LED driving power supply intelligent commissioning and testing method based on deep learning

Abstract

The invention discloses an LED driving power supply intelligent debugging and testing method based on deep learning, and relates to the technical field of deep learning, and the method comprises the steps: obtaining the historical working data of an LED driving power supply in different working environments and working states; performing data annotation distinguishing on the historical working data to obtain historical normal working data under normal working states of different working environments; performing standardization processing on the historical normal working data to obtain standard historical data; according to the standard historical data, constructing a deep learning power supply intelligent commissioning and testing model, and further obtaining a power supply intelligent commissioning and testing index; according to the intelligent power supply adjusting and testing index, intelligent adjusting and testing are carried out on the LED driving power supply, and the optical performance index of the LED after intelligent adjusting and testing is obtained; and optical performance indexes are compared and judged, so that the performance and efficiency of the LED driving power supply are improved. And the intelligent commissioning and testing refinement degree is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of deep learning, and specifically to an intelligent debugging method for LED driver power supplies based on deep learning. Background Technique

[0002] With the rapid development of LED technology, LED lamps are widely used in various fields such as lighting, display, and automotive lamps due to their high efficiency, long lifespan, and environmental friendliness. As an important part of LED lamps, the performance of LED driver power supplies directly affects the working efficiency and lifespan of LEDs. Therefore, how to effectively debug and optimize the performance of LED driver power supplies has become a research hotspot.

[0003] Currently, the debugging methods for LED driver power supplies mainly include manual debugging and automatic debugging based on simple algorithms. Although these methods can achieve the debugging of LED driver power supplies to a certain extent, they still have the following deficiencies: 1. Manual debugging: Traditional manual debugging relies on the experience and technical level of engineers. It is not only time-consuming and laborious, but also the stability and consistency of the debugging results are poor, making it difficult to meet the requirements of large-scale production.

[0004] 2. Simple algorithm debugging: Some automatic debugging methods based on simple algorithms, such as PID (Proportional-Integral-Derivative) control, although they can achieve a certain degree of automation, for complex power supply working states and changing working environments, their debugging effects are limited, and it is difficult to achieve refined control.

[0005] As an important technology in the field of artificial intelligence, deep learning has powerful data processing and pattern recognition capabilities. In recent years, deep learning has achieved remarkable results in fields such as image recognition, natural language processing, and speech recognition, demonstrating its advantages in dealing with complex problems. Introducing deep learning technology into the debugging of LED driver power supplies can effectively improve the efficiency and accuracy of debugging. The specific advantages are as follows: 1. Self-learning ability: The deep learning model can automatically extract and identify the characteristics of the power supply working state by learning a large amount of historical data and actual operation data, realizing intelligent debugging.

[0006] 2. High-precision prediction: The deep learning model can accurately predict the performance parameters of the LED driver power supply under different working conditions, thereby achieving precise adjustment and optimization.

[0007] 3. Real-time regulation: Combining online learning and adjustment of real-time data, the deep learning model can dynamically adapt to changes in the working environment of the power supply, ensuring that the power supply is always in the best working state.

[0008] Therefore, an intelligent debugging method for LED driver power supplies based on deep learning is provided. Summary of the Invention

[0009] In order to solve the above technical problems, the object of the present invention is to provide an intelligent debugging method for LED drive power supplies based on deep learning.

[0010] In order to achieve the above object, the present invention provides the following technical solution: An intelligent debugging method for LED drive power supplies based on deep learning, comprising the following steps: Step S1: Obtain historical working data of the LED drive power supply under different working environments and working states; perform data annotation and differentiation on the historical working data to obtain historical normal working data under normal working states in different working environments; Step S2: Perform standardization processing on the historical normal working data to obtain standard historical data; construct a deep learning power supply intelligent debugging model based on the standard historical data, and then obtain a power supply intelligent debugging index; Step S3: Perform intelligent debugging on the LED drive power supply according to the power supply intelligent debugging index to obtain the optical performance index of the LED after intelligent debugging; perform comparison and judgment on the optical performance index to improve the performance and efficiency of the LED drive power supply.

[0011] Further, the process of obtaining historical working data of the LED drive power supply under different working environments and working states includes: Set up a data acquisition unit; The data acquisition unit is composed of several acquisition nodes, which are respectively set at the data acquisition point positions of the LED drive power supply; the several acquisition nodes are used to collect real-time working data of the LED drive power supply under different working environments and working states in real time, and set the acquisition period and acquisition time; The real-time working data is a set of input voltage, input current, output voltage, output current, ambient temperature, and ambient humidity of the LED drive power supply collected at the same acquisition time; the acquisition period includes several acquisition times, and each acquisition time collects a corresponding set of real-time working data; Set up a real-time database and a historical database; Upload the real-time working data collected in real time during the acquisition period to the real-time database for storage. When the real-time database receives the real-time working data of the next acquisition period, send the saved real-time working data of the previous acquisition period to the historical database for storage, and mark the real-time working data saved in the historical database as historical working data; The historical working data includes historical input voltage, historical input current, historical output voltage, historical output current, historical ambient temperature, and historical ambient humidity.

[0012] Further, the process of data annotation and differentiation of historical work data to obtain historical normal work data under normal working conditions in different working environments includes: Set the standard environmental power difference threshold; According to the historical work data corresponding to each acquisition moment within the acquisition period, obtain the environmental power difference of the LED driver power supply in the working state; If the environmental power difference is greater than the standard environmental power difference threshold, mark the working state of the LED driver power supply as an abnormal working state; If the environmental power difference is less than or equal to the standard environmental power difference threshold, mark the working state of the LED driver power supply as a normal working state, and mark the historical work data in the normal working state as historical normal work data; The historical normal work data includes normal historical input voltage, normal historical input current, normal historical output voltage, normal historical output current, normal historical environmental temperature, and normal historical environmental humidity.

[0013] Further, the process of obtaining the environmental power difference of the LED driver power supply in the working state includes: Mark the historical input voltage, historical input current, historical output voltage, historical output current, historical environmental temperature, and historical environmental humidity in the historical work data corresponding to each acquisition moment within the acquisition period respectively; Obtain the environmental power difference according to the marked historical input voltage, historical input current, historical output voltage, historical output current, historical environmental temperature, and historical environmental humidity.

[0014] Further, the process of standardizing the historical normal work data to obtain standard historical data includes: Number the historical normal work data, denoted as ; is a natural number; Mark the normal historical input voltage, normal historical input current, normal historical output voltage, normal historical output current, normal historical environmental temperature, and normal historical environmental humidity respectively; Calculate the mean value of the marked normal historical input voltage, normal historical input current, normal historical output voltage, normal historical output current, normal historical environmental temperature, and normal historical environmental humidity respectively to obtain the mean value of the historical normal work data; Obtain the standard deviation of the historical normal work data according to the historical normal work data and the mean value of the historical normal work data; Obtain the standard historical data according to the mean value of the historical normal work data and the standard deviation of the historical normal work data.

[0015] Further, the process of constructing a deep learning power intelligent debugging model based on standard historical data and then obtaining the power intelligent debugging index includes: Based on a convolutional neural network, construct a standard deep learning power intelligent debugging model; Divide the obtained several groups of standard historical data into a training set, a validation set, and a test set, and input the training set, validation set, and test set into the standard deep learning power intelligent debugging model to train the standard deep learning power intelligent debugging model, obtain the trained standard deep learning power intelligent debugging model, and denote the trained standard deep learning power intelligent debugging model as the deep learning power intelligent debugging model; According to the deep learning power intelligent debugging model, obtain the power intelligent debugging index.

[0016] Further, the process of performing intelligent debugging on the LED driver power supply according to the power intelligent debugging index and obtaining the optical performance index of the LED after intelligent debugging includes: Obtain the real-time working data of the LED driver power supply under different working environments in real time, and mark the real-time working data; According to the marked real-time working data and the power intelligent debugging index, obtain the optical performance index.

[0017] Further, the process of comparing and judging the optical performance index includes: Preset the standard optical performance index; If the optical performance index is greater than or equal to the standard optical performance index, the optical performance of the LED reaches the expected effect, indicating that the intelligent debugging of the LED driver power supply meets the requirements; If the optical performance index is less than the standard optical performance index, the optical performance of the LED does not reach the expected effect, indicating that the intelligent debugging of the LED driver power supply does not meet the requirements.

