CN117150891B - Intelligent prediction method and system for LED lamp bead power based on data driving - Google Patents

Intelligent prediction method and system for LED lamp bead power based on data driving Download PDF

Info

Publication number
CN117150891B
CN117150891B CN202311026862.7A CN202311026862A CN117150891B CN 117150891 B CN117150891 B CN 117150891B CN 202311026862 A CN202311026862 A CN 202311026862A CN 117150891 B CN117150891 B CN 117150891B
Authority
CN
China
Prior art keywords
target
power
lamp bead
correlation
beads
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202311026862.7A
Other languages
Chinese (zh)
Other versions
CN117150891A (en
Inventor
楼鸿玮
楼信满
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Migang New Materials Technology Shanghai Co ltd
Original Assignee
Migang New Materials Technology Shanghai Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Migang New Materials Technology Shanghai Co ltd filed Critical Migang New Materials Technology Shanghai Co ltd
Priority to CN202311026862.7A priority Critical patent/CN117150891B/en
Publication of CN117150891A publication Critical patent/CN117150891A/en
Application granted granted Critical
Publication of CN117150891B publication Critical patent/CN117150891B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R21/00—Arrangements for measuring electric power or power factor
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/27—Regression, e.g. linear or logistic regression
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G06N20/20—Ensemble learning
    • H—ELECTRICITY
    • H05—ELECTRIC TECHNIQUES NOT OTHERWISE PROVIDED FOR
    • H05B—ELECTRIC HEATING; ELECTRIC LIGHT SOURCES NOT OTHERWISE PROVIDED FOR; CIRCUIT ARRANGEMENTS FOR ELECTRIC LIGHT SOURCES, IN GENERAL
    • H05B45/00—Circuit arrangements for operating light-emitting diodes [LED]
    • H05B45/30—Driver circuits
    • H—ELECTRICITY
    • H05—ELECTRIC TECHNIQUES NOT OTHERWISE PROVIDED FOR
    • H05B—ELECTRIC HEATING; ELECTRIC LIGHT SOURCES NOT OTHERWISE PROVIDED FOR; CIRCUIT ARRANGEMENTS FOR ELECTRIC LIGHT SOURCES, IN GENERAL
    • H05B45/00—Circuit arrangements for operating light-emitting diodes [LED]
    • H05B45/50—Circuit arrangements for operating light-emitting diodes [LED] responsive to malfunctions or undesirable behaviour of LEDs; responsive to LED life; Protective circuits
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02B—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO BUILDINGS, e.g. HOUSING, HOUSE APPLIANCES OR RELATED END-USER APPLICATIONS
    • Y02B20/00—Energy efficient lighting technologies, e.g. halogen lamps or gas discharge lamps
    • Y02B20/40—Control techniques providing energy savings, e.g. smart controller or presence detection

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Software Systems (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Informatics (AREA)
  • Computing Systems (AREA)
  • Mathematical Analysis (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Optimization (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Computational Mathematics (AREA)
  • Evolutionary Biology (AREA)
  • Geometry (AREA)
  • Power Engineering (AREA)
  • Operations Research (AREA)
  • Probability & Statistics with Applications (AREA)
  • Computer Hardware Design (AREA)
  • Algebra (AREA)
  • Databases & Information Systems (AREA)
  • Circuit Arrangement For Electric Light Sources In General (AREA)

Abstract

The application relates to the field of electronic data processing, in particular to an intelligent LED lamp bead power prediction method and system based on data driving, comprising the following steps: the power information corresponding to the target lamp beads is used as a single calculation parameter to conduct power prediction, first prediction power corresponding to the target lamp beads is confirmed, then the area correlation degree is calculated based on the correlation between the target distribution area where the target lamp beads are located and the historical reference area, then the lamp bead correlation degree is calculated according to the correlation between the target lamp beads and other lamp beads in the target distribution area, second prediction power corresponding to the target lamp beads is calculated based on the area correlation degree and the lamp bead correlation degree, finally the first prediction power corresponding to the target lamp beads and the second prediction power are subjected to weight distribution through the accuracy weight corresponding to the target lamp beads, final prediction power corresponding to the target lamp beads is confirmed, and compared with a traditional power prediction mode, the accuracy of power prediction is greatly improved, and the working cost is reduced.

Description

Intelligent prediction method and system for LED lamp bead power based on data driving
Technical Field
The application relates to the field of electronic data processing, in particular to an intelligent LED lamp bead power prediction method and system based on data driving.
Background
An LED light bead is an electronic component that emits light using a semiconductor material. The LED lamp is a core component of an LED lamp, and is also a light source widely applied in the field of modern illumination. The working principle of the LED lamp bead is that electric energy is converted into light energy through electronic energy level structural change of a semiconductor material. When current passes through the LED beads, electrons and holes recombine in the semiconductor material, releasing energy, producing optical radiation. The color of the light emitted by the LED beads depends on the band structure of the semiconductor material used. LED light beads have many advantages including high efficiency, long life, low energy consumption, dimming, quick start, vibration resistance, etc. Compared with the traditional incandescent lamp and fluorescent lamp, the LED lamp bead is more energy-saving and environment-friendly, and can provide better light quality and color temperature control. The LED lamp beads are widely applied, and comprise indoor illumination, outdoor illumination, automobile illumination, display screens, backlight sources and the like. With the continuous progress of LED technology and the reduction of cost, LED beads have become the main stream light source in the lighting industry, and are widely used in various fields.
The power requirement of the LED lamp beads is an important silver color of the LED lamp beads serving various fields. The power requirements of an LED light bulb refer to the amount of electrical power required by the LED light bulb in a particular lighting application. And the power requirements of the LED light beads may vary significantly under different environmental conditions. For example, fluctuations in input voltage, changes in ambient temperature, different lighting requirements, etc. all affect the power requirements of the LED light beads. Therefore, in order to obtain the power requirement of the LED lamp bead more accurately, power prediction is required to be performed on the LED lamp bead. Traditional LED lamp pearl power prediction relies on the parameter and the loss of LED lamp pearl itself more to carry out artifical experience judgement, does not consider more objective factors, leads to traditional LED lamp pearl power prediction accuracy lower.
Disclosure of Invention
In view of the above, it is necessary to provide an intelligent prediction method and system for power of LED lamp beads based on data driving, which improves accuracy of power prediction of LED lamp beads and further reduces working cost of power prediction of LED lamp beads compared with the conventional prediction method for power of LED lamp beads.
The application provides an intelligent LED lamp bead power prediction method based on data driving, which is applied to the field of lamp bead power prediction, and comprises the following steps: taking the power information corresponding to the target lamp bead as a single calculation parameter to carry out power prediction, and confirming the first prediction power corresponding to the target lamp bead; calculating the region correlation degree based on the correlation between a target distribution region where the target lamp beads are located and a historical reference region, wherein the historical reference region is other distribution regions with the correlation with the target distribution region being larger than a preset correlation threshold; calculating the degree of correlation of the lamp beads according to the correlation of the target lamp beads and other lamp beads in the target distribution area; calculating second predicted power corresponding to the target lamp bead based on the region correlation degree and the lamp bead correlation degree; and carrying out weight distribution on the first predicted power and the second predicted power corresponding to the target lamp beads through the accuracy weight corresponding to the target lamp beads, and confirming the final predicted power corresponding to the target lamp beads.
In one embodiment, the calculating the area correlation degree based on the correlation between the target distribution area where the target lamp beads are located and the historical reference area, where the historical reference area is another distribution area with a correlation with the target distribution area greater than a preset correlation threshold specifically includes: calculating the power ratio value corresponding to the target distribution area and other distribution areas at each moment, and calculating the average value of the power ratio values of the target distribution area and other distribution areas at all moments of a target analysis sequence to confirm the difference of the areas, wherein the target analysis sequence is a power value sequence constructed by power values at preset quantity moments before the target lamp beads; comparing the regional difference with a preset correlation threshold, and taking other distribution regions larger than the preset correlation threshold as history reference regions; and calculating the ratio of the sum of the correlation between the target distribution area and each historical reference area to the sum of the correlation between every two historical reference areas, and confirming the area correlation degree.
In one embodiment, the calculating the power ratio value of the target distribution area corresponding to each time with the other distribution areas in the target analysis sequence, and determining the difference of the areas, where the target analysis sequence refers to a power value sequence constructed by power values of a preset number of times before the target lamp beads specifically includes:
wherein, For the region difference between the target distribution region where the target lamp bead a is located and the other distribution regions d, c a is the target distribution region where the target lamp bead a is located, d is the other distribution regions, L is the time length corresponding to the target analysis sequence, and Δlv (i,c,d) is the difference between the power ratio value of the target distribution region c a corresponding to the i-th time and the power ratio average value of the target distribution region c a and the other distribution region d at all times of the target analysis sequence.
In one embodiment, the calculating the degree of bead correlation according to the correlation between the target bead and other beads in the target distribution area specifically includes: calculating the power ratio of the target lamp bead to other lamp beads in the target distribution area at each moment, and calculating the average value of the power ratios of the target lamp bead to other lamp beads in the target distribution area at all moments of the target analysis sequence, so as to confirm the difference of the initial lamp beads; calculating the final lamp bead difference based on the curve correlation of the initial lamp bead difference and other lamp beads corresponding to the target distribution area; and confirming the degree of correlation of the lamp beads according to the proportion of the final lamp bead difference in the target distribution area.
In one embodiment, the calculating the power ratio of the target bead to other beads in the target distribution area at each moment, and the average value of the power ratios of the target bead to other beads in the target distribution area at all moments in the target analysis sequence, to confirm the initial bead difference specifically includes:
Co (a,w) is the initial bead difference between the target bead a and the w other beads in the target distribution area, and DeltaGv (i,a,w) is the difference between the power ratio of the target bead a to the w other beads in the target distribution area at the i-th moment and the average value of the power ratios of the target bead a to the other beads in the target distribution area at all moments in the target analysis sequence.
In one embodiment, the calculating the final bead difference based on the curve correlation between the initial bead difference and other beads in the target distribution area specifically includes:
Wherein Con' (a,w) is the final bead difference between the target bead a and the w other beads in the target distribution area, DTW "j is the curve correlation between the j-th referent sequence of the other beads w in the target distribution area and the corresponding curve, x is the number of referent sequences, and Co ((d,a,w),j) is the initial bead difference between the target bead a and the w other beads in the target distribution area in the j-th referent sequence.
In one embodiment, the calculating the second predicted power corresponding to the target lamp bead based on the region correlation degree and the lamp bead correlation degree specifically includes:
wherein Pe is the second predicted power corresponding to the target lamp bead a, For the region difference between the target distribution region c a where the target lamp bead a is located and other distribution regions d, con' (a,w) is the final lamp bead difference between the target lamp bead a and the w other lamp beads in the target distribution region, q is the number of historical reference regions, r is the number of lamp beads in the target distribution region c a, and the predicted power value is obtained according to the power prediction of the w other lamp beads in the d historical reference region.
In one embodiment, the weight distribution is performed on the first predicted power and the second predicted power corresponding to the target lamp bead according to the accuracy weight corresponding to the target lamp bead, and the determining the final predicted power corresponding to the target lamp bead specifically includes: predicting an ideal power value corresponding to the target lamp bead according to the current value and the voltage value corresponding to the referent sequence; calculating an accuracy weight corresponding to the target lamp bead based on the difference value between the ideal power value and the acquired actual power value; and inputting the accuracy weight, the first predicted power and the second predicted power corresponding to the target lamp beads into a preset calculation formula, and confirming the final predicted power corresponding to the target lamp beads.
In one embodiment, the calculating the accuracy weight corresponding to the target lamp bead based on the difference between the ideal power value and the acquired actual power value specifically includes:
Wherein, pre a is the accuracy weight corresponding to the target lamp bead a, ΔPr (j,a) is the difference between the ideal power value of the j-th referenceable data corresponding to the target lamp bead a and the acquired actual power value; correspondingly, inputting the accuracy weight, the first predicted power and the second predicted power corresponding to the target lamp bead into a preset calculation formula, and confirming the final predicted power corresponding to the target lamp bead, wherein the method specifically comprises the following steps:
P＝Prea*P0+(1-Prea)*Pe
Wherein, P is the final predicted power corresponding to the target lamp bead a, pre a is the accuracy weight corresponding to the target lamp bead a, P 0 is the first predicted power, and Pe is the second predicted power.
The second aspect of the application provides an intelligent LED lamp bead power prediction system based on data driving, which is applied to the field of lamp bead power prediction and comprises: the confirming module is used for carrying out power prediction by taking the power information corresponding to the target lamp bead as a single calculation parameter and confirming the first predicted power corresponding to the target lamp bead; the first calculation module is used for calculating the area correlation degree based on the correlation between the target distribution area where the target lamp beads are located and the historical reference area, wherein the historical reference area is other distribution areas with the correlation with the target distribution area being larger than a preset correlation threshold; the second calculating module is used for calculating the correlation degree of the lamp beads according to the correlation between the target lamp beads and other lamp beads in the target distribution area; the determining module is used for calculating second predicted power corresponding to the target lamp bead based on the region correlation degree and the lamp bead correlation degree; and the prediction module is used for carrying out weight distribution on the first predicted power and the second predicted power corresponding to the target lamp beads through the accuracy weight corresponding to the target lamp beads, and confirming the final predicted power corresponding to the target lamp beads.
According to the embodiment of the application, power information corresponding to the target lamp beads is used as a single calculation parameter to carry out power prediction, first prediction power corresponding to the target lamp beads is confirmed, then the correlation degree of the region is calculated based on the correlation between a target distribution region where the target lamp beads are located and a historical reference region, wherein the historical reference region is other distribution regions with the correlation with the target distribution region being larger than a preset correlation threshold value, and then the correlation degree of the lamp beads is calculated according to the correlation between the target lamp beads and other lamp beads in the target distribution region; and calculating second predicted power corresponding to the target lamp bead based on the region correlation degree and the lamp bead correlation degree, and finally carrying out weight distribution on the first predicted power and the second predicted power corresponding to the target lamp bead through the accuracy weight corresponding to the target lamp bead to confirm final predicted power corresponding to the target lamp bead. The calculation parameters of the lamp bead power prediction are used as the calculation parameters of the lamp bead power prediction through the region correlation degree of the target distribution region where the target lamp bead is located and other corresponding distribution regions and the lamp bead correlation degree between the target lamp bead and other corresponding lamp beads, and compared with the traditional mode of power prediction by means of experience values, the accuracy of power prediction is greatly improved, and the working cost of power prediction is reduced.
Drawings
Fig. 1 is a schematic flow chart of an intelligent prediction method for power of an LED lamp bead based on data driving according to an embodiment of the present application.
Fig. 2 is a schematic diagram of a first sub-flow of an intelligent prediction method for LED lamp bead power based on data driving according to an embodiment of the present application.
Fig. 3 is a schematic diagram of a second sub-flow of an intelligent prediction method for LED lamp bead power based on data driving according to an embodiment of the present application.
Fig. 4 is a schematic diagram of a third sub-flow of an intelligent prediction method for LED lamp bead power based on data driving according to an embodiment of the present application.
Fig. 5 is a block schematic diagram of an intelligent prediction system for LED lamp bead power based on data driving according to an embodiment of the present application.
Fig. 6 is a schematic diagram of a preferred embodiment of an intelligent prediction method for power of LED lamp beads based on data driving according to an embodiment of the present application.
Fig. 7 is a schematic diagram of a first calculation formula of an intelligent prediction method for power of an LED lamp bead based on data driving according to an embodiment of the present application.
Fig. 8 is a schematic diagram of a second calculation formula of an intelligent prediction method for power of an LED lamp bead based on data driving according to an embodiment of the present application.
Fig. 9 is a schematic diagram of a third calculation formula of an intelligent prediction method for power of an LED lamp bead based on data driving according to an embodiment of the present application.
Fig. 10 is a schematic diagram of a fourth calculation formula of an intelligent prediction method for power of an LED lamp bead based on data driving according to an embodiment of the present application.
Fig. 11 is a schematic diagram of a fifth calculation formula of an intelligent prediction method for power of an LED lamp bead based on data driving according to an embodiment of the present application.
Fig. 12 is a schematic diagram of a sixth calculation formula of an intelligent prediction method for power of LED lamp beads based on data driving according to an embodiment of the present application.
Fig. 13 is a schematic diagram of a seventh calculation formula of an intelligent prediction method for power of an LED lamp bead based on data driving according to an embodiment of the present application.
Fig. 14 is a schematic diagram of an eighth calculation formula of an intelligent prediction method for power of LED lamp beads based on data driving according to an embodiment of the present application.
Detailed Description
In describing embodiments of the present application, words such as "exemplary," "or," "such as," and the like are used to mean serving as an example, instance, or illustration. Any embodiment or design described herein as "exemplary" or "for example" should not be construed as preferred or advantageous over other embodiments or designs. Rather, the use of words such as "exemplary," "or," "such as," and the like are intended to present related concepts in a concrete fashion.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in the description of the application herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. It is to be understood that, unless otherwise indicated, a "/" means or. For example, A/B may represent A or B. The "and/or" in the present application is merely one association relationship describing the association object, indicating that three relationships may exist. For example, a and/or B may represent: a exists alone, A and B exist simultaneously, and B exists alone. "at least one" means one or more. "plurality" means two or more than two. For example, at least one of a, b or c may represent: seven cases of a, b, c, a and b, a and c, b and c, a, b and c.
It should be further noted that the terms "first" and "second" in the description and claims of the present application and the accompanying drawings are used for respectively similar objects, and are not used for describing a specific order or sequence. The method disclosed in the embodiments of the present application or the method shown in the flowchart, including one or more steps for implementing the method, may be performed in an order that the steps may be interchanged with one another, and some steps may be deleted without departing from the scope of the claims.
The embodiment of the application firstly provides an intelligent LED lamp bead power prediction method based on data driving, which is applied to the field of lamp bead power prediction, and referring to the attached figure 1, the method comprises the following steps:
S101, power prediction is carried out by taking power information corresponding to a target lamp bead as a single calculation parameter, and first prediction power corresponding to the target lamp bead is confirmed.
The step of performing power prediction by taking the power information corresponding to the target lamp bead as a single calculation parameter refers to performing subsequent power demand prediction on the target lamp bead by considering only one factor of power information change corresponding to the target lamp bead. Specifically, the collected historical power information of the target lamp bead can be processed according to an exponential smoothing algorithm, so as to obtain a first predicted power corresponding to the target lamp bead.
It should be noted that, the exponential smoothing algorithm is a commonly used time series prediction algorithm for smoothing and predicting time series data. The exponential smoothing algorithm calculates future predictions by weighted averaging past observations based on the principle of weighted averaging. It assumes that there is an exponentially decaying relationship between future values and past values, i.e., the closer observations are weighted the more and the farther observations are weighted the less. The basic formula of the exponential smoothing algorithm is:
F(t+1＝k*Y(t-1-k)*Y(t)
wherein F (t+1) represents a predicted value at time t+1, Y (t) represents an actual observed value at time t, F (t) represents a predicted value at time t, k represents a smoothing coefficient (smoothing factor), and the value range is 0 to 1.
The exponential smoothing algorithm steps are as follows:
1. initializing: an initial smoothing coefficient k and an initial predicted value F (0) are selected.
2. Updating the predicted value: from the current observation YY (t) and the previous prediction F (t), a new prediction FF (t+1) is calculated by weighted averaging.
3. Updating the smoothing coefficient: the weights of the past observations and the recent observations are balanced by adjusting the magnitude of k based on the historical error (the difference between the actual observations and the predicted values) and the previous smoothing coefficient k.
4. And (3) repeating the step (2) and the step (3) until the predicted value converges or reaches the specified iteration times.
The advantage of the exponential smoothing algorithm is that it is simple, fast, easy to understand and implement. It is applicable to data without obvious trends and seasonality, and can be used for short-term prediction and smoothing data. Different exponential smoothing algorithm variants, such as simple exponential smoothing, double exponential smoothing, and three exponential smoothing, etc., may also be used to better accommodate different data patterns and trends, depending on the specific data characteristics and prediction requirements. The collected historical power information of the target lamp beads is processed according to an exponential smoothing algorithm to obtain first predicted power corresponding to the target lamp beads, the first predicted power can be substituted correspondingly according to the steps, and the scheme of the specific steps is not limited further.
S102, calculating the area correlation degree based on the correlation between the target distribution area where the target lamp beads are located and a historical reference area, wherein the historical reference area is other distribution areas with the correlation with the target distribution area being larger than a preset correlation threshold value.
The target distribution area refers to a distribution area where the target beads are located, the distribution area of the beads refers to area division according to row and column distribution conditions of the beads on a PCB board, and referring to fig. 6, the PCB board includes m×n beads, and the equal area division of a preset number is performed on each row of the beads, and each area includes a preset number of the beads. The history reference area is a history reference area defined by calculating the correlation according to a target distribution area where the target lamp beads are located and other distribution areas in the same PCB, and the correlation with the target distribution area is larger than a preset correlation threshold. The area correlation degree refers to the intensity or tightness degree of the relation between the target distribution area where the target lamp beads are located and the historical reference area.
It should be noted that, the correlation is a statistical index for measuring the relationship between two variables. Common calculation methods include correlation coefficients and covariance. Correlation coefficient: the correlation coefficient measures the strength of the linear correlation degree between two variables, and the common correlation coefficients are a pearson correlation coefficient and a spearman correlation coefficient. The pearson correlation coefficient applies to continuous variables and the spearman correlation coefficient applies to sequential variables or non-linear relationships. The correlation coefficient has a value ranging from-1 to 1, a value near 1 indicating positive correlation, a value near-1 indicating negative correlation, and a value near 0 indicating no correlation. Covariance: covariance measures the relationship of the overall trend of change between two variables, and the overall covariance can be estimated by the sample covariance. The covariance has a value ranging from negative infinity to positive infinity, positive values for positive correlations, negative values for negative correlations, and 0 for no correlations.
S103, calculating the degree of correlation of the lamp beads according to the correlation of the target lamp beads and other lamp beads in the target distribution area.
The correlation between the target lamp bead and other lamp beads in the target distribution area refers to the representation of the power relation of the target lamp bead and other lamp beads in the target distribution area, and it is noted that the size relation between the lamp beads in the same distribution area is influenced and has a relation. The degree of bead correlation refers to the intensity or tightness of the relationship between the target bead and other beads in the target distribution area.
It should be noted that, the correlation between the target lamp bead and other lamp beads in the target distribution area is calculated by the historical referent power information of the total LED lamp corresponding to the target lamp bead. The obtaining of the history referenceable power information of the total LED lamp may be the following steps: firstly, a power time curve eta 0 of the LED lamp is drawn, and the length of the curve is 50. Wherein the abscissa indicates time and the ordinate indicates total power of all the beads in the LED lamp. The historical data LED lamp is drawn into a power time curve eta, the length of the curve is L, and the L is 1000 in the invention. And sliding on the curve eta by taking the curve eta 0 as a template and taking the time t=1 as a step length, and calculating the similarity between the template curve and the window curve during each sliding, wherein the similarity is acquired by using a DTW algorithm, and the acquired similarity is marked as DTW. Meanwhile, in order to avoid that adjacent sections of one section of curve are similar curves, non-maximum suppression is adopted for the obtained similarity DTW. And finally, screening according to the size of the similarity DTW, for example, marking a window curve which is larger than a preset threshold value, so as to be used as a referenceable sequence curve, namely historical referenceable power information of the total LED lamp.
And S104, calculating second predicted power corresponding to the target lamp bead based on the region correlation degree and the lamp bead correlation degree.
And calculating the second predicted power corresponding to the target lamp bead based on the region correlation degree and the lamp bead correlation degree, wherein the calculation of the second predicted power corresponding to the target lamp bead is to further predict and obtain the second predicted power corresponding to the target lamp bead by taking the region correlation degree and the lamp bead correlation degree as calculation parameters. Compared with the first predicted power, the second predicted power has the advantages that the calculated parameters are more than the correlation degree between the lamp beads and the correlation degree between the areas where the lamp beads are located, namely, the consideration factors are more than the first predicted power, and the corresponding accuracy is higher. Specifically, when the correlation between the target analysis area where the target lamp bead is located and other relevant areas is stronger, the correlation between the second predicted power corresponding to the target lamp bead is also stronger, that is, the second predicted power obtained by deleting the target analysis area and other relevant areas is closer to the actual power result, that is, the obtained predicted power has higher accuracy.
And S105, carrying out weight distribution on the first predicted power and the second predicted power corresponding to the target lamp beads through the accuracy weight corresponding to the target lamp beads, and confirming the final predicted power corresponding to the target lamp beads.
The accuracy weight is a weight calculated according to the corresponding information of the target lamp bead, and the purpose of the weight is to reasonably distribute the first predicted power and the second predicted power corresponding to the target lamp bead so as to obtain the final predicted power corresponding to the target lamp bead.
It should be noted that, the power prediction of the target lamp bead can be doped with a plurality of objective factors and self lamp bead factors, so that the power influence caused by each factor is reasonably distributed, the final predicted power is comprehensively obtained, and the power prediction accuracy of the target lamp bead can be further increased. And the factors to be considered in power prediction can include the following: 1. based on specifications and technical parameters: first, the power may be estimated according to specifications and technical parameters of the LED lamp. This includes the power rating, the light efficiency (lumens/watt) and the brightness level of the LED lamp. By knowing the specifications and technical parameters of the lamp, the power output of the lamp can be estimated initially. 2. Based on the measured data and the history: the actual LED lamp power data and the historical record are collected and analyzed, and the power output of the LED lamp can be predicted more accurately. This may be done by monitoring and recording the power consumption of the LED lamp, or using a power monitoring device to monitor the power consumption in real time. 3. Consider a usage scenario and a control strategy: different usage scenarios and control strategies can have an impact on the power requirements of the LED lamp. For example, brightness adjustment, switching status, dimming control, etc. of the light fixture can affect the power output. Therefore, these factors need to be considered when predicting the power of the LED lamp, and analyzed in connection with actual use. Further, machine learning and statistical models are used: machine learning and statistical models can be utilized to predict the power output of the LED lamp. These models may learn and build relationships between power output and input characteristics based on historical data and other relevant factors. For example, regression analysis, support vector machines, random forests, etc. may be used to construct the predictive model. The power prediction model is made by referring to the influence factors, and specific common steps are not further limited and can be recorded by referring to the prior art.
According to the embodiment of the application, power information corresponding to the target lamp beads is used as a single calculation parameter to carry out power prediction, first prediction power corresponding to the target lamp beads is confirmed, then the correlation degree of the region is calculated based on the correlation between a target distribution region where the target lamp beads are located and a historical reference region, wherein the historical reference region is other distribution regions with the correlation with the target distribution region being larger than a preset correlation threshold value, and then the correlation degree of the lamp beads is calculated according to the correlation between the target lamp beads and other lamp beads in the target distribution region; and calculating second predicted power corresponding to the target lamp bead based on the region correlation degree and the lamp bead correlation degree, and finally carrying out weight distribution on the first predicted power and the second predicted power corresponding to the target lamp bead through the accuracy weight corresponding to the target lamp bead to confirm final predicted power corresponding to the target lamp bead. The calculation parameters of the lamp bead power prediction are used as the calculation parameters of the lamp bead power prediction through the region correlation degree of the target distribution region where the target lamp bead is located and other corresponding distribution regions and the lamp bead correlation degree between the target lamp bead and other corresponding lamp beads, and compared with the traditional mode of power prediction by means of experience values, the accuracy of power prediction is greatly improved, and the working cost of power prediction is reduced.
In one embodiment of the present application, and referring to fig. 2, the step S102: calculating the region correlation degree based on the correlation between the target distribution region where the target lamp beads are located and a history reference region, wherein the history reference region is other distribution regions with the correlation with the target distribution region being larger than a preset correlation threshold, and the method specifically comprises the following steps:
S201, calculating the power ratio of the target distribution area to other distribution areas at each moment, and calculating the average value of the power ratio of the target distribution area to the power ratio of the other distribution areas at all moments of a target analysis sequence to confirm the difference of the areas, wherein the target analysis sequence is a power value sequence constructed by power values at preset quantity moments in front of a target lamp bead.
The target distribution area refers to a distribution area where the target lamp beads are located, and the other distribution areas refer to other distribution areas except the target distribution area in the PCB where the target lamp beads are located. The target analysis sequence refers to a power value sequence constructed by power values at preset number of moments before the target lamp beads, and the target analysis sequence can also be a power value sequence which is obtained after the corresponding similarity screening and is more relevant to the target lamp beads.
Specifically, referring to fig. 7, the difference calculation is performed between the power ratio of the target distribution area corresponding to each time and the power ratio average of the target distribution area corresponding to each time of the other distribution areas in the target analysis sequence, and the difference of the areas is determined, where the target analysis sequence refers to a power value sequence constructed by power values at preset number of times before the target lamp beads, and specifically includes:
wherein, For the region difference between the target distribution region where the target lamp bead a is located and the other distribution regions d, c a is the target distribution region where the target lamp bead a is located, d is the other distribution regions, L is the time length corresponding to the target analysis sequence, and Δlv (i,c,d) is the difference between the power ratio value of the target distribution region c a corresponding to the i-th time and the power ratio average value of the target distribution region c a and the other distribution region d at all times of the target analysis sequence.
S202, comparing the regional difference with a preset correlation threshold, and taking other distribution regions larger than the preset correlation threshold as history reference regions.
After the regional difference between the target distribution region and other distribution regions is obtained, the corresponding regional difference is compared with a preset correlation threshold value, and other distribution regions larger than the preset correlation threshold value are used as history reference regions. The history reference area is an area with strong correlation with the target distribution area, and performs power prediction of the target lamp beads based on the lamp bead power condition of the distribution area with strong correlation, so that the accuracy of power prediction can be further increased.
S203, calculating the ratio of the sum of the correlation between the target distribution area and each historical reference area to the sum of the correlation between every two historical reference areas, and confirming the area correlation degree.
After the historical reference areas are confirmed, calculating the correlation between the target distribution area and each historical reference area, counting the sum of the correlation between the target distribution area and each historical reference area, calculating the correlation between every two areas of all the historical reference areas, counting the sum of the correlation between every two areas of all the historical reference areas, and calculating the ratio of the two areas to obtain the area correlation range.
In one embodiment of the present application, referring to fig. 3, the calculating the degree of correlation between the target bead and other beads in the target distribution area according to the correlation between the target bead and other beads specifically includes:
And S301, calculating the power ratio of the target lamp bead to other lamp beads in the target distribution area at each moment, and calculating the average value of the power ratio of the target lamp bead to other lamp beads in the target distribution area at all moments of the target analysis sequence, so as to confirm the difference of the initial lamp beads.
And calculating the power value corresponding to the target lamp bead at each moment, simultaneously calculating the power values corresponding to other lamp beads in the target distribution area at the moment, calculating the ratio of the power values to each other to obtain the power ratio corresponding to the target lamp bead and the other lamp beads in the target distribution area at each moment, calculating the average value of the power ratios of the target lamp bead and the other lamp beads in the target distribution area at all moments of the target analysis sequence, and calculating the difference between the ratio of the target lamp bead and the average value of the ratio of the target lamp bead to obtain the difference of the initial lamp beads. The initial bead difference refers to the difference or variation degree between the target bead and other beads in the target distribution area.
Specifically, referring to fig. 8, the calculating the power ratio of the target bead corresponding to other beads in the target distribution area at each moment, and the average value of the power ratios of the target bead and other beads in the target distribution area at all moments in the target analysis sequence, to confirm the initial bead difference specifically includes:
Co (a,w) is the initial bead difference between the target bead a and the w other beads in the target distribution area, and DeltaGv (i,a,w) is the difference between the power ratio of the target bead a to the w other beads in the target distribution area at the i-th moment and the average value of the power ratios of the target bead a to the other beads in the target distribution area at all moments in the target analysis sequence.
S302, calculating the final lamp bead difference based on the curve correlation of the initial lamp bead difference and other lamp beads in the target distribution area.
The other beads in the target distribution area correspond to a curve correlation, and the curve correlation refers to whether the relationship between the two variables of the target bead and the other beads in the target distribution area shows a curve type correlation. In statistics, a common approach is to measure the linear correlation between variables by a correlation coefficient, however, sometimes the relationship between variables is not a simple linear relationship, but rather exhibits a curvilinear form of correlation. Typically, the curve correlation can be obtained by some common calculation algorithm, specifically: 1. collecting data: first, relevant data needs to be collected, including observations of both variables. The quality and accuracy of the data are ensured. 2. Drawing a scatter diagram: observations of two variables are plotted on a scatter plot, with the horizontal axis representing one variable and the vertical axis representing the other. The form of the relationship between the variables can be preliminarily observed through the scatter diagram. 3. Observing the curve form: the relationship morphology between the variables is observed on the scatter plot. Based on the shape of the scattergram, it can be determined whether or not there is a curve correlation, such as a positive correlation curve, a negative correlation curve, and a U-shaped or inverted U-shaped curve. 4. Fitting a curve: if curve correlation is observed, a suitable method may be used to fit the curve. Common methods include polynomial regression, nonlinear regression, or other curve fitting methods. These methods may estimate parameters of the curve by techniques such as least squares. 5. Calculating a correlation coefficient: after fitting the curve, correlation coefficients may be calculated to quantify the strength of the curve correlation. Common correlation coefficients are pearson correlation coefficients, spearman correlation coefficients, and the like. The correlation coefficient has a value ranging from-1 to 1, a value close to-1 indicating a negative correlation, a value close to 1 indicating a positive correlation, and a value close to 0 indicating no correlation. 6. And (3) checking statistical significance: after calculating the correlation coefficients, a statistical significance test may be performed to determine if the curve correlation is significant. Common methods include calculating a p-value or confidence interval to determine the significance level of the correlation coefficient. The subsequent analysis of the specific calculation method in the individual embodiments is not repeated here.
Specifically, referring to fig. 9, the calculating the final bead difference based on the curve correlation between the initial bead difference and other beads in the target distribution area specifically includes:
Wherein Con' (a,w) is the final bead difference between the target bead a and the w other beads in the target distribution area, DTW "j is the curve correlation between the j-th referent sequence of the other beads w in the target distribution area and the corresponding curve, x is the number of referent sequences, and Co ((d,a,w),j) is the initial bead difference between the target bead a and the w other beads in the target distribution area in the j-th referent sequence.
S303, confirming the degree of correlation of the lamp beads according to the proportion of the final lamp bead difference in the target distribution area.
After the final bead difference is obtained, the proportion of the final bead difference in the target distribution area is calculated to further confirm the bead correlation degree.
In an embodiment of the present application, referring to fig. 10, in step S104, the calculating, based on the region correlation degree and the bead correlation degree, the second predicted power corresponding to the target bead specifically includes:
wherein Pe is the second predicted power corresponding to the target lamp bead a, For the region difference between the target distribution region c a where the target lamp bead a is located and other distribution regions d, con' (a,w) is the final lamp bead difference between the target lamp bead a and the w other lamp beads in the target distribution region, q is the number of historical reference regions, r is the number of lamp beads in the target distribution region c a, and the predicted power value is obtained according to the power prediction of the w other lamp beads in the d historical reference region.
It should be noted that the data variability refers to the difference or degree of variation between different data in the data set. In LED lamp power prediction, data variability can be manifested in several aspects: 1. variability of power data: there may be a large difference in power output of the LED lamp, i.e. the power values of different lamps may be different in the same use situation. Such variability may result from factors such as the specifications, technical parameters, branding, manufacturing process, etc. of the different fixtures. 2. Time difference: the power output of an LED lamp may vary over time. For example, the lamp may have a higher power output when it is just turned on, and the power output may gradually decrease as the usage time increases. Therefore, in power prediction, time variability needs to be considered and time factors are included in the prediction model. 3. Environmental variability: the power output of an LED lamp may vary due to environmental factors. For example, environmental factors such as temperature, humidity, illumination, etc. may affect the power requirements and output of the luminaire. Therefore, in performing power prediction, it is necessary to consider the environmental variability and to incorporate environmental factors into the prediction model. 4. Use context variability: there may be differences in the power requirements of the LED lamp under different usage scenarios. For example, different locations, different use fixtures may have different demands for power. Therefore, in power prediction, it is necessary to consider the difference in the use situation and to incorporate the use situation factor into the prediction model. In summary, data variability is an important consideration in LED lamp power prediction. By knowing and analyzing the specifications, technical parameters, brands and manufacturing processes of different lamps and considering factors such as time, environment and use situation, the power output of the LED lamp can be predicted more accurately. In this embodiment, the difference between the lamp beads and the difference between the distribution areas of the lamp beads are used as a calculation parameter of power prediction, that is, part of the factors of the above factors are integrated, so that the accuracy of subsequent power prediction can be further increased.
In some possible embodiments, referring to fig. 11, the second predicted power needs to be adjusted by an error, and in an actual circuit, the detected power value may not be completely equal to the product of the voltage and the current value due to measurement errors, non-linear characteristics of circuit elements, non-idealities of the circuit, and the like. These factors may lead to errors or deviations in the measurement of power. Therefore, although power can be obtained from the current-voltage detection result, an error needs to be considered. The present embodiment uses the correlation between the real-time detected current and voltage product and the real-time detected power value to represent the error:
where x represents the number of referenceable history curves based on power draw, L represents the template length, The power value of the current analysis lamp bead is calculated by representing the current voltage value obtained at the ith moment in the jth referenceable data, and P j,i represents the power value obtained according to the api data collection at the ith moment in the jth referenceable data. That is, the closer the product of the current and the voltage value in the acquired data is to the corresponding acquired power, the more accurate the prediction result obtained based on the current and the voltage value is.
The voltage time series and the current time series are obtained by using the method. And acquiring corresponding referenceable historical data, and simultaneously acquiring correlations among different lamp beads corresponding to the corresponding voltage time sequence, and regional correlations. And predicting the voltage and current data of each lamp bead by using the exponential smoothing algorithm, acquiring a prediction result of the lamp bead to be analyzed based on the region correlation and the correlation among the lamp beads, and marking the prediction result as U 'a、I′a, wherein the power prediction result obtained according to the current and voltage value is U' a*I′a.
Further, referring to fig. 12, the power prediction value obtained based on the analysis of the other lamp bead data is finally obtained by analyzing the prediction result obtained based on the other current and voltage data information:
Where Com ' represents a predicted power error obtained based on the current-voltage data, P ' a represents the above-obtained adjusted second predicted power, U ' a*I′a represents a predicted power value of the current lamp bead obtained based on predictive analysis of other lamp bead current-voltage information, and Pe is the final second predicted power obtained after adjustment.
In an embodiment of the present application, referring to fig. 4, in step S105, the step of determining the final predicted power corresponding to the target lamp bead by weight distribution of the first predicted power and the second predicted power corresponding to the target lamp bead according to the accuracy weight corresponding to the target lamp bead specifically includes:
S401, predicting an ideal power value corresponding to a target lamp bead according to a current value and a voltage value corresponding to a referent sequence;
s402, calculating an accuracy weight corresponding to the target lamp bead based on the difference value between the ideal power value and the acquired actual power value.
Specifically, referring to fig. 13, the calculating the accuracy weight corresponding to the target lamp bead based on the difference between the ideal power value and the acquired actual power value specifically includes:
Wherein, pre a is the accuracy weight corresponding to the target lamp bead a, ΔPr (j,a) is the difference between the ideal power value of the j-th referenceable data corresponding to the target lamp bead a and the actual power value acquired.
S403, inputting the accuracy weight value, the first predicted power and the second predicted power corresponding to the target lamp bead into a preset calculation formula, and confirming the final predicted power corresponding to the target lamp bead.
Specifically, referring to fig. 14, the inputting the accuracy weight, the first predicted power and the second predicted power corresponding to the target lamp bead into a preset calculation formula to determine the final predicted power corresponding to the target lamp bead specifically includes:
P＝Prea*P0+(1-Prea)*Pe
Wherein, P is the final predicted power corresponding to the target lamp bead a, pre a is the accuracy weight corresponding to the target lamp bead a, P 0 is the first predicted power, and Pe is the second predicted power.
According to the embodiment of the application, power information corresponding to the target lamp beads is used as a single calculation parameter to carry out power prediction, first prediction power corresponding to the target lamp beads is confirmed, then the correlation degree of the region is calculated based on the correlation between a target distribution region where the target lamp beads are located and a historical reference region, wherein the historical reference region is other distribution regions with the correlation with the target distribution region being larger than a preset correlation threshold value, and then the correlation degree of the lamp beads is calculated according to the correlation between the target lamp beads and other lamp beads in the target distribution region; and calculating second predicted power corresponding to the target lamp bead based on the region correlation degree and the lamp bead correlation degree, and finally carrying out weight distribution on the first predicted power and the second predicted power corresponding to the target lamp bead through the accuracy weight corresponding to the target lamp bead to confirm final predicted power corresponding to the target lamp bead. The calculation parameters of the lamp bead power prediction are used as the calculation parameters of the lamp bead power prediction through the region correlation degree of the target distribution region where the target lamp bead is located and other corresponding distribution regions and the lamp bead correlation degree between the target lamp bead and other corresponding lamp beads, and compared with the traditional mode of power prediction by means of experience values, the accuracy of power prediction is greatly improved, and the working cost of power prediction is reduced.
In another embodiment of the present application, referring to fig. 5, an embodiment of the present application firstly proposes an intelligent prediction system for predicting LED lamp bead power based on data driving, which is applied to the field of lamp bead power prediction, and the system includes:
the confirming module 1 is used for carrying out power prediction by taking the power information corresponding to the target lamp bead as a single calculation parameter and confirming the first predicted power corresponding to the target lamp bead;
The first calculating module 2 is configured to calculate a region correlation degree based on a correlation between a target distribution region where the target lamp beads are located and a historical reference region, where the historical reference region is another distribution region whose correlation with the target distribution region is greater than a preset correlation threshold;
the second calculating module 3 is configured to calculate a degree of correlation between the target bead and other beads in the target distribution area according to the correlation between the target bead and the other beads;
the determining module 4 is configured to calculate a second predicted power corresponding to the target lamp bead based on the region correlation degree and the lamp bead correlation degree;
And the prediction module 5 is used for carrying out weight distribution on the first predicted power and the second predicted power corresponding to the target lamp beads through the accuracy weight corresponding to the target lamp beads, and confirming the final predicted power corresponding to the target lamp beads.
According to the embodiment of the application, power prediction is carried out by taking power information corresponding to the target lamp beads as a single calculation parameter through a confirmation module, first prediction power corresponding to the target lamp beads is confirmed, then the correlation degree of the region is calculated through a first calculation module based on the correlation between a target distribution region where the target lamp beads are located and a historical reference region, wherein the historical reference region is other distribution regions with the correlation larger than a preset correlation threshold, then the correlation degree of the lamp beads is calculated through a second calculation module according to the correlation between the target lamp beads and other lamp beads in the target distribution region, second prediction power corresponding to the target lamp beads is calculated through a determination module based on the correlation degree of the region and the lamp beads, and finally the final prediction power corresponding to the target lamp beads is confirmed by carrying out weight distribution on the first prediction power corresponding to the target lamp beads and the second prediction power through an accuracy weight corresponding to the target lamp beads based on the prediction module. The calculation parameters of the lamp bead power prediction are used as the calculation parameters of the lamp bead power prediction through the region correlation degree of the target distribution region where the target lamp bead is located and other corresponding distribution regions and the lamp bead correlation degree between the target lamp bead and other corresponding lamp beads, and compared with the traditional mode of power prediction by means of experience values, the accuracy of power prediction is greatly improved, and the working cost of power prediction is reduced.
The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. In the description corresponding to the flowcharts and block diagrams in the figures, operations or steps corresponding to different blocks may also occur in different orders than that disclosed in the description, and sometimes no specific order exists between different operations or steps. For example, two consecutive operations or steps may actually be performed substantially in parallel, they may sometimes be performed in reverse order, which may be dependent on the functions involved. Each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
It will be evident to those skilled in the art that the application is not limited to the details of the foregoing illustrative embodiments, and that the present application may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. The above-described embodiments of the application are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the application being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.

Claims (10)

1. The intelligent LED lamp bead power prediction method based on data driving is applied to the field of lamp bead power prediction, and is characterized by comprising the following steps of:
taking the power information corresponding to the target lamp bead as a single calculation parameter to carry out power prediction, and confirming the first prediction power corresponding to the target lamp bead;
Calculating the region correlation degree based on the correlation between a target distribution region where the target lamp beads are located and a historical reference region, wherein the historical reference region is other distribution regions with the correlation with the target distribution region being larger than a preset correlation threshold;
Calculating the degree of correlation of the lamp beads according to the correlation of the target lamp beads and other lamp beads in the target distribution area;
calculating second predicted power corresponding to the target lamp bead based on the region correlation degree and the lamp bead correlation degree;
And carrying out weight distribution on the first predicted power and the second predicted power corresponding to the target lamp beads through the accuracy weight corresponding to the target lamp beads, and confirming the final predicted power corresponding to the target lamp beads.
2. The intelligent prediction method of LED lamp bead power based on data driving according to claim 1, wherein the calculating the area correlation degree is based on the correlation between the target distribution area where the target lamp bead is located and a history reference area, wherein the history reference area is another distribution area with a correlation with the target distribution area greater than a preset correlation threshold, specifically includes:
Calculating the power ratio value corresponding to the target distribution area and other distribution areas at each moment, and calculating the average value of the power ratio values of the target distribution area and other distribution areas at all moments of a target analysis sequence to confirm the difference of the areas, wherein the target analysis sequence is a power value sequence constructed by power values at preset quantity moments before the target lamp beads;
comparing the regional difference with a preset correlation threshold, and taking other distribution regions larger than the preset correlation threshold as history reference regions;
and calculating the ratio of the sum of the correlation between the target distribution area and each historical reference area to the sum of the correlation between every two historical reference areas, and confirming the area correlation degree.
3. The intelligent prediction method for power of LED lamp beads based on data driving according to claim 2, wherein the power ratio of the target distribution area to other distribution areas at each moment is calculated by difference from the average value of the power ratios of the target distribution area to other distribution areas at all moments of the target analysis sequence, and the regional difference is confirmed, wherein the target analysis sequence is a power value sequence constructed by power values at preset number of moments in front of the target lamp beads, and specifically comprises:
wherein, For the region difference between the target distribution region where the target lamp bead a is located and the other distribution regions d, c a is the target distribution region where the target lamp bead a is located, d is the other distribution regions, L is the time length corresponding to the target analysis sequence, and Δlv (i,c,d) is the difference between the power ratio value of the target distribution region c a corresponding to the i-th time and the power ratio average value of the target distribution region c a and the other distribution region d at all times of the target analysis sequence.
4. The intelligent prediction method for power of LED lamp beads based on data driving according to claim 3, wherein the calculating the degree of lamp bead correlation according to the correlation between the target lamp bead and other lamp beads in the target distribution area specifically comprises:
Calculating the power ratio of the target lamp bead to other lamp beads in the target distribution area at each moment, and calculating the average value of the power ratios of the target lamp bead to other lamp beads in the target distribution area at all moments of the target analysis sequence, so as to confirm the difference of the initial lamp beads;
Calculating the final lamp bead difference based on the curve correlation of the initial lamp bead difference and other lamp beads corresponding to the target distribution area;
and confirming the degree of correlation of the lamp beads according to the proportion of the final lamp bead difference in the target distribution area.
5. The intelligent prediction method for power of LED lamp beads based on data driving according to claim 4, wherein the step of calculating the power ratio of the target lamp bead to other lamp beads in the target distribution area at each time, and the average value of the power ratios of the target lamp bead to other lamp beads in the target distribution area at all times in the target analysis sequence, to confirm the initial lamp bead difference, comprises the following steps:
Co (a,w) is the initial bead difference between the target bead a and the w other beads in the target distribution area, and DeltaGv (i,a,w) is the difference between the power ratio of the target bead a to the w other beads in the target distribution area at the i-th moment and the average value of the power ratios of the target bead a to the other beads in the target distribution area at all moments in the target analysis sequence.
6. The intelligent prediction method for LED lamp bead power based on data driving of claim 5, wherein said calculating final lamp bead differences based on curve correlation of said initial lamp bead differences with other lamp beads of the target distribution area comprises:
Wherein Con' (a,w) is the final bead difference between the target bead a and the w other beads in the target distribution area, DTW "j is the curve correlation between the j-th referent sequence of the other beads w in the target distribution area and the corresponding curve, x is the number of referent sequences, and Co ((d,a,w),j) is the initial bead difference between the target bead a and the w other beads in the target distribution area in the j-th referent sequence.
7. The method for intelligently predicting power of LED lamp beads based on data driving according to claim 6, wherein said calculating the second predicted power corresponding to the target lamp bead based on the region correlation degree and the lamp bead correlation degree specifically comprises:
wherein Pe is the second predicted power corresponding to the target lamp bead a, For the region difference between the target distribution region c a where the target lamp bead a is located and other distribution regions d, con' (a,w) is the final lamp bead difference between the target lamp bead a and the w other lamp beads in the target distribution region, q is the number of historical reference regions, r is the number of lamp beads in the target distribution region c a, and the predicted power value is obtained according to the power prediction of the w other lamp beads in the d historical reference region.
8. The intelligent prediction method for LED lamp bead power based on data driving of claim 7, wherein said determining the final predicted power corresponding to the target lamp bead by weight distribution of the first predicted power and the second predicted power corresponding to the target lamp bead according to the accuracy weight corresponding to the target lamp bead, specifically comprises:
predicting an ideal power value corresponding to the target lamp bead according to the current value and the voltage value corresponding to the referent sequence;
Calculating an accuracy weight corresponding to the target lamp bead based on the difference value between the ideal power value and the acquired actual power value;
And inputting the accuracy weight, the first predicted power and the second predicted power corresponding to the target lamp beads into a preset calculation formula, and confirming the final predicted power corresponding to the target lamp beads.
9. The intelligent prediction method for power of LED lamp beads based on data driving of claim 8, wherein said calculating the accuracy weight corresponding to the target lamp beads based on the difference between the ideal power value and the actual power value obtained by collection specifically comprises:
Wherein, pre a is the accuracy weight corresponding to the target lamp bead a, ΔPr (j,a) is the difference between the ideal power value of the j-th referenceable data corresponding to the target lamp bead a and the acquired actual power value;
Correspondingly, inputting the accuracy weight, the first predicted power and the second predicted power corresponding to the target lamp bead into a preset calculation formula, and confirming the final predicted power corresponding to the target lamp bead, wherein the method specifically comprises the following steps:
P＝Prea*P0+(1-Prea)*Pe
Wherein, P is the final predicted power corresponding to the target lamp bead a, pre a is the accuracy weight corresponding to the target lamp bead a, P 0 is the first predicted power, and Pe is the second predicted power.
10. LED lamp pearl power intelligence prediction system based on data drive is applied to lamp pearl power prediction field, its characterized in that, the system includes:
The confirming module is used for carrying out power prediction by taking the power information corresponding to the target lamp bead as a single calculation parameter and confirming the first predicted power corresponding to the target lamp bead;
The first calculation module is used for calculating the area correlation degree based on the correlation between the target distribution area where the target lamp beads are located and the historical reference area, wherein the historical reference area is other distribution areas with the correlation with the target distribution area being larger than a preset correlation threshold;
the second calculating module is used for calculating the correlation degree of the lamp beads according to the correlation between the target lamp beads and other lamp beads in the target distribution area;
The determining module is used for calculating second predicted power corresponding to the target lamp bead based on the region correlation degree and the lamp bead correlation degree;
and the prediction module is used for carrying out weight distribution on the first predicted power and the second predicted power corresponding to the target lamp beads through the accuracy weight corresponding to the target lamp beads, and confirming the final predicted power corresponding to the target lamp beads.
CN202311026862.7A 2023-08-15 2023-08-15 Intelligent prediction method and system for LED lamp bead power based on data driving Active CN117150891B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202311026862.7A CN117150891B (en) 2023-08-15 2023-08-15 Intelligent prediction method and system for LED lamp bead power based on data driving

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202311026862.7A CN117150891B (en) 2023-08-15 2023-08-15 Intelligent prediction method and system for LED lamp bead power based on data driving

Publications (2)

Publication Number Publication Date
CN117150891A CN117150891A (en) 2023-12-01
CN117150891B true CN117150891B (en) 2024-04-26

Family

ID=88905314

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202311026862.7A Active CN117150891B (en) 2023-08-15 2023-08-15 Intelligent prediction method and system for LED lamp bead power based on data driving

Country Status (1)

Country Link
CN (1) CN117150891B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN118589294A (en) * 2024-08-06 2024-09-03 大连中科超硅集成技术有限公司 Power control method, system and equipment of semiconductor laser

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112446554A (en) * 2020-12-18 2021-03-05 阳光电源股份有限公司 Power prediction model establishing method, power prediction method and device
WO2021052156A1 (en) * 2019-09-18 2021-03-25 平安科技（深圳）有限公司 Data analysis method, apparatus and device, and computer readable storage medium
WO2021213192A1 (en) * 2020-04-22 2021-10-28 国网江苏省电力有限公司苏州供电分公司 Load prediction method and load prediction system employing general distribution
CN114154697A (en) * 2021-11-19 2022-03-08 中国建设银行股份有限公司 House maintenance resource prediction method and device, computer equipment and storage medium
CN116316542A (en) * 2022-11-14 2023-06-23 国网浙江省电力有限公司宁波供电公司 Regional distributed photovoltaic power prediction method, regional distributed photovoltaic power prediction device, computer equipment and storage medium

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2021052156A1 (en) * 2019-09-18 2021-03-25 平安科技（深圳）有限公司 Data analysis method, apparatus and device, and computer readable storage medium
WO2021213192A1 (en) * 2020-04-22 2021-10-28 国网江苏省电力有限公司苏州供电分公司 Load prediction method and load prediction system employing general distribution
CN112446554A (en) * 2020-12-18 2021-03-05 阳光电源股份有限公司 Power prediction model establishing method, power prediction method and device
CN114154697A (en) * 2021-11-19 2022-03-08 中国建设银行股份有限公司 House maintenance resource prediction method and device, computer equipment and storage medium
CN116316542A (en) * 2022-11-14 2023-06-23 国网浙江省电力有限公司宁波供电公司 Regional distributed photovoltaic power prediction method, regional distributed photovoltaic power prediction device, computer equipment and storage medium

Also Published As

Publication number Publication date
CN117150891A (en) 2023-12-01

Similar Documents

Publication Publication Date Title
CN117150891B (en) Intelligent prediction method and system for LED lamp bead power based on data driving
CN106059492B (en) Photovoltaic module shade fault type judges method based on power prediction
CN109063366A (en) A kind of building performance online data preprocess method based on time and spatial weighting
CN110333962B (en) Electronic component fault diagnosis model based on data analysis and prediction
CN112989706B (en) Tunnel lamp illumination attenuation prediction method
CN107644297B (en) Energy-saving calculation and verification method for motor system
CN102095572A (en) Product performance test method based on benchmarking product comparison
CN112084717A (en) Ultraviolet light emitting diode performance degradation prediction model construction and service life prediction method
CN110757510A (en) Method and system for predicting remaining life of robot
CN114897241A (en) Intelligent building energy efficiency supervision and prediction method based on digital twins
Ibrahim et al. Bayesian based lifetime prediction for high-power white LEDs
Guo et al. A data-driven evaluating method on the defrosting effect of the air source heat pump system in Beijing
CN110880044A (en) Markov chain-based load prediction method
CN112989585B (en) Color prediction calibration method and system for RGB atmosphere lamp
CN117493932A (en) Management system based on LED operation information
CN115828744A (en) White light LED fault on-line diagnosis and service life prediction method
CN112996193B (en) Self-adaptive switching LED service life testing system and method based on vehicle arrival
CN115169707A (en) Equipment energy consumption prediction method and device based on multiple linear regression
Xia et al. Optimal metering plan of measurement and verification for energy efficiency lighting projects
CN110187294B (en) Fault diagnosis method and device for piecewise linear constant current drive LED light source
CN114781265B (en) Service performance evaluation method for highway tunnel lighting facility
CN117784736B (en) Intelligent building energy management method based on Internet of things technology
Ghafoori et al. Estimating Electricity Consumption of Buildings Using Information Theory and Machine Learning Methods
CN109635345A (en) A kind of lighting apparatus method for early warning based on cloud computing
Guo-guang et al. Prognostics and Health Management Technology of LED Lamp

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant