A method and system for adjusting power consumption of an RFID tag chip

Abstract

The application discloses a power consumption adjustment method and system for an RFID tag chip. The method comprises the following steps: analyzing the thermal effect interference distribution information of a target RFID tag chip and corresponding power consumption distribution imbalance modules; and considering the chip power consumption constraint condition and the data interaction configuration scheme of each power consumption distribution imbalance module when performing energy distribution optimization on each power consumption distribution imbalance module. Therefore, the power consumption distribution imbalance modules can be accurately and effectively optimized in the presence of thermal effect interference, while ensuring the smooth data interaction of each power consumption distribution imbalance module, thereby ensuring the stability and performance of the target RFID tag chip, and helping to meet the demand for coexistence of high performance and miniaturization of the RFID tag chip.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of chips, in particular to a power consumption adjustment method and system for an RFID tag chip. BACKGROUND

[0002] In the field of modern science and technology, the research on radio frequency identification tag chips based on artificial intelligence is of great significance, which is widely used in logistics, medical treatment and intelligent manufacturing scenes, and directly affects the promotion of informatization and intelligentization process. However, in the prior art, the radio frequency identification tag chip often exposes obvious shortcomings when facing complex application environment, especially in the aspects of function integration and stability guarantee in space limited scene, the prior art cannot meet the demand of coexistence of high performance and miniaturization of radio frequency identification tag chip, the main reason is that in the space limited scene, the high integration of radio frequency identification tag chip will cause the thermal effect of elements to be superimposed, thereby generating a non-negligible interference, which will directly cause the power consumption imbalance between multiple functional modules inside the radio frequency identification tag chip, and the prior art cannot effectively cope with the power consumption imbalance problem caused by thermal effect interference, thereby seriously affecting the stability and performance of the radio frequency identification tag chip. SUMMARY

[0003] The present application provides a power consumption adjustment method and system for an RFID tag chip to solve the technical problem that the prior art cannot effectively cope with the power consumption imbalance problem caused by thermal effect interference, thereby seriously affecting the stability and performance of the radio frequency identification tag chip.

[0004] In order to solve the above technical problems, the first aspect of the embodiment of the present application provides a power consumption adjustment method for an RFID tag chip, comprising:

[0005] According to the structure parameters and surface temperature distribution data of the target RFID tag chip, the thermal effect interference distribution information of the target RFID tag chip is determined;

[0006] Based on the thermal effect interference distribution information and the temperature detection data of each functional module in the target RFID tag chip, a plurality of power consumption allocation imbalance modules are determined;

[0007] According to the power consumption data of each power consumption allocation imbalance module and the preset chip configuration parameters, a data interaction configuration scheme of each power consumption allocation imbalance module is determined;

[0008] According to the data interaction configuration scheme and the preset chip power consumption constraint condition, the energy allocation optimization of each power consumption allocation imbalance module is carried out, and the target energy allocation proportion of each power consumption allocation imbalance module is determined;

[0009] According to the target energy distribution proportion, a power consumption limit value of each power consumption distribution imbalance module is determined, and output power consumption of each power consumption distribution imbalance module is adjusted according to the power consumption limit value.

[0010] As a preferred solution, the target RFID tag chip thermal effect interference distribution information is determined according to the structure parameters and surface temperature distribution data of the target RFID tag chip, and specifically includes:

[0011] According to the structure parameters, the assembly position, heat generation power, surface area, heat transfer distance and thermal conductivity coefficient of each functional module in the target RFID tag chip are determined;

[0012] According to the heat generation power, the surface area, the heat transfer distance and the thermal conductivity coefficient, the heat generation influence radius of each functional module is determined;

[0013] According to the assembly position and the heat generation influence radius, the heat generation influence area of each functional module is determined, and at least one thermal effect superposition area in the target RFID tag chip is determined based on the overlapping area between each heat generation influence area;

[0014] Based on the surface temperature distribution data, a plurality of surface temperature collection points in the thermal effect superposition area and a surface temperature value corresponding to each surface temperature collection point are determined;

[0015] According to the surface temperature value and a preset surface temperature threshold value, it is judged whether each thermal effect superposition area is a thermal effect interference area;

[0016] When there is any one thermal effect superposition area as the thermal effect interference area, the thermal effect interference distribution information is determined according to the position information of the thermal effect interference area and a plurality of target functional modules corresponding to the thermal effect interference area.

[0017] As a preferred solution, the plurality of power consumption distribution imbalance modules are determined based on the thermal effect interference distribution information and temperature detection data of each functional module in the target RFID tag chip, and specifically include:

[0018] According to the thermal effect interference distribution information, a plurality of target functional modules corresponding to each thermal effect interference area are determined;

[0019] From the temperature detection data of each functional module, a target temperature detection value sequence of each target functional module in a preset temperature detection period is obtained;

[0020] input the target temperature detection value sequence into a preset autoregressive moving average model to obtain a power consumption value sequence corresponding to the target temperature detection value sequence of each target function module; the autoregressive moving average model is obtained based on historical temperature detection data and historical power consumption data of each function module in a preset historical time period;

[0021] determine whether the output power consumption distribution of a plurality of target function modules corresponding to each thermal effect interference region is unbalanced according to the real-time chip total power consumption at each time in the power consumption value sequence, a preset upper limit of power consumption proportion of each target function module, a preset lower limit of power consumption proportion, and the power consumption value sequence;

[0022] when the output power consumption distribution of a plurality of target function modules corresponding to any one thermal effect interference region is unbalanced, determine the plurality of target function modules corresponding to the any one thermal effect interference region as the power consumption distribution imbalance module.

[0023] As a preferred solution, the data interaction configuration scheme of each power consumption distribution imbalance module is determined according to the power consumption data of each power consumption distribution imbalance module and a preset chip configuration parameter, specifically including:

[0024] determine a data interaction priority coefficient of each power consumption distribution imbalance module according to the power consumption data of each power consumption distribution imbalance module;

[0025] determine the data interaction configuration scheme of each power consumption distribution imbalance module according to the data interaction priority coefficient and the chip configuration parameter.

[0026] As a preferred solution, the data interaction priority coefficient of each power consumption distribution imbalance module is determined according to the power consumption data of each power consumption distribution imbalance module, specifically including:

[0027] determine the actual output power consumption and the rated output power consumption of each power consumption distribution imbalance module according to the power consumption data;

[0028] determine the data interaction priority coefficient of each power consumption distribution imbalance module according to the ratio between the actual output power consumption and the rated output power consumption.

[0029] As a preferred solution, the data interaction configuration scheme of each power consumption distribution imbalance module is determined according to the data interaction priority coefficient and the chip configuration parameter, specifically including:

[0030] According to the chip configuration parameter, determine the interconnection bandwidth parameter between each of the power consumption allocation imbalance modules, the task emergency coefficient to be processed and the data interaction coefficient; wherein, the task emergency coefficient to be processed is determined based on the difference between the preset processing deadline of the current task to be processed of the power consumption allocation imbalance module and the current time; the data interaction coefficient is used to represent the data interaction probability between each of the power consumption allocation imbalance modules, and the data interaction coefficient is determined based on the historical data interaction record of the power consumption allocation imbalance module;

[0031] According to the interconnection bandwidth parameter, the task emergency coefficient to be processed, the data interaction coefficient and the data interaction priority coefficient, determine the data interaction demand weight between each of the power consumption allocation imbalance modules;

[0032] According to the data interaction demand weight and the current residual bandwidth resource, determine the bandwidth configuration parameter of the data interaction path between each of the power consumption allocation imbalance modules;

[0033] According to the bandwidth configuration parameter of each of the data interaction paths, determine the data interaction configuration scheme of each of the power consumption allocation imbalance modules.

[0034] As a preferred scheme, according to the data interaction configuration scheme and the preset chip power consumption constraint condition, the energy allocation optimization of each of the power consumption allocation imbalance modules is performed to determine the target energy allocation proportion of each of the power consumption allocation imbalance modules, specifically including:

[0035] According to the data interaction configuration scheme, determine the bandwidth configuration parameter of the data interaction path between each of the power consumption allocation imbalance modules;

[0036] Based on the bandwidth configuration parameter and the preset unit bandwidth power consumption coefficient, determine the unit time energy consumption value of each of the data interaction paths;

[0037] According to the actual output power consumption of each of the functional modules and the unit time energy consumption value, determine the current total chip power consumption of the target RFID tag chip;

[0038] According to the actual output power consumption of each of the power consumption allocation imbalance modules, the current total chip power consumption and the chip power consumption constraint condition, the genetic algorithm is used to perform energy allocation optimization on each of the power consumption allocation imbalance modules to determine the target energy allocation proportion of each of the power consumption allocation imbalance modules.