[0018] Compared with the prior art, the beneficial effects of the present invention are: obtaining the historical working data of the LED driver power supply under different working environments; performing data annotation and differentiation on the historical working data to obtain the historical normal working data under the normal working state of different working environments; performing standardization processing on the historical normal working data to obtain standard historical data; constructing a deep learning power intelligent debugging model according to the standard historical data, and then obtaining the power intelligent debugging index; performing intelligent debugging on the LED driver power supply according to the power intelligent debugging index to obtain the optical performance index of the LED after intelligent debugging; comparing and judging the optical performance index to improve the performance and efficiency of the LED driver power supply. The refinement degree of intelligent debugging is improved. Description of the Drawings

[0019] To more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the accompanying drawings required in the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments recorded in the present invention. For those of ordinary skill in the art, other accompanying drawings can also be obtained based on these drawings.

[0020] Figure 1 This is the schematic diagram of the present invention. Specific embodiments

[0021] To make the objectives, technical solutions, and advantages of the present invention clearer, the following will describe the technical solutions of the present invention in detail. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope protected by the present invention.

[0022] As Figure 1 shown, an intelligent debugging method for an LED driver power supply based on deep learning includes the following steps: Step S1: Obtain the historical working data of the LED driver power supply under different working environments; perform data annotation and differentiation on the historical working data to obtain the historical normal working data under the normal working state of different working environments; Step S2: Perform standardization processing on the historical normal working data to obtain standard historical data; construct a deep learning power supply intelligent debugging model based on the standard historical data, and then obtain the power supply intelligent debugging index; Step S3: Perform intelligent debugging on the LED driver power supply according to the power supply intelligent debugging index to obtain the optical performance indicators of the LED after intelligent debugging; perform comparison and judgment on the optical performance indicators to improve the performance and efficiency of the LED driver power supply.

[0023] It should be further noted that in the specific implementation process, the specific process of obtaining the historical working data of the LED driver power supply under different working environments includes: Set up a data acquisition unit; The data acquisition unit consists of several acquisition nodes, which are respectively set at the data acquisition point positions of the LED driver power supply; the several acquisition nodes are used to collect the real-time working data of the LED driver power supply under different working environments in real time, and set the acquisition period and acquisition time; It should be further noted that the real-time working data is a set of input voltage, input current, output voltage, output current, ambient temperature, and ambient humidity of the LED driver power supply collected at the same acquisition moment; the acquisition period includes several acquisition moments, and each acquisition moment collects a corresponding set of real-time working data; Set up a real-time database and a historical database; Upload the real-time working data collected in real time during the acquisition period to the real-time database for storage. When the real-time database receives the real-time working data of the next acquisition period, send the saved real-time working data of the previous acquisition period to the historical database for storage, and mark the real-time working data saved in the historical database as historical working data; It should be further noted that the historical database stores several sets of historical working data saved in the time sequence of the acquisition period; the historical working data includes historical input voltage, historical input current, historical output voltage, historical output current, historical ambient temperature, and historical ambient humidity.

[0024] It should be further noted that in the specific implementation process, the specific process of performing data annotation and differentiation on the historical working data to obtain the historical normal working data under the normal working state of different working environments includes: Obtain the historical working data from the historical database; Set the standard environmental power difference threshold ; According to the historical working data corresponding to each acquisition moment within the acquisition period, obtain the environmental power difference of the LED driver power supply in the working state; It should be further noted that in the specific implementation process, the specific process of obtaining the environmental power difference of the LED driver power supply in the working state includes: Mark the historical input voltage, historical input current, historical output voltage, historical output current, historical ambient temperature, and historical ambient humidity in the historical working data corresponding to each acquisition moment within the acquisition period respectively, and denote them as 、 、 、 、 and ; According to the historical input voltage 、historical input current 、historical output voltage 、historical output current 、historical ambient temperature and historical ambient humidity , obtain the calculation formula for the environmental power difference as: ; If the environmental power difference is greater than the standard environmental power difference threshold , mark the working state of the LED driver power supply as an abnormal working state; If the environmental power difference is less than or equal to the standard environmental power difference threshold , mark the working state of the LED driver power supply as a normal working state, and mark the historical working data of the normal working state as historical normal working data; It should be further noted that the historical normal working data includes normal historical input voltage, normal historical input current, normal historical output voltage, normal historical output current, normal historical environmental temperature, and normal historical environmental humidity.

[0025] It should be further noted that in the specific implementation process, the specific process of standardizing the historical normal working data to obtain standard historical data includes: Number the historical normal working data, denoted as ; is a natural number; Mark the normal historical input voltage, normal historical input current, normal historical output voltage, normal historical output current, normal historical environmental temperature, and normal historical environmental humidity respectively, denoted as , , , , and ; Calculate the mean value of the normal historical input voltage , normal historical input current , normal historical output voltage , normal historical output current , normal historical environmental temperature and normal historical environmental humidity respectively to obtain the mean value of historical normal working data; It should be further noted that the mean value of historical normal working data includes the mean value of input voltage , mean value of input current , mean value of output voltage , mean value of output current , mean value of environmental temperature and mean value of environmental humidity ; Obtain the standard deviation of historical normal working data according to the historical normal working data and the mean value of historical normal working data; It should be further noted that the standard deviation of the historical normal operating data includes the standard deviation of the input voltage , the standard deviation of the input current , the standard deviation of the output voltage , the standard deviation of the output current , the standard deviation of the ambient temperature and the standard deviation of the ambient humidity ; According to the mean value of the historical normal operating data and the standard deviation of the historical normal operating data, standard historical data is obtained; It should be further noted that the standard historical data includes the standard input voltage , the standard input current , the standard output voltage , the standard output current , the standard ambient temperature and the standard ambient humidity ; It should be further noted that the calculation formula of the standard input voltage is: ; Repeating the calculation process of the standard input voltage , the standard input current , the standard output voltage , the standard output current , the standard ambient temperature and the standard ambient humidity are respectively obtained.

[0026] It should be further noted that in the specific implementation process, according to the standard historical data, the specific process of constructing a deep learning power intelligent debugging model and then obtaining the power intelligent debugging index includes: Based on the convolutional neural network, a standard deep learning power intelligent debugging model is constructed; The obtained several groups of standard historical data are divided into a training set, a validation set and a test set, and the training set, the validation set and the test set are input into the standard deep learning power intelligent debugging model to train the standard deep learning power intelligent debugging model, and the trained standard deep learning power intelligent debugging model is obtained, and the trained standard deep learning power intelligent debugging model is denoted as the deep learning power intelligent debugging model; According to the deep learning power intelligent debugging model, the calculation formula of the power intelligent debugging index is: .

[0027] It should be further noted that in the specific implementation process, according to the intelligent debugging index of the power supply, the specific process of intelligent debugging of the LED driving power supply and obtaining the optical performance index of the LED after intelligent debugging includes: Obtain the real-time working data of the LED driving power supply under different working environments and working states in real time; Mark the input voltage, input current, output voltage, output current, ambient temperature, and ambient humidity as , , , , and respectively; According to the intelligent debugging index of the power supply , input voltage , input current , output voltage , output current , ambient temperature and ambient humidity , the calculation formula for obtaining the optical performance index is: ; It should be further noted that in the specific implementation process, the specific process of comparing and judging the optical performance index includes: Preset the standard optical performance index ; If the optical performance index is greater than or equal to the standard optical performance index , it means that the optical performance of the LED reaches the expected effect, indicating that the intelligent debugging of the LED driving power supply meets the requirements; If the optical performance index is less than the standard optical performance index , it means that the optical performance of the LED does not reach the expected effect, indicating that the intelligent debugging of the LED driving power supply does not meet the requirements, and professionals are reminded to maintain the LED driving power supply.

[0028] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical method of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical method of the present invention.

Claims

1. An intelligent debugging method for LED driving power supplies based on deep learning, characterized in that, It includes the following steps: Step S1: Obtain the historical working data of the LED driving power supply under different working environments and working states; perform data annotation and differentiation on the historical working data to obtain the historical normal working data under the normal working states of different working environments; Step S2: Perform standardization processing on the historical normal working data to obtain standard historical data; construct a deep learning power intelligent debugging model based on the standard historical data, and then obtain the power intelligent debugging index; Step S3: Perform intelligent debugging on the LED driving power supply according to the power intelligent debugging index to obtain the optical performance indicators of the LED after intelligent debugging; perform comparison and judgment on the optical performance indicators to improve the performance and efficiency of the LED driving power supply.

2. The intelligent debugging method for an LED driver power supply based on deep learning according to claim 1, wherein The process of obtaining the historical working data of the LED driving power supply under different working environments and working states includes: Set up a data acquisition unit; The data acquisition unit is composed of several acquisition nodes, which are respectively set at the data acquisition point positions of the LED driving power supply; the several acquisition nodes are used to collect the real-time working data of the LED driving power supply under different working environments and working states in real time, and set the acquisition period and acquisition time; The real-time working data is a set of input voltage, input current, output voltage, output current, ambient temperature, and ambient humidity of the LED driving power supply collected at the same acquisition time; the acquisition period contains several acquisition times, and each acquisition time collects a corresponding set of real-time working data; Set up a real-time database and a historical database; Upload the real-time working data collected in real time during the acquisition period to the real-time database for storage. When the real-time database receives the real-time working data of the next acquisition period, send the saved real-time working data of the previous acquisition period to the historical database for storage, and mark the real-time working data saved in the historical database as historical working data; The historical working data includes historical input voltage, historical input current, historical output voltage, historical output current, historical ambient temperature, and historical ambient humidity.

