs further analysis on the time-varying correlation model. The system trains a time-varying correlation model; a prediction curve generation unit acquires slice data of the 3D printing model to be printed, predicts the temperature of any slice data using the trained time-varying correlation model, and plots the predicted temperature change curve L3(t); a risk point setting unit sets the temperature monitoring time point for the current printing task based on the obtained predicted temperature change curve L3(t), and monitors the temperature according to the set time point during actual printing; a cooling strategy execution unit determines whether the temperature is within the temperature range of the two temperature curves plotted by the LCD temperature range determination unit, and if not, directly executes the cooling strategy.

8. The LCD temperature detection system based on a 3D printer as described in claim 7, characterized in that, The LCD temperature prediction unit further includes: inputting the correlated LCD temperature value and sliced ​​image data into a convolutional neural network for spatiotemporal relationship learning, wherein the sliced ​​image data includes extractable image features: OUT Δt =F(θ(IMAGE(Δt),T)+μ)where, OUT Δt θ represents the output data after learning by the convolutional network; F is the activation function of the time-varying correlation model; IMAGE(Δt) represents the image features of the slice data as it changes over time; T represents the temperature value of the LCD; θ and μ are the adjustable parameters of the convolutional network to be learned, and the weights of the adjustable parameters in the convolutional neural network are determined by gradient descent.

9. The LCD temperature detection system based on a 3D printer as described in claim 8, characterized in that, The time point is determined as follows: L3(t)=α(L2(t)-L1(t))+β, where α and β represent the bias coefficients for setting the time point, which are preset by the 3D printing equipment manufacturer according to the heat resistance of the LCD. The time t calculated according to the equation represents the set time point.

10. The LCD temperature detection system based on a 3D printer as described in claim 9, characterized in that, The cooling strategy selection method is as follows: first, the dataset for a single cooling method is initialized, and the collected cooling data set is normalized to output a data set in a unified format. Then, based on the changing trend of L3(t) after the set time point, the remaining data information is used to form a data association matrix and a judgment matrix of the influencing factors of the cooling method, and the corresponding strategy is executed based on the judgment matrix.

Images