[0039] As a preferred solution, the target energy distribution ratio of each of the power distribution imbalance modules is determined by using a genetic algorithm to optimize energy distribution of each of the power distribution imbalance modules according to actual output power consumption of each of the power distribution imbalance modules, the current total chip power consumption and the chip power consumption constraint condition, and specifically includes:

[0040] An initial energy distribution ratio of each of the power distribution imbalance modules is determined according to a ratio between the actual output power consumption of each of the power distribution imbalance modules and the current total chip power consumption.

[0041] An output power consumption upper limit of each of the functional modules and a total chip power consumption upper limit are determined according to the chip power consumption constraint condition.

[0042] A current energy distribution constraint condition is determined based on the output power consumption upper limit and the total chip power consumption upper limit.

[0043] An initial population containing a plurality of chromosomes is randomly generated in a preset initialization interval based on the initial energy distribution ratio, wherein the chromosomes contain a plurality of coding genes for representing energy distribution ratios of each of the power distribution imbalance modules and energy distribution ratios of the remaining functional modules other than the power distribution imbalance modules.

[0044] A fitness evaluation result is obtained by performing fitness evaluation on each of the chromosomes in the initial population based on a fitness function, wherein the fitness function is an inverse of a sum of squares of differences between energy distribution values of the functional modules and the output power consumption upper limit; the energy distribution value is a product of the energy distribution ratio and the total chip power consumption upper limit.

[0045] The initial population is iteratively updated by using a genetic algorithm based on the fitness evaluation result until the fitness evaluation result meets a preset iteration termination condition, and the target energy distribution ratio of each of the power distribution imbalance modules is determined according to a chromosome with the highest current fitness value.

[0046] As a preferred solution, the target energy distribution ratio of each of the power distribution imbalance modules is determined by using a genetic algorithm to optimize energy distribution of each of the power distribution imbalance modules according to actual output power consumption of each of the power distribution imbalance modules, the current total chip power consumption and the chip power consumption constraint condition, and specifically includes:

[0047] The power consumption limit value of each of the power distribution imbalance modules is determined according to a product of the total chip power consumption upper limit in the chip power consumption constraint condition and each of the target energy distribution ratios.

[0048] The second aspect of the embodiment of the present application provides a power consumption adjustment system of an RFID tag chip, which comprises:

[0049] a thermal effect interference distribution analysis module configured to determine thermal effect interference distribution information of the target RFID tag chip according to structure parameters and surface temperature distribution data of the target RFID tag chip;

[0050] a power consumption allocation imbalance module determination module configured to determine a plurality of power consumption allocation imbalance modules based on the thermal effect interference distribution information and temperature detection data of each functional module in the target RFID tag chip;

[0051] a data interaction configuration module configured to determine a data interaction configuration scheme of each power consumption allocation imbalance module according to power consumption data of each power consumption allocation imbalance module and preset chip configuration parameters;

[0052] an energy allocation optimization module configured to perform energy allocation optimization on each power consumption allocation imbalance module according to the data interaction configuration scheme and preset chip power consumption constraints to determine a target energy allocation ratio of each power consumption allocation imbalance module;

[0053] an output power adjustment module configured to determine a power consumption limit value of each power consumption allocation imbalance module according to the target energy allocation ratio and adjust output power of each power consumption allocation imbalance module according to the power consumption limit value.

[0054] Compared with the prior art, the embodiment of the present application has the beneficial effects that by analyzing the thermal effect interference distribution information of the target RFID tag chip and the corresponding power consumption allocation imbalance modules, and considering the chip power consumption constraints and the data interaction configuration scheme of each power consumption allocation imbalance module when performing energy allocation optimization on each power consumption allocation imbalance module, the power consumption allocation imbalance modules can be accurately and effectively optimized for power consumption allocation when there is thermal effect interference, while ensuring smooth data interaction of each power consumption allocation imbalance module, thereby ensuring the stability and performance of the target RFID tag chip, and helping to meet the demand for coexistence of high performance and miniaturization of RFID tag chips. BRIEF DESCRIPTION OF DRAWINGS

[0055] Figure 1 is a flowchart of the power consumption adjustment method of the RFID tag chip in the embodiment of the present application;

[0056] Figure 2 is a structural schematic diagram of the power consumption adjustment system of the RFID tag chip in the embodiment of the present application. DETAILED DESCRIPTION

[0057] With reference to the drawings of the embodiments of the present application, the technical solutions in the embodiments of the present application will be described clearly and completely. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments of the present application, all the other embodiments obtained by those skilled in the art without creative work fall within the scope of the present application.

[0058] Please refer to Figure 1 The first aspect of the embodiments of the present application provides a power consumption adjustment method of an RFID tag chip, comprising the following steps S1 to S5.

[0059] Step S1, according to the structure parameters and surface temperature distribution data of a target RFID tag chip, determining the thermal effect interference distribution information of the target RFID tag chip;

[0060] Step S2, based on the thermal effect interference distribution information and the temperature detection data of each functional module in the target RFID tag chip, determining a plurality of power consumption allocation imbalance modules;

[0061] Step S3, according to the power consumption data of each power consumption allocation imbalance module and the preset chip configuration parameters, determining the data interaction configuration scheme of each power consumption allocation imbalance module;

[0062] Step S4, according to the data interaction configuration scheme and the preset chip power consumption constraint condition, performing energy allocation optimization on each power consumption allocation imbalance module, and determining the energy allocation proportion of each power consumption allocation imbalance module;

[0063] Step S5, according to the energy allocation proportion, determining the power consumption limit value of each power consumption allocation imbalance module, and adjusting the output power consumption of each power consumption allocation imbalance module according to the power consumption limit value.

[0064] Specifically, the thermal effect interference of the RFID (Radio Frequency Identification) tag chip is usually caused by the thermal effect of elements in a space-limited scenario, such as some highly integrated or miniaturized RFID tag chips. The thermal effect interference not only causes an imbalance in power consumption among multiple functional modules, but also affects the stability of data transmission of the RFID tag chip. The reason is that temperature changes can cause changes in the dielectric constant of materials, which is sufficient to cause a significant deviation in the originally stable signal propagation path. Therefore, it is necessary to optimize the power consumption distribution of the power consumption imbalance module under the thermal effect interference. It can be understood that the assembly position of each functional module in the RFID tag chip determines the thermal effect superposition area between the functional modules. That is, for different RFID tag chips, the functional modules contained therein may be different, the assembly position of the functional modules may be different, and thus the thermal effect superposition area may be different. In order to accurately identify the thermal effect interference information of the target RFID tag chip, the embodiment needs to first obtain the structure parameters and surface temperature distribution data of the target RFID tag chip to determine the thermal effect interference distribution information thereof.

[0065] Further, after determining the thermal effect interference distribution information, the current several power consumption imbalance modules can be determined based on the temperature detection data of each functional module corresponding to the thermal effect interference distribution area.

[0066] Further, when optimizing the power consumption distribution of the power consumption imbalance module, not only is it necessary to ensure that the output power consumption of each power consumption imbalance module after optimization is within a suitable interval, thereby achieving power consumption distribution balance, but also it is necessary to ensure the high performance of the target RFID tag chip. In order to ensure the high performance of the target RFID tag chip, it is necessary to ensure that the data interaction of each power consumption imbalance module proceeds smoothly, and avoid the execution of the to-be-processed tasks of each power consumption imbalance module at each moment from being affected. Therefore, the embodiment determines the data interaction configuration scheme of each power consumption imbalance module according to the power consumption data and chip configuration parameters of each power consumption imbalance module, so as to provide an optimization basis for subsequent energy distribution optimization. It can be understood that the power consumption data of each power consumption imbalance module can directly reflect the computing tasks undertaken by each power consumption imbalance module, and the chip configuration parameters have real-time and accuracy, which store parameters related to the data interaction of each power consumption imbalance module, such as interconnection bandwidth parameters, information of to-be-processed tasks, etc., thereby providing a basis for determining the data interaction configuration scheme.

[0067] Further, since different cooling systems of different RFID tag chips correspond to different heat dissipation capacities, the overall power consumption of the RFID tag chip and the power consumption of each functional module thereof have different power consumption constraints under the heat dissipation capacity of the different cooling systems, so as to ensure that the cooling system of the current RFID tag chip can effectively dissipate heat under the power consumption constraint, and avoid continuous superposition of thermal effects. Therefore, based on the above data interaction configuration scheme and the preset chip power consumption constraint condition of the target RFID tag chip, the energy distribution of each power distribution imbalance module is optimized to determine the target energy distribution ratio of each power distribution imbalance module. According to the target energy distribution ratio, the power consumption limit value of each power distribution imbalance module can be converted, and the power consumption configuration instruction containing the power consumption limit value of each power distribution imbalance module is sent to each power distribution imbalance module through the power consumption control interface inside the target RFID tag chip, so as to adjust the output power consumption of each power distribution imbalance module to meet the allocated power consumption limit value.

[0068] The power consumption adjustment method of the RFID tag chip provided by the embodiment of the application can accurately and effectively optimize the power consumption distribution of the power distribution imbalance module when there is thermal effect interference, while ensuring smooth data interaction of each power distribution imbalance module, thereby ensuring the stability and performance of the target RFID tag chip, and helping to meet the demand for coexistence of high performance and miniaturization of the RFID tag chip.

[0069] As a preferred scheme, the thermal effect interference distribution information of the target RFID tag chip is determined according to the structure parameters and the surface temperature distribution data of the target RFID tag chip, and specifically includes:

[0070] According to the structure parameters, the assembly position, the heat generation power, the surface area, the heat transfer distance and the thermal conductivity of each functional module in the target RFID tag chip are determined;

[0071] According to the heat generation power, the surface area, the heat transfer distance and the thermal conductivity, the heat generation influence radius of each functional module is determined;

[0072] According to the assembly position and the heat generation influence radius, the heat generation influence area of each functional module is determined, and at least one thermal effect superposition area in the target RFID tag chip is determined based on the overlapping area between each heat generation influence area;

[0073] determining a plurality of surface temperature collection points in the thermal effect superposition region and a surface temperature value corresponding to each of the surface temperature collection points based on the surface temperature distribution data;

[0074] judging whether each of the thermal effect superposition regions is a thermal effect interference region according to the surface temperature value and a preset surface temperature threshold value;

[0075] when any one of the thermal effect superposition regions is the thermal effect interference region, determining the thermal effect interference distribution information according to position information of the thermal effect interference region and a plurality of target functional modules corresponding to the thermal effect interference region.

[0076] Specifically, the embodiment obtains the assembly position, heat generation power, surface area, heat transfer distance and thermal conductivity of each functional module of the target RFID tag chip based on the structural parameters of the target RFID tag chip. Further, the heat flow density can be obtained according to the ratio of the heat generation power of each functional module to the surface area thereof, and then the heat generation influence radius of each functional module can be obtained according to the multiplication of the heat flow density and the heat transfer distance and the ratio of the multiplication result to the thermal conductivity. It can be understood that the heat generation power refers to the heat generated by the functional module per unit time, which is usually measured in watts (W) and reflects the rate of the conversion of the electric energy consumed by the functional module in the running process into heat energy; the thermal conductivity is a physical quantity for measuring the heat conduction capacity of a material, which is usually represented by k and measured in watts per meter·kelvin (W / m·K), and reflects the heat passing through a unit area per unit time under a unit temperature gradient; the greater the thermal conductivity, the stronger the heat conduction capacity of the material; the heat transfer distance refers to the path length of the heat transfer from the heat source to the surrounding environment or the heat dissipation structure, which is usually measured in meters (m).

[0077] Further, the heat generation influence region of each functional module can be determined based on the assembly position and the heat generation influence radius of each functional module, and when there is an overlapping region between the heat generation influence regions of at least two functional modules, the overlapping region is the thermal effect superposition region.

[0078] Further, the embodiment pre-arranges a plurality of temperature sensors on the surface of the target RFID tag chip to collect the surface temperature distribution data of the target RFID tag chip. For example, the embodiment can divide the surface area of the target RFID tag chip into a plurality of grids according to a preset grid length, such as 10x10 grid area, and then arrange one temperature sensor in each grid area, so that there are 100 temperature collection points on the surface of the target RFID tag chip, and each temperature sensor has a unique position coordinate, which can determine the surface area corresponding to each surface temperature value. For each heat effect superposition area, the surface temperature collection points contained therein are counted and the surface temperature value corresponding to each surface temperature collection point is obtained, and then based on a preset surface temperature threshold value, such as 90°C, 92°C, 94°C, 95°C, etc., which is not specifically limited in the embodiment, the proportion of the surface temperature collection points in the heat effect superposition area that exceed the surface temperature threshold value is counted, and the proportion is used to determine whether each heat effect superposition area is a heat effect interference area. It can be understood that when the proportion of the surface temperature collection points in the heat effect superposition area that exceed the surface temperature threshold value is greater than a preset proportion threshold value, it indicates that there is excessive heat aggregation in the heat effect superposition area, and it is determined as a heat effect interference area. Since each heat effect superposition area is determined by the overlapping of the heat-affected areas of at least two functional modules, each heat effect interference area has at least two corresponding functional modules. According to the position information of each heat effect interference area and the corresponding plurality of target functional modules, the heat effect interference distribution information can be determined.

[0079] As a preferred solution, the determination of a plurality of power consumption allocation imbalance modules based on the heat effect interference distribution information and the temperature detection data of each functional module in the target RFID tag chip specifically includes:

[0080] determining a plurality of target functional modules corresponding to each heat effect interference area according to the heat effect interference distribution information;

[0081] obtaining a target temperature detection value sequence of each target functional module in a preset temperature detection period from the temperature detection data of each functional module;

[0082] inputting the target temperature detection value sequence into a preset autoregressive moving average model to obtain a power consumption value sequence corresponding to the target temperature detection value sequence of each target functional module; wherein the autoregressive moving average model is obtained based on historical temperature detection data and historical power consumption data of each functional module in a preset historical time period;

[0083] determine whether the output power distribution of the target function modules corresponding to each thermal effect interference region is unbalanced according to the real-time total chip power consumption at each time in the power consumption value sequence, the preset upper limit of power consumption proportion, the preset lower limit of power consumption proportion of each target function module, and the power consumption value sequence;

[0084] When the output power distribution of the target function modules corresponding to any one thermal effect interference region is unbalanced, the target function modules corresponding to the any one thermal effect interference region are determined as the power consumption distribution imbalance modules.

[0085] Specifically, after determining the thermal effect interference distribution information, the target function modules corresponding to each thermal effect interference region can be determined. Since there is a positive correlation between the operating temperature and the output power consumption of each function module, in order to determine whether the output power distribution of the target function modules corresponding to each thermal effect interference region is unbalanced, the embodiment obtains a target temperature detection value sequence of each target function module in a preset temperature detection period from the temperature detection data of each function module.

[0086] Further, the embodiment derives a corresponding power consumption value sequence from the target temperature detection value sequence of each target function module by using an autoregressive moving average model. It can be understood that there is a nonlinear relationship between the operating temperature and the output power consumption of a function module, for example, the output power consumption of a function module may increase by 15% to 20% when its operating temperature increases by 10℃. The embodiment finds that the autoregressive moving average model has a unique advantage in processing the corresponding relationship between temperature and power consumption. The autoregressive moving average model is trained by using historical temperature detection data and historical power consumption data of each function module in a preset historical time period. It is worth noting that at each time in the historical time period, a corresponding temperature detection value and power consumption value can be obtained from the historical temperature detection data and historical power consumption data. Therefore, training the autoregressive moving average model based on the historical temperature detection data and historical power consumption data can ensure that the autoregressive moving average model fits the nonlinear relationship between the operating temperature and the output power consumption of a function module, and realizes the conversion from the temperature sequence to the power consumption sequence.

[0087] Further, by dividing the difference between the power consumption values of adjacent time points in the power consumption value sequence by the time interval, the instantaneous power consumption change rate of each target functional module at each time point can be obtained. When the instantaneous power consumption change rate is positive, it indicates that the power consumption is rising. When the instantaneous power consumption change rate is negative, it indicates that the power consumption is falling. In this embodiment, the proportion analysis method is used for the determination of power consumption allocation imbalance. Under normal circumstances, for CPU (Central Processing Unit), GPU (Graphics Processing Unit) and other functional modules, the power consumption of CPU accounts for about 40% of the total chip power consumption, the power consumption of GPU accounts for about 30% of the total chip power consumption, and the power consumption of the remaining functional modules accounts for about 30% of the total chip power consumption. Based on the real-time total chip power consumption at each time point in the power consumption value sequence, the upper limit of the preset power consumption proportion, the lower limit of the preset power consumption proportion and the power consumption value sequence, it can be determined whether the output power consumption distribution of a plurality of target functional modules corresponding to each thermal effect interference region is imbalanced. Specifically, the output power consumption distribution imbalance usually occurs in a specific spatial region, which is caused by thermal effect interference. When it is detected that among a plurality of target functional modules corresponding to a thermal effect interference region, there is at least one target functional module whose power consumption value exceeds the upper limit of the preset power consumption proportion of the real-time total chip power consumption and lasts for a preset time, while the power consumption value of the remaining target functional modules is compressed to be less than the lower limit of the preset power consumption proportion and lasts for a preset time, it is determined that the plurality of target functional modules corresponding to the thermal effect interference region have power consumption allocation imbalance.

[0088] As a preferred solution, the data interaction configuration scheme of each power consumption allocation imbalance module is determined according to the power consumption data of each power consumption allocation imbalance module and the preset chip configuration parameters, specifically including:

[0089] According to the power consumption data of each power consumption allocation imbalance module, a data interaction priority coefficient of each power consumption allocation imbalance module is determined.

[0090] According to the data interaction priority coefficient and the chip configuration parameters, the data interaction configuration scheme of each power consumption allocation imbalance module is determined.

[0091] Specifically, based on the power consumption data of each power consumption allocation imbalance module, the computing tasks undertaken by each power consumption allocation imbalance module can be determined. The heavier the computing tasks undertaken, the higher the urgency of resource allocation. Therefore, the data interaction priority coefficient of each power consumption allocation imbalance module is determined.

[0092] Further, in addition to considering the computing tasks undertaken by each power consumption allocation imbalance module, the chip configuration parameters also need to be considered for data interaction configuration of each power consumption allocation imbalance module, so as to simultaneously take into account the hardware capability, task demand and correlation between modules, achieve the dual goals of power consumption balance and performance optimization, and effectively alleviate the local overheating problem.

[0093] As a preferred solution, the data interaction priority coefficient of each power consumption allocation imbalance module is determined according to the power consumption data of each power consumption allocation imbalance module, specifically including:

[0094] According to the power consumption data, the actual output power consumption and the rated output power consumption of each power consumption allocation imbalance module are determined;

[0095] According to the ratio between the actual output power consumption and the rated output power consumption, the data interaction priority coefficient of each power consumption allocation imbalance module is determined.

[0096] Specifically, each functional module will determine a rated output power consumption when designed. According to the ratio between the actual output power consumption and the rated output power consumption of each power consumption allocation imbalance module, the load pressure of each power consumption allocation imbalance module can be intuitively reflected, so that this ratio is taken as the data interaction priority coefficient of each power consumption allocation imbalance module in this embodiment. For example, assuming that there are three power consumption allocation imbalance modules, GPU, CPU and cache controller, and after calculation, the ratio between the actual output power consumption and the rated output power consumption of GPU is 1.5, the ratio between the actual output power consumption and the rated output power consumption of CPU is 1.3, and the ratio between the actual output power consumption and the rated output power consumption of the cache controller is 1.1, then the data interaction priority coefficients of GPU, CPU and cache controller are 1.5, 1.3 and 1.1 respectively, and the size of the data interaction priority coefficient reflects the urgency of the demand for data transmission resources of the power consumption allocation imbalance module.

[0097] As a preferred solution, the data interaction configuration scheme of each power consumption allocation imbalance module is determined according to the data interaction priority coefficient and the chip configuration parameters, specifically including:

[0098] According to the chip configuration parameters, the interconnection bandwidth parameters between each power consumption allocation imbalance module, the task emergency coefficient and the data interaction coefficient are determined; wherein, the task emergency coefficient is determined based on the difference between the preset processing deadline of the current task to be processed of the power consumption allocation imbalance module and the current time; the data interaction coefficient is used to represent the data interaction probability between each power consumption allocation imbalance module, and the data interaction coefficient is determined based on the historical data interaction record of the power consumption allocation imbalance module;

[0099] determine a data interaction demand weight between each of the power consumption allocation imbalance modules according to the interconnection bandwidth parameter, the to-be-processed task emergency coefficient, the data interaction coefficient and the data interaction priority coefficient;

[0100] determine a data interaction path bandwidth configuration parameter between each of the power consumption allocation imbalance modules according to the data interaction demand weight and a current residual bandwidth resource;

[0101] determine the data interaction configuration scheme of each of the power consumption allocation imbalance modules according to the bandwidth configuration parameter of each of the data interaction paths.

[0102] Specifically, the embodiment first obtains, from the chip configuration parameter, an interconnection bandwidth parameter, a to-be-processed task emergency coefficient and a data interaction coefficient between each of the power consumption allocation imbalance modules. It can be understood that the interconnection bandwidth parameter is used to represent the maximum transmission capacity of the data transmission channel between each of the power consumption allocation imbalance modules, and is usually in units of GB / s; the to-be-processed task emergency coefficient is determined based on the difference between the preset processing deadline of the current to-be-processed task of the power consumption allocation imbalance module and the current time, and for example, the embodiment can set a corresponding to-be-processed task emergency coefficient for different difference intervals, and the smaller the difference, the higher the degree of urgency of the to-be-processed task, and the larger the corresponding to-be-processed task emergency coefficient; the data interaction coefficient is used to represent the data interaction probability between each of the power consumption allocation imbalance modules, which can provide a basis for subsequent allocation of data interaction resources, wherein the data interaction coefficient is determined based on the historical data interaction record of the power consumption allocation imbalance module. It can be understood that the embodiment can count the total data interaction times of the power consumption allocation imbalance module in a certain historical time period and the data interaction times between the power consumption allocation imbalance module and each of the remaining functional modules through the historical data interaction record, and determine the data interaction probability between the power consumption allocation imbalance module and each of the remaining functional modules based on the ratio of the data interaction times between the power consumption allocation imbalance module and each of the remaining functional modules to the total data interaction times. If the data interaction probability between two power consumption allocation imbalance modules is larger, it means that the data interaction demand between the two power consumption allocation imbalance modules is larger, and more data interaction resources need to be allocated.

[0103] Further, a product of the interconnection bandwidth parameter, the emergency coefficient of the task to be processed, the data interaction coefficient and the data interaction priority coefficient is taken as a data interaction demand coefficient of each data interaction path between the power consumption allocation imbalance modules, then the data interaction demand coefficients of each data interaction path are summed to obtain a total data interaction demand coefficient of each data interaction path, and then a data interaction demand weight between the power consumption allocation imbalance modules is determined based on a ratio between the data interaction demand coefficient and the total data interaction demand coefficient. It can be understood that, when the data interaction demand coefficient is calculated, the first power consumption allocation imbalance module and the second power consumption allocation imbalance module are respectively used to represent two power consumption allocation imbalance modules involved in the data interaction path, and if the data interaction path is used for transmitting data from the first power consumption allocation imbalance module to the second power consumption allocation imbalance module, the data interaction priority coefficient used when the data interaction demand coefficient is calculated is the data interaction priority coefficient of the first power consumption allocation imbalance module.

[0104] Further, the data interaction path with a larger data interaction demand weight indicates that the priority of the task for transmitting data on the data interaction path is higher, and therefore a product of each interaction demand weight and the current residual bandwidth resource can be used to determine a bandwidth configuration parameter of the data interaction path between the power consumption allocation imbalance modules, so that the data interaction path with a larger data interaction demand weight can be allocated more bandwidth resources.

[0105] As a preferred solution, the energy allocation optimization of each power consumption allocation imbalance module is performed according to the data interaction configuration scheme and the preset chip power consumption constraint condition, and a target energy allocation ratio of each power consumption allocation imbalance module is determined, and the energy allocation optimization specifically includes:

[0106] The bandwidth configuration parameter of the data interaction path between each power consumption allocation imbalance module is determined according to the data interaction configuration scheme;

[0107] A unit time energy consumption value of each data interaction path is determined based on the bandwidth configuration parameter and a preset unit bandwidth power consumption coefficient;

[0108] A current total chip power consumption of the target RFID tag chip is determined according to the actual output power consumption of each functional module and the unit time energy consumption value;

[0109] The genetic algorithm is used to perform the energy allocation optimization of each power consumption allocation imbalance module according to the actual output power consumption of each power consumption allocation imbalance module, the current total chip power consumption and the chip power consumption constraint condition, and the target energy allocation ratio of each power consumption allocation imbalance module is determined.

[0110] Specifically, after obtaining the data interaction configuration scheme of each power consumption allocation imbalance module, the bandwidth configuration parameters of the data interaction path between each power consumption allocation imbalance module can be determined, for example, the data interaction path for data transmission from a GPU with a higher priority to a