Optimization control system and method for driving power supply

Abstract

The invention relates to the technical field of power supply control, and discloses an optimal control system and method for a driving power supply, and the system comprises a monitoring module which is used for setting an initial working parameter of the driving power supply according to the type of a current light-emitting device, monitoring a real-time state parameter of the light-emitting device according to a preset time interval, and calculating a real-time state coefficient; the judgment module is used for judging whether optimization control is carried out or not according to the real-time state coefficient, if yes, a first optimization control strategy is generated, and a test time period is set; the calculation module is used for generating a first state feature, a second state feature and an efficiency feature in a test period and calculating an optimization evaluation value; the control module is used for judging whether the first optimization control strategy is corrected or not according to the optimization evaluation value, if yes, a second optimization control strategy is generated, it is guaranteed that the light-emitting device is in the optimal operation state, meanwhile, the operation state and the conversion efficiency of the driving power source are accurately evaluated, and the optimization control effect is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power supply control, in particular to an optimized control system and method of driving power supply. BACKGROUND

[0002] Light emitting devices such as LEDs, laser diodes, etc. have been widely used in the fields of lighting, display, communication, etc. due to their advantages of high efficiency, energy saving, long service life, etc. Driving power supply as a core component of light emitting devices directly affects the working stability, service life and energy efficiency of light emitting devices.

[0003] In the prior art, the driving power supply of light emitting devices has poor adaptability, and it is difficult to automatically adjust the output parameters according to different types and different rules of light emitting devices, which leads to the fact that the light emitting devices cannot work in the best state, and the running state of the driving power supply cannot be accurately diagnosed, which is easy to cause damage to the driving power supply and the light emitting devices or the power conversion efficiency to decrease, resulting in energy waste. SUMMARY

[0004] To solve the above technical problems, the present application provides an optimized control system and method of driving power supply, which sets the initial working parameters of the driving power supply, calculates the real-time state coefficient, judges whether to perform optimized control, if yes, generates a first optimized control strategy and sets a verification period, calculates the optimized evaluation value in the verification period, judges whether to perform correction, if yes, generates a second optimized control strategy, ensures that the light emitting device is in the best running state, accurately evaluates the running state and conversion efficiency of the driving power supply, and improves the optimized control effect.

[0005] In some embodiments of the present application, an optimized control system of driving power supply is provided, which comprises: A monitoring module is configured to set the initial working parameters of the driving power supply according to the current light emitting device category, monitor the real-time state parameters of the light emitting device according to a preset time interval, and calculate the real-time state coefficient; A judgment module is configured to judge whether to perform optimized control on the initial working parameters according to the real-time state coefficient, if yes, generate a first optimized control strategy and set a verification period; A calculation module is configured to generate the first state feature, the second state feature and the efficiency feature in the verification period, and calculate the optimized evaluation value; A control module is configured to judge whether to perform correction on the first optimized control strategy according to the optimized evaluation value, if yes, generate a second optimized control strategy, and issue the control instruction after the verification period.

[0006] In some embodiments of the present application, before setting the initial working parameters of the driving power supply according to the current light emitting device category, the method comprises: Constructing a plurality of light emitting device categories; acquire a plurality of historical working logs of each light emitting device category, extract historical state parameters of the light emitting device and historical working parameters of the driving power supply in each historical working log; construct a plurality of historical state parameter groups according to the historical state parameters at a plurality of same historical time nodes in the same historical working log, and each historical state parameter group is mapped with a historical working parameter group of the driving power supply at the same historical time node; set a plurality of state evaluation indexes based on the application scene of each light emitting device category; perform state evaluation on the plurality of historical state parameter groups of the corresponding light emitting device category based on the plurality of state evaluation indexes, obtain historical state evaluation values of each state evaluation index, and perform weight processing to obtain a historical comprehensive state evaluation value of each historical state parameter group; select a historical state parameter group with a historical comprehensive state evaluation value greater than a preset comprehensive state evaluation value threshold, and mark the historical working parameter group of the driving power supply mapped by each selected historical state parameter group as a to-be-determined working parameter group; compare the to-be-determined working parameter groups of different historical working logs of each light emitting device category, and calculate an optimal coefficient of each to-be-determined working parameter group according to the comparison result; set the to-be-determined working parameter group with the largest optimal coefficient as the optimal working parameter group of the corresponding light emitting device category.

[0007] In some embodiments of the present application, the initial working parameters of the driving power supply are set according to the current light emitting device category, including: acquire characteristic parameters of the current light emitting device, including self parameters and scene parameters; perform similarity analysis on the characteristic parameters of the current light emitting device and the historical characteristic parameters of each light emitting device category to obtain a similarity degree; wherein the historical characteristic parameters include historical self parameters and historical scene parameters; set the light emitting device category with the largest similarity degree as the current light emitting device category, and set the optimal working parameter group of the light emitting device category with the largest similarity degree as the initial working parameters of the current light emitting device category.

[0008] In some embodiments of the present application, the similarity degree includes: set a plurality of first characteristic evaluation indexes and a plurality of second characteristic evaluation indexes according to the characteristic parameters of the current light emitting device; evaluate the self parameters of the current light emitting device and the historical self parameters of each light emitting device category according to the plurality of first characteristic evaluation indexes respectively to obtain a plurality of first reference values and a plurality of second reference values; According to the historical scene parameters of each light emitting device category and the historical scene parameters of each light emitting device category, a plurality of second reference values and a plurality of fourth reference values are obtained by evaluating the historical scene parameters of each light emitting device category according to a plurality of second characteristic evaluation indexes; The similarity between the current light emitting device and each light emitting device category is calculated according to the first reference value, the second reference value corresponding to each first characteristic evaluation index, and the third reference value, the fourth reference value corresponding to each second characteristic evaluation index. The calculation formula of the similarity is: ; Wherein, D is the similarity, d1 is the first similarity conversion coefficient, q1 is the first weight coefficient, d2 is the second similarity conversion coefficient, q2 is the second weight coefficient, n1 is the number of first characteristic evaluation indexes, n2 is the number of second characteristic evaluation indexes, k1i is the first reference value of the i th first characteristic evaluation index, k2i is the second reference value of the i th first characteristic evaluation index, k3s is the third reference value of the s th second characteristic evaluation index, k4s is the fourth reference value of the s th second characteristic evaluation index, a1i is the weight coefficient of the i th first characteristic evaluation index, and a2s is the weight coefficient of the s th second characteristic evaluation index.

[0009] In some embodiments of the present application, the real-time state parameters of the light emitting device are monitored at a preset time interval, and the real-time state coefficient is calculated, including: According to a plurality of state evaluation indexes of the current light emitting device category, the real-time state parameters are evaluated to obtain a plurality of real-time state evaluation values of the state evaluation indexes; Based on the state evaluation value-state coefficient mapping table of each state evaluation index, the real-time state coefficient corresponding to the real-time state evaluation value of the corresponding state evaluation index is obtained; According to the real-time state coefficient corresponding to a plurality of state evaluation indexes and the weight coefficient of each state evaluation index, the real-time state coefficient is calculated; The state coefficient threshold is preset; If the real-time state coefficient is less than the state coefficient threshold, the initial working parameter is optimized and controlled; If the real-time state coefficient is not less than the state coefficient threshold, the initial working parameter is not optimized and controlled.

[0010] In some embodiments of the present application, a first optimization control strategy is generated and a test period is set, including: The historical optimization control log of the driving power supply of the current light emitting device category is obtained, a plurality of historical adjustment working parameters of the driving power supply in each historical optimization control log are extracted, and a historical attention period of each historical adjustment working parameter is set; extract a plurality of historical state parameters of the light-emitting device in the historical attention period of each historical adjustment working parameter, the historical change value of which is greater than a preset change value threshold; determine a historical adjustment value of each historical adjustment working parameter in different historical optimization control logs, and combine the historical change value of the extracted historical state parameter in the corresponding historical attention period in the corresponding historical optimization control log to construct an adjustment value-state parameter-change value matrix of each historical adjustment working parameter; determine whether there is an association relationship between the historical change value of each historical state parameter and the historical adjustment value of the historical adjustment working parameter according to the adjustment value-state parameter-change value matrix of each historical adjustment working parameter; If yes, set the historical adjustment working parameter associated with each historical state parameter as an associated working parameter, and construct an associated working parameter set of each historical state parameter; Each associated working parameter is mapped with an associated feature of the corresponding historical state parameter, and the associated feature includes a plurality of historical change values of the corresponding historical state parameter and a plurality of historical adjustment values of the associated working parameter. Each historical state parameter and a plurality of historical change values of each historical state parameter are used as training input data, and the associated working parameter set of each historical state parameter and a plurality of historical adjustment values of each associated working parameter are used as training output data, and neural network training is performed to obtain an association model; determine the abnormal state parameter of the current light-emitting device and the to-be-adjusted change value of each abnormal state parameter, and input them into the association model to obtain a plurality of to-be-adjusted working parameters and corresponding to-be-adjusted values; generate a first optimization control strategy according to the plurality of to-be-adjusted working parameters and the corresponding to-be-adjusted values; predict the optimization control feature of the first optimization control strategy, the optimization control feature including time feature, difficulty feature and precision feature; generate a prediction optimization coefficient according to the predicted optimization control feature; set a test time length according to the prediction optimization coefficient, and generate a test period.

[0011] In some embodiments of the present application, the first state feature, the second state feature and the efficiency feature in the test period are generated, including: set a plurality of acquisition time nodes according to the test time length of the test period and a preset time interval; acquire real-time first state parameters, real-time second state parameters and real-time efficiency parameters in the test period based on the plurality of acquisition time nodes, and construct a plurality of first state parameter change curves, a plurality of second state parameter change curves and a plurality of efficiency parameter change curves in the test period. pre-constructing a plurality of first standard state parameter variation curves, a plurality of second state parameter threshold lines and a plurality of efficiency parameter threshold lines in a test period; mapping each first state parameter variation curve and the corresponding first standard state parameter variation curve into the same reference graph to obtain a first state feature corresponding to each real-time first state parameter, the first state feature including a curve trend similarity coefficient, a parameter similarity coefficient and a slope similarity coefficient; mapping each second state parameter variation curve and the corresponding second state parameter threshold line into the same reference graph to obtain a second state feature corresponding to each real-time second state parameter, the second state feature including a first node number less than the second state parameter threshold line, a second real-time state parameter difference at each first node and a first time length; mapping each efficiency parameter variation curve and the corresponding efficiency parameter threshold line into the same reference graph to obtain an efficiency feature corresponding to each real-time efficiency parameter, the efficiency feature including a second node number less than the efficiency parameter threshold line, an efficiency parameter difference at each second node and a second time length.

[0012] In some embodiments of the present application, an optimization evaluation value is calculated, including: generating a comprehensive similarity coefficient according to the first state feature corresponding to each real-time first state parameter, and setting a first optimization sub-evaluation value according to the comprehensive similarity coefficient; generating a first optimization evaluation value according to the first optimization sub-evaluation values of a plurality of real-time first state parameters and the weight coefficients of the corresponding real-time first state parameters; generating an abnormality coefficient according to the second state feature corresponding to each real-time second state parameter, and setting a second optimization sub-evaluation value according to the abnormality coefficient; generating a second optimization evaluation value according to the second optimization sub-evaluation values of a plurality of real-time second state parameters and the weight coefficients of the corresponding real-time second state parameters; generating an energy efficiency coefficient according to the efficiency feature corresponding to each real-time efficiency parameter, and setting a third optimization sub-evaluation value according to the energy efficiency coefficient; generating a third optimization evaluation value according to the third optimization sub-evaluation values of a plurality of real-time efficiency parameters and the weight coefficients of the corresponding real-time efficiency parameters; generating an optimization evaluation value according to the first optimization evaluation value, the second optimization evaluation value and the third optimization evaluation value; The calculation formula of the optimization evaluation value is: ; wherein Y is the optimization evaluation value, y1 is the first optimization evaluation value, b1 is the first weight coefficient, b2 is the second weight coefficient, y2 is the second optimization evaluation value, b3 is the third weight coefficient, and y3 is the third optimization evaluation value.

[0013] In some embodiments of the present application, it is determined whether to modify the first optimization control strategy according to the optimization evaluation value, if yes, a second optimization control strategy is generated, comprising: an optimization evaluation value threshold is preset; if the optimization evaluation value is not less than the optimization evaluation value threshold, it is determined not to modify the first optimization control strategy; if the optimization evaluation value is less than the optimization evaluation value threshold, an optimization evaluation value difference is calculated, and it is determined to modify the first optimization control strategy; a plurality of preset optimization evaluation value difference intervals are set, and each preset optimization evaluation value difference interval is mapped with a specific modification span; the current modification span is determined based on the preset optimization evaluation value difference interval to which the optimization evaluation value difference belongs; a plurality of modification strategies of the proportional adjustment coefficient, the integral adjustment coefficient and the differential adjustment coefficient in the first optimization control strategy are generated based on the pre-constructed modification model and the current modification span, and a plurality of to-be-determined optimization control strategies are determined; the predicted optimization coefficient of each to-be-determined optimization control strategy is calculated, and the to-be-determined optimization control strategy corresponding to the maximum predicted optimization coefficient is set as the second optimization control strategy.

[0014] In some embodiments of the present application, an optimization control method of a driving power supply is further included, comprising: initial working parameters of the driving power supply are set according to the current light emitting device category, real-time state parameters of the light emitting device are monitored at a preset time interval, and a real-time state coefficient is calculated; it is determined whether to perform optimization control on the initial working parameters according to the real-time state coefficient, if yes, a first optimization control strategy is generated and a verification period is set; a first state feature, a second state feature and an efficiency feature in the verification period are generated, and an optimization evaluation value is calculated; it is determined whether to modify the first optimization control strategy according to the optimization evaluation value, if yes, a second optimization control strategy is generated, and a control instruction after the verification period is issued.

[0015] Compared with the prior art, the optimization control system and method of the driving power supply according to the embodiments of the present application have the following beneficial effects: By setting the initial working parameters of the driving power supply and calculating the real-time state coefficient, it is determined whether to perform optimization control, if yes, a first optimization control strategy is generated and a verification period is set, the optimization evaluation value in the verification period is calculated, it is determined whether to modify, if yes, a second optimization control strategy is generated, which ensures that the light emitting device is in the best operating state, accurately evaluates the operating state and conversion efficiency of the driving power supply, and improves the optimization control effect. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 is a schematic diagram of an optimization control system of a driving power supply in an embodiment of the present application; Figure 2 is a flowchart of an optimization control method of a driving power supply in an embodiment of the present application. DETAILED DESCRIPTION

[0017] The specific embodiments of the present application will be further described in conjunction with the accompanying drawings and embodiments. The following embodiments are used to illustrate the present application, but not to limit the scope of the present application.

[0018] In the description of the present application, it should be understood that the terms "center", "upper", "lower", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the purpose of facilitating the description of the present application and simplifying the description, and therefore cannot be understood as indicating or implying that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application.

[0019] The terms "first", "second" are only for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the technical features indicated. Therefore, the features defined with "first", "second" can explicitly or implicitly include one or more of the features. In the description of the present application, unless otherwise specified, the meaning of "a plurality of" is two or more.

[0020] In the description of the present application, it should be noted that, unless otherwise explicitly specified and limited, the terms "mounting", "connecting", "connection" should be understood in a broad sense, for example, it can be fixed connection, or detachable connection, or integral connection; it can be mechanical connection, or electrical connection; it can be direct connection, or indirect connection through intermediate medium, or internal communication of two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0021] As shown in Figure 1 The optimization control system of the driving power supply in an embodiment of the present application comprises: A monitoring module is configured to set initial working parameters of the driving power supply according to the current light emitting device category, monitor real-time state parameters of the light emitting device at a preset time interval, and calculate a real-time state coefficient; A judgment module is configured to determine whether to perform optimization control on the initial working parameters according to the real-time state coefficient, and if so, generate a first optimization control strategy and set a verification period; The computing module is configured to generate the first state feature, the second state feature and the efficiency feature in the test period, and calculate an optimization evaluation value; The control module is configured to determine whether to modify the first optimization control strategy according to the optimization evaluation value, and if so, generate a second optimization control strategy and issue a control instruction after the test period.

[0022] In the embodiment, the first state change feature refers to a state change feature of a real-time state parameter of the light emitting device, the second state change feature refers to a state change feature of a real-time state parameter of the driving power supply, and the efficiency change feature refers to an efficiency change feature of a real-time efficiency parameter of the driving power supply.

[0023] In some embodiments of the present application, before setting the initial working parameters of the driving power supply according to the current light emitting device category, the following steps are included: Constructing a plurality of light emitting device categories; Obtaining a plurality of historical working logs of each light emitting device category, and extracting historical state parameters of the light emitting device and historical working parameters of the driving power supply in each historical working log; Constructing a plurality of historical state parameter groups according to the historical state parameters at a plurality of same historical time nodes in the same historical working log, and each historical state parameter group is mapped with a historical working parameter group of the driving power supply at the same historical time node; Setting a plurality of state evaluation indexes based on the application scenarios of each light emitting device category; Performing state evaluation on a plurality of historical state parameter groups of the corresponding light emitting device category based on a plurality of state evaluation indexes, obtaining a historical state evaluation value of each state evaluation index, and performing weight processing to obtain a historical comprehensive state evaluation value of each historical state parameter group; Screening out historical state parameter groups with a historical comprehensive state evaluation value greater than a preset comprehensive state evaluation value threshold, and marking the historical working parameter group of the driving power supply mapped by each screened historical state parameter group as a to-be-determined working parameter group; Comparing the to-be-determined working parameter groups of different historical working logs of each light emitting device category, and calculating an optimal coefficient of each to-be-determined working parameter group according to the comparison result; Setting the to-be-determined working parameter group with the largest optimal coefficient as the optimal working parameter group of the corresponding light emitting device category.

[0024] In the embodiment, the light emitting device category is set according to the functional requirements, which include but are not limited to lighting, display, medical treatment, etc. The application scenarios correspond to the functional requirements, and the performance requirements of the light emitting device are different in different application scenarios, so the plurality of state evaluation indexes are also different.

[0025] In the embodiment, the historical state parameter of the light emitting device category refers to a parameter capable of characterizing the working state of the corresponding light emitting device, and the historical working parameter of the driving power supply includes output current, voltage, working frequency, and output power, etc.

[0026] In the embodiment, the standard state parameter interval of the state evaluation index of each light emitting device category is preset, and compared with the historical state parameter in the historical state parameter group. If it is not in the standard state parameter interval and the parameter difference is greater, the historical state evaluation value is smaller, and vice versa.

[0027] In the embodiment, the priority coefficient is calculated according to the application frequency of the historical working parameter of each to-be-determined working parameter group and the corresponding historical state evaluation value. When the application frequency of the historical working parameter is greater and the historical state evaluation value is greater, the corresponding priority coefficient is greater, and vice versa.

[0028] In the embodiment, by constructing a plurality of light emitting device categories and determining the preferred working parameter group of each light emitting device category, the working efficiency of the driving power supply of each light emitting device category is improved, so as to ensure that the running state of the light emitting device reaches the optimal working state.

[0029] In some embodiments of the present application, the initial working parameter of the driving power supply is set according to the current light emitting device category, including: obtaining the characteristic parameter of the current light emitting device, the characteristic parameter including the self parameter and the scene parameter; performing similarity analysis on the characteristic parameter of the current light emitting device and the historical characteristic parameter of each light emitting device category to obtain the similarity; wherein the historical characteristic parameter includes the historical self parameter and the historical scene parameter; setting the light emitting device category with the greatest similarity as the current light emitting device category, and setting the preferred working parameter group of the light emitting device category with the greatest similarity as the initial working parameter of the current light emitting device category.

[0030] In the embodiment, the self parameter of the light emitting device refers to the type, specification, etc., and the scene parameter is collected according to the scene where the current light emitting device is located. The historical self parameter and the historical scene parameter of each light emitting device category are obtained by aggregating the related historical parameters in the historical working log.

[0031] In the embodiment, the similarity between the current light emitting device and different light emitting device categories is calculated to determine the initial working parameter of the current light emitting device, thereby improving the working efficiency of the driving power supply of the current light emitting device and laying a foundation for subsequent optimization control.

[0032] In some embodiments of the present application, the similarity includes: According to the characteristic parameters of the current light emitting device, a plurality of first characteristic evaluation indexes and a plurality of second characteristic evaluation indexes are set; According to the plurality of first characteristic evaluation indexes, the self parameters of the current light emitting device and the historical self parameters of each light emitting device category are evaluated respectively, to obtain a plurality of first reference values and a plurality of second reference values; According to the plurality of second characteristic evaluation indexes, the historical scene parameters of each light emitting device category and the historical scene parameters of each light emitting device category are evaluated, to obtain a plurality of third reference values and a plurality of fourth reference values; According to the first reference value, the second reference value corresponding to each first characteristic evaluation index, and the third reference value, the fourth reference value corresponding to each second characteristic evaluation index, the similarity between the current light emitting device and each light emitting device category is calculated; The calculation formula of the similarity is: ; Wherein, D is the similarity, d1 is the first similarity conversion coefficient, q1 is the first weight coefficient, d2 is the second similarity conversion coefficient, q2 is the second weight coefficient, n1 is the number of first characteristic evaluation indexes, n2 is the number of second characteristic evaluation indexes, k1i is the first reference value of the i th first characteristic evaluation index, k2i is the second reference value of the i th first characteristic evaluation index, k3s is the third reference value of the s th second characteristic evaluation index, k4s is the fourth reference value of the s th second characteristic evaluation index, a1i is the weight coefficient of the i th first characteristic evaluation index, and a2s is the weight coefficient of the s th second characteristic evaluation index.

[0033] In some embodiments of the present application, the real-time state parameters of the light emitting device are monitored according to the preset time interval, and the real-time state coefficient is calculated, including: According to the plurality of state evaluation indexes of the current light emitting device category, the real-time state parameters are evaluated, to obtain the real-time state evaluation values of the plurality of state evaluation indexes; Based on the state evaluation value-state coefficient mapping table of each state evaluation index, the real-time state coefficient corresponding to the real-time state evaluation value of the corresponding state evaluation index is obtained; According to the real-time state coefficient corresponding to the plurality of state evaluation indexes and the weight coefficient of each state evaluation index, the real-time state coefficient is calculated; The state coefficient threshold is set in advance; If the real-time state coefficient is less than the state coefficient threshold, the initial working parameter is optimized and controlled; If the real-time state coefficient is not less than the state coefficient threshold, the initial working parameter is not optimized and controlled.

[0034] In the embodiment, the state evaluation value-state coefficient mapping table of each state evaluation index is constructed according to a plurality of historical state evaluation values and corresponding historical state coefficients of the corresponding light emitting device category. When the historical state evaluation value is larger, the mapped historical state coefficient is larger, and vice versa.

[0035] In the embodiment, the real-time state coefficient of the light emitting device is calculated to determine whether the light emitting device is in the optimal working state. If not, the first optimization control strategy is generated to adjust the working parameters of the driving power supply, improve the optimization control efficiency of the driving power supply, and ensure that the light emitting device is in the optimal working state.

[0036] In some embodiments of the present application, the first optimization control strategy is generated and the verification period is set, including: The historical optimization control logs of the driving power supply of the current light emitting device category are obtained, a plurality of historical adjustment working parameters of the driving power supply in each historical optimization control log are extracted, and a historical attention period of each historical adjustment working parameter is set; A plurality of historical state parameters of the light emitting device in the historical attention period of each historical adjustment working parameter are extracted, and the historical change value of each historical state parameter is greater than a preset change value threshold; The historical adjustment value of each historical adjustment working parameter in different historical optimization control logs is determined, and the historical change value of the extracted historical state parameter in the corresponding historical attention period in the corresponding historical optimization control log is combined to construct an adjustment value-state parameter-change value matrix of each historical adjustment working parameter; According to the adjustment value-state parameter-change value matrix of each historical adjustment working parameter, it is determined whether the historical change value of each historical state parameter and the historical adjustment value of the historical adjustment working parameter exist an association relationship; If yes, the historical adjustment working parameter associated with each historical state parameter is set as an associated working parameter, and an associated working parameter set of each historical state parameter is constructed; Each associated working parameter is mapped with an associated feature of the corresponding historical state parameter, and the associated feature includes a plurality of historical change values of the corresponding historical state parameter and a plurality of historical adjustment values of the associated working parameter; Each historical state parameter and a plurality of historical change values of each historical state parameter are taken as training input data, and the associated working parameter set of each historical state parameter and a plurality of historical adjustment values of each associated working parameter are taken as training output data, and neural network training is performed to obtain an association model; The abnormal state parameters of the current light emitting device and the to-be-adjusted change values of each abnormal state parameter are determined and input into the association model to obtain a plurality of to-be-adjusted working parameters and corresponding to-be-adjusted values; generate a first optimization control strategy according to a plurality of to-be-adjusted working parameters and corresponding to-be-adjusted values; predict an optimization control feature of the first optimization control strategy, the optimization control feature including a time feature, a difficulty feature, and a precision feature; generate a predicted optimization coefficient according to the predicted optimization control feature; set a test time length according to the predicted optimization coefficient, and generate a test period.

[0037] In this embodiment, the historical adjustment working parameter refers to a historical working parameter that is adjusted in a historical optimization control log, and the historical attention period refers to a period in which the historical adjustment working parameter has an impact on a historical state parameter of the light emitting device.

[0038] In this embodiment, the column of the adjustment value-state parameter-change value matrix of each historical adjustment working parameter is a historical change value of each extracted historical state parameter corresponding to different historical adjustment values of each historical adjustment working parameter, which is the same as the historical change value of the same historical state parameter.

[0039] In this embodiment, it is determined whether the historical change value of each extracted historical state parameter changes with the change of the historical adjustment value of the corresponding historical adjustment working parameter, and if so, the correlation coefficient is obtained.

[0040] In this embodiment, the time feature, the difficulty feature, and the precision feature refer to an optimization control time length, an optimization control difficulty, and an optimization control precision of the predicted first optimization control strategy. When the optimization control time length is longer, the optimization control difficulty is greater, and the optimization control precision is lower, the corresponding predicted optimization coefficient is smaller, and the set test time length is shorter, and vice versa.

[0041] In this embodiment, the optimization control feature of the first optimization control strategy is predicted according to the historical optimization control feature of the similar historical optimization control strategy, so that a reasonable test period is set, the optimization effect of the first optimization control strategy in the test period is evaluated, the optimization control strategy is adjusted in time, and the operation efficiency of the driving power supply and the light emitting device is improved.

[0042] In some embodiments of the present application, the first state feature, the second state feature, and the efficiency feature in the test period are generated, including: set a plurality of acquisition time nodes according to the test time length of the test period and a preset time interval; acquire real-time first state parameters, real-time second state parameters, and real-time efficiency parameters in the test period based on the plurality of acquisition time nodes, and construct a plurality of first state parameter change curves, a plurality of second state parameter change curves, and a plurality of efficiency parameter change curves in the test period; a plurality of first standard state parameter change curves, a plurality of second state parameter threshold lines and a plurality of efficiency parameter threshold lines in the pre-constructed inspection period are determined; each first state parameter change curve is mapped to the corresponding first standard state parameter change curve in the same reference graph to obtain a first state feature corresponding to each real-time first state parameter, and the first state feature includes a curve trend similarity coefficient, a parameter similarity coefficient and a slope similarity coefficient; each second state parameter change curve is mapped to the corresponding second state parameter threshold line in the same reference graph to obtain a second state feature corresponding to each real-time second state parameter, and the second state feature includes a first node number less than the second state parameter threshold line, a second real-time state parameter difference at each first node and a first time length; each efficiency parameter change curve is mapped to the corresponding efficiency parameter threshold line in the same reference graph to obtain an efficiency feature corresponding to each real-time efficiency parameter, and the efficiency feature includes a second node number less than the efficiency parameter threshold line, an efficiency parameter difference at each second node and a second time length.

[0043] In the embodiment, the real-time first state parameter refers to a parameter representing the running state of the light emitting device in the inspection period, the real-time second state parameter refers to a parameter representing whether the driving power supply fails in the inspection period, and the real-time efficiency parameter refers to a parameter representing the conversion efficiency of the driving power supply in the inspection period.

[0044] In the embodiment, the first standard state parameter change curve refers to a standard change curve of each first real-time state parameter meeting the best running state of the light emitting device, the second state parameter threshold line is constructed by the minimum value of each second real-time state parameter meeting the normal running of the driving power supply, and the efficiency parameter threshold line is constructed by the minimum value of each real-time efficiency parameter meeting the high conversion efficiency of the driving power supply, wherein the second state parameter threshold line and the efficiency parameter threshold line are straight lines.

[0045] In the embodiment, the curve trend similarity coefficient refers to the similarity degree of the overall shape and the change direction of each first state parameter change curve and the corresponding first standard state parameter change curve, the parameter similarity coefficient refers to the parameter similarity degree at each acquisition time node, and the slope similarity coefficient refers to the slope similarity degree at each acquisition time node.

[0046] In the embodiment, the first, second and efficiency features are determined to accurately evaluate the optimization efficiency of the first optimization control strategy on the running state of the light emitting device, the running state of the driving power supply and the conversion efficiency of the driving power supply, find the deficiencies in time and adjust to maximize the optimization control effect of the driving power supply.

[0047] In some embodiments of the present application, the optimization evaluation value is calculated, comprising: generating a comprehensive similarity coefficient according to the first state feature corresponding to each real-time first state parameter, and setting a first optimization sub-evaluation value according to the comprehensive similarity coefficient; generating a first optimization evaluation value according to the first optimization sub-evaluation values of the several real-time first state parameters and the weight coefficients of the corresponding real-time first state parameters; generating an abnormality coefficient according to the second state feature corresponding to each real-time second state parameter, and setting a second optimization sub-evaluation value according to the abnormality coefficient; generating a second optimization evaluation value according to the second optimization sub-evaluation values of the several real-time second state parameters and the weight coefficients of the corresponding real-time second state parameters; generating an energy efficiency coefficient according to the efficiency feature corresponding to each real-time efficiency parameter, and setting a third optimization sub-evaluation value according to the energy efficiency coefficient; generating a third optimization evaluation value according to the third optimization sub-evaluation values of the several real-time efficiency parameters and the weight coefficients of the corresponding real-time efficiency parameters; generating an optimization evaluation value according to the first optimization evaluation value, the second optimization evaluation value and the third optimization evaluation value; The calculation formula of the optimization evaluation value is: ; Wherein, Y is the optimization evaluation value, y1 is the first optimization evaluation value, b1 is the first weight coefficient, b2 is the second weight coefficient, y2 is the second optimization evaluation value, b3 is the third weight coefficient, and y3 is the third optimization evaluation value.

[0048] In the present embodiment, the larger the comprehensive similarity coefficient is, the larger the corresponding first optimization sub-evaluation value is, and vice versa.

[0049] In the present embodiment, the real-time second state parameter difference refers to the difference between the real-time second state parameter and the corresponding second state parameter threshold line. When the number of first nodes is larger, the real-time second state parameter difference at each first node is larger, and the time length of the continuous first nodes is longer, the abnormality coefficient is larger, and the corresponding second optimization sub-evaluation value is smaller, and vice versa.

[0050] In the present embodiment, the real-time efficiency parameter difference refers to the difference between the real-time efficiency parameter and the corresponding efficiency parameter threshold line. When the number of second nodes is larger, the real-time efficiency parameter difference at each second node is larger, and the time length of the continuous second nodes is longer, the energy efficiency coefficient is smaller, and the corresponding third optimization sub-evaluation value is smaller, and vice versa.

[0051] In some embodiments of the present application, it is judged whether to modify the first optimization control strategy according to the optimization evaluation value. If yes, a second optimization control strategy is generated, comprising: pre-set optimization evaluation value threshold value; if the optimization evaluation value is not less than the optimization evaluation value threshold value, it is determined that the first optimization control strategy is not modified; if the optimization evaluation value is less than the optimization evaluation value threshold value, the optimization evaluation value difference is calculated, and it is determined that the first optimization control strategy is modified; a plurality of pre-set optimization evaluation value difference intervals are set, and each pre-set optimization evaluation value difference interval is mapped with a specific modification span; the current modification span is determined based on the pre-set optimization evaluation value difference interval to which the optimization evaluation value difference belongs; a plurality of modification strategies of the proportional adjustment coefficient, the integral adjustment coefficient and the differential adjustment coefficient in the first optimization control strategy are generated based on the pre-constructed modification model and the current modification span, and a plurality of to-be-determined optimization control strategies are determined; the predicted optimization coefficient of each to-be-determined optimization control strategy is calculated, and the to-be-determined optimization control strategy corresponding to the maximum predicted optimization coefficient is set as the second optimization control strategy.

[0052] In the embodiment, the pre-constructed modification model is constructed according to the historical modification span and the corresponding historical optimization control strategy.

[0053] In the embodiment, by setting the second optimization control strategy, the shortcomings of the first optimization control strategy are timely adjusted, the optimization control efficiency of the driving power supply is improved, and the best working state of the light emitting device, the normal operation of the driving power supply and the best conversion efficiency are ensured.

[0054] In some embodiments of the present application, as shown in Figure 2 the optimization control method of the driving power supply further comprises the steps of: Step S201: setting the initial working parameters of the driving power supply according to the current light emitting device category, monitoring the real-time state parameters of the light emitting device at a pre-set time interval, and calculating the real-time state coefficient; Step S202: determining whether to perform optimization control on the initial working parameters according to the real-time state coefficient, if yes, generating the first optimization control strategy and setting the inspection period; Step S203: generating the first state feature, the second state feature and the efficiency feature in the inspection period, and calculating the optimization evaluation value; Step S204: determining whether to modify the first optimization control strategy according to the optimization evaluation value, if yes, generating the second optimization control strategy, and issuing the control instruction after the inspection period.

[0055] The above merely describes the preferred embodiments of the present application, and it should be pointed out that, for those skilled in the art, some improvements and replacements can be made without departing from the technical principles of the present application, and these improvements and replacements should also be considered as the protection scope of the present application.

Claims

1. An optimal control system for driving a power source, characterized by, The method comprises the following steps: a monitoring module is configured to set initial operating parameters of a driving power supply according to a current light emitting device category, monitor real-time state parameters of the light emitting device at preset time intervals, and calculate a real-time state coefficient; a judgment module is configured to determine whether to optimize the initial operating parameters according to the real-time state coefficient, and if so, generate a first optimization control strategy and set a test period; a calculation module is configured to generate a first state feature, a second state feature and an efficiency feature in the test period, and calculate an optimization evaluation value; a control module is configured to determine whether to modify the first optimization control strategy according to the optimization evaluation value, and if so, generate a second optimization control strategy and issue a control instruction after the test period.

2. The optimal control system of a driving power source according to claim 1, characterized by, Before setting the initial operating parameters of the driving power supply according to the current light emitting device category, the method comprises the following steps: constructing a plurality of light emitting device categories; obtaining a plurality of historical operating logs of each light emitting device category, extracting historical state parameters of the light emitting device and historical operating parameters of the driving power supply in each historical operating log; constructing a plurality of historical state parameter groups according to the historical state parameters at a plurality of same historical time nodes in the same historical operating log, and each historical state parameter group is mapped with a historical operating parameter group of the driving power supply at the same historical time node; setting a plurality of state evaluation indexes based on the application scenarios of each light emitting device category; performing state evaluation on a plurality of historical state parameter groups of the corresponding light emitting device category based on a plurality of state evaluation indexes, obtaining historical state evaluation values of each state evaluation index, and performing weight processing to obtain a historical comprehensive state evaluation value of each historical state parameter group; screening out historical state parameter groups with a historical comprehensive state evaluation value greater than a preset comprehensive state evaluation value threshold, and marking the historical operating parameter group of the driving power supply mapped by each screened historical state parameter group as a to-be-determined operating parameter group; comparing the to-be-determined operating parameter groups of different historical operating logs of each light emitting device category, and calculating an optimization coefficient of each to-be-determined operating parameter group according to the comparison result; setting the to-be-determined operating parameter group with the largest optimization coefficient as the optimization operating parameter group of the corresponding light emitting device category.

3. The optimal control system of a driving power source according to claim 2, characterized in that, Setting the initial operating parameters of the driving power supply according to the current light emitting device category comprises the following steps: obtaining characteristic parameters of the current light emitting device, wherein the characteristic parameters include self parameters and scene parameters; performing similarity analysis on the characteristic parameters of the current light emitting device and the historical characteristic parameters of each light emitting device category to obtain a similarity degree; wherein the historical characteristic parameters include historical self parameters and historical scene parameters; setting the light emitting device category with the largest similarity degree as the current light emitting device category, and setting the optimization operating parameter group of the light emitting device category with the largest similarity degree as the initial operating parameters of the current light emitting device category.

4. The optimum control system of a driving power source according to claim 3, characterized in that, The similarity degree comprises the following steps: setting a plurality of first feature evaluation indexes and a plurality of second feature evaluation indexes according to the characteristic parameters of the current light emitting device; evaluating the self parameters of the current light emitting device and the historical self parameters of each light emitting device category according to the plurality of first feature evaluation indexes, respectively, to obtain a plurality of first reference values and a plurality of second reference values; According to the historical scene parameters of each light emitting device category and the historical scene parameters of each light emitting device category, the historical scene parameters of each light emitting device category are evaluated according to a plurality of second characteristic evaluation indexes, and a plurality of third reference values and a plurality of fourth reference values are obtained; According to the first reference value, the second reference value corresponding to each first characteristic evaluation index and the third reference value, the fourth reference value corresponding to each second characteristic evaluation index, the similarity between the current light emitting device and each light emitting device category is calculated; The calculation formula of the similarity is: ； Wherein, D is the similarity, d1 is the first similarity conversion coefficient, q1 is the first weight coefficient, d2 is the second similarity conversion coefficient, q2 is the second weight coefficient, n1 is the number of first characteristic evaluation indexes, n2 is the number of second characteristic evaluation indexes, k1i is the first reference value of the i th first characteristic evaluation index, k2i is the second reference value of the i th first characteristic evaluation index, k3s is the third reference value of the s th second characteristic evaluation index, k4s is the fourth reference value of the s th second characteristic evaluation index, a1i is the weight coefficient of the i th first characteristic evaluation index, and a2s is the weight coefficient of the s th second characteristic evaluation index.

5. The optimal control system of a driving power source according to claim 4, characterized in that, According to the preset time interval, the real-time state parameters of the light emitting device are monitored, and the real-time state coefficient is calculated, including: According to the real-time state parameters, the real-time state evaluation value of a plurality of state evaluation indexes is obtained according to the current light emitting device category; Based on the state evaluation value-state coefficient mapping table of each state evaluation index, the real-time state coefficient corresponding to the real-time state evaluation value of the corresponding state evaluation index is obtained; According to the real-time state coefficient corresponding to a plurality of state evaluation indexes and the weight coefficient of each state evaluation index, the real-time state coefficient is calculated; The state coefficient threshold is set in advance; If the real-time state coefficient is less than the state coefficient threshold, the initial working parameter is optimized and controlled; If the real-time state coefficient is not less than the state coefficient threshold, the initial working parameter is not optimized and controlled.

6. The optimal control system of a driving power source according to claim 5, characterized in that, A first optimization control strategy is generated and a test period is set, including: The historical optimization control log of the driving power supply of the current light emitting device category is obtained, a plurality of historical adjustment working parameters of the driving power supply in each historical optimization control log are extracted, and the historical attention period of each historical adjustment working parameter is set; The historical change value of the light emitting device in the historical attention period of each historical adjustment working parameter is greater than the preset change value threshold, and a plurality of historical state parameters are extracted; The historical adjustment value of each historical adjustment working parameter in different historical optimization control logs is determined, and the historical change value of the extracted historical state parameter in the corresponding historical attention period in the corresponding historical optimization control log is combined to construct the adjustment value-state parameter-change value matrix of each historical adjustment working parameter; According to the adjustment value-state parameter-change value matrix of each historical adjustment working parameter, it is judged whether there is an association between the historical change value of each historical state parameter and the historical adjustment value of the historical adjustment working parameter; If so, set the historical adjustment working parameter associated with each historical state parameter as an associated working parameter, and construct an associated working parameter set of each historical state parameter; Each associated working parameter is mapped with an associated feature of the corresponding historical state parameter, and the associated feature includes a plurality of historical change values of the corresponding historical state parameter and a plurality of historical adjustment values of the associated working parameter; Each historical state parameter and a plurality of historical change values of each historical state parameter are taken as training input data, and the associated working parameter set of each historical state parameter and a plurality of historical adjustment values of each associated working parameter are taken as training output data, and neural network training is performed to obtain an associated model; Determine the abnormal state parameters of the current light emitting device and the to-be-adjusted change values of each abnormal state parameter, and input them into the associated model to obtain a plurality of to-be-adjusted working parameters and corresponding to-be-adjusted values; Generate a first optimization control strategy according to the plurality of to-be-adjusted working parameters and the corresponding to-be-adjusted values; Predict the optimization control features of the first optimization control strategy, including time features, difficulty features, and precision features; Generate a prediction optimization coefficient according to the predicted optimization control features; Set a test time length according to the prediction optimization coefficient, and generate a test period.

7. The optimal control system of a driving power source according to claim 6, characterized in that, Generate first state features, second state features, and efficiency features in the test period, including: Set a plurality of acquisition time nodes according to the test time length of the test period and a preset time interval; Acquire real-time first state parameters, real-time second state parameters, and real-time efficiency parameters in the test period based on the plurality of acquisition time nodes, and construct a plurality of first state parameter change curves, a plurality of second state parameter change curves, and a plurality of efficiency parameter change curves in the test period; Pre-construct a plurality of first standard state parameter change curves, a plurality of second state parameter threshold lines, and a plurality of efficiency parameter threshold lines in the test period; Map each first state parameter change curve and the corresponding first standard state parameter change curve to the same reference graph to obtain a first state feature corresponding to each real-time first state parameter, including a curve trend similarity coefficient, a parameter similarity coefficient, and a slope similarity coefficient; Map each second state parameter change curve and the corresponding second state parameter threshold line to the same reference graph to obtain a second state feature corresponding to each real-time second state parameter, including a first node number less than the second state parameter threshold line, a second real-time state parameter difference at each first node, and a first time length; Map each efficiency parameter change curve and the corresponding efficiency parameter threshold line to the same reference graph to obtain an efficiency feature corresponding to each real-time efficiency parameter, including a second node number less than the efficiency parameter threshold line, an efficiency parameter difference at each second node, and a second time length.

8. The optimal control system of a driving power source according to claim 7, characterized in that, Calculate an optimization evaluation value, including: Generate a comprehensive similarity coefficient according to the first state feature corresponding to each real-time first state parameter, and set a first optimization sub-evaluation value according to the comprehensive similarity coefficient; The first optimization evaluation value is generated according to the first optimization sub-evaluation values of the several real-time first state parameters and the weight coefficients of the corresponding real-time first state parameters; The abnormality coefficient is generated according to the second state characteristics corresponding to each real-time second state parameter, and the second optimization sub-evaluation value is set according to the abnormality coefficient; The second optimization evaluation value is generated according to the second optimization sub-evaluation values of the several real-time second state parameters and the weight coefficients of the corresponding real-time second state parameters; The energy efficiency coefficient is generated according to the efficiency characteristics corresponding to each real-time efficiency parameter, and the third optimization sub-evaluation value is set according to the energy efficiency coefficient; The third optimization evaluation value is generated according to the third optimization sub-evaluation values of the several real-time efficiency parameters and the weight coefficients of the corresponding real-time efficiency parameters; The optimization evaluation value is generated according to the first optimization evaluation value, the second optimization evaluation value and the third optimization evaluation value; The calculation formula of the optimization evaluation value is: ； Wherein, Y is the optimization evaluation value, y1 is the first optimization evaluation value, b1 is the first weight coefficient, b2 is the second weight coefficient, y2 is the second optimization evaluation value, b3 is the third weight coefficient, and y3 is the third optimization evaluation value.

9. The optimal control system of a driving power source according to claim 8, characterized in that, Whether the first optimization control strategy is modified is determined according to the optimization evaluation value, if yes, the second optimization control strategy is generated, including: The optimization evaluation value threshold is set in advance; If the optimization evaluation value is not less than the optimization evaluation value threshold, it is determined that the first optimization control strategy is not modified; If the optimization evaluation value is less than the optimization evaluation value threshold, the optimization evaluation value difference is calculated, and it is determined that the first optimization control strategy is modified; There are several preset optimization evaluation value difference intervals, and each preset optimization evaluation value difference interval is mapped with a specific modification span; The current modification span is determined based on the preset optimization evaluation value difference interval to which the optimization evaluation value difference belongs; Based on the pre-constructed modification model and the current modification span, several modification strategies of the proportional adjustment coefficient, the integral adjustment coefficient and the differential adjustment coefficient in the first optimization control strategy are generated, and several to-be-determined optimization control strategies are determined; The prediction optimization coefficient of each to-be-determined optimization control strategy is calculated, and the to-be-determined optimization control strategy corresponding to the maximum prediction optimization coefficient is set as the second optimization control strategy.

10. An optimal control method of driving a power source, characterized by, Including: The initial working parameter of the driving power supply is set according to the current light emitting device category, the real-time state parameter of the light emitting device is monitored at a preset time interval, and the real-time state coefficient is calculated; Whether the initial working parameter is optimized is determined according to the real-time state coefficient, if yes, the first optimization control strategy is generated and the verification period is set; The first state characteristics, the second state characteristics and the efficiency characteristics in the verification period are generated, and the optimization evaluation value is calculated; Whether the first optimization control strategy is modified is determined according to the optimization evaluation value, if yes, the second optimization control strategy is generated, and the control instruction after the verification period is issued.

Images