3. The intelligent debugging method for an LED driving power supply based on deep learning according to claim 2, characterized in that The process of performing data annotation and differentiation on the historical working data to obtain the historical normal working data under the normal working states of different working environments includes: Set the standard environmental power difference threshold; According to the historical working data corresponding to each acquisition time during the acquisition period, obtain the environmental power difference of the LED driving power supply in the working state; If the environmental power difference is greater than the standard environmental power difference threshold, mark the working state of the LED driving power supply as an abnormal working state; If the environmental power difference is less than or equal to the standard environmental power difference threshold, mark the working state of the LED driving power supply as a normal working state, and mark the historical working data in the normal working state as historical normal working data; The historical normal working data includes normal historical input voltage, normal historical input current, normal historical output voltage, normal historical output current, normal historical ambient temperature, and normal historical ambient humidity.

4. The intelligent debugging method for an LED driving power supply based on deep learning according to claim 3, wherein The process of obtaining the environmental power difference of the LED driving power supply in the working state includes: Mark the historical input voltage, historical input current, historical output voltage, historical output current, historical ambient temperature, and historical ambient humidity in the historical working data corresponding to each acquisition moment within the acquisition period respectively; Obtain the environmental power difference based on the marked historical input voltage, historical input current, historical output voltage, historical output current, historical ambient temperature, and historical ambient humidity.

5. The intelligent debugging method for an LED driving power supply based on deep learning according to claim 4, characterized in that The process of performing normalization processing on the historical normal working data to obtain the standard historical data includes: Number the historical normal working data, denoted as i = 1, 2, 3, ……, n; n is a natural number; Mark the normal historical input voltage, normal historical input current, normal historical output voltage, normal historical output current, normal historical ambient temperature, and normal historical ambient humidity respectively; Perform mean value calculations on the marked normal historical input voltage, normal historical input current, normal historical output voltage, normal historical output current, normal historical ambient temperature, and normal historical ambient humidity respectively to obtain the mean value of the historical normal working data; Obtain the standard deviation of the historical normal working data based on the historical normal working data and the mean value of the historical normal working data; Obtain the standard historical data based on the mean value of the historical normal working data and the standard deviation of the historical normal working data.

6. The intelligent debugging method for an LED driving power supply based on deep learning according to claim 5, wherein The process of constructing a deep learning power intelligent debugging model based on the standard historical data and then obtaining the power intelligent debugging index includes: Construct a standard deep learning power intelligent debugging model based on a convolutional neural network; Divide the obtained several groups of standard historical data into a training set, a validation set, and a test set, and input the training set, validation set, and test set into the standard deep learning power intelligent debugging model to train the standard deep learning power intelligent debugging model, obtain the trained standard deep learning power intelligent debugging model, and denote the trained standard deep learning power intelligent debugging model as the deep learning power intelligent debugging model; Obtain the power intelligent debugging index based on the deep learning power intelligent debugging model.

7. The intelligent debugging method for an LED driving power supply based on deep learning according to claim 6, wherein The process of performing intelligent debugging on the LED driver power supply based on the power intelligent debugging index and obtaining the optical performance index of the LED after intelligent debugging includes: Obtain the real-time working data of the LED driver power supply under different working environments in real time and mark the real-time working data; Obtain the optical performance index based on the marked real-time working data and the power intelligent debugging index.

8. The intelligent debugging method for an LED driving power supply based on deep learning according to claim 7, wherein The process of performing comparison and judgment on the optical performance index includes: Preset the standard optical performance index; If the optical performance index is greater than or equal to the standard optical performance index, the optical performance of the LED reaches the expected effect, indicating that the intelligent debugging of the LED driver power supply meets the requirements; If the optical performance index is less than the standard optical performance index, the optical performance of the LED does not reach the expected effect, indicating that the intelligent debugging of the LED driver power supply does not meet the requirements.

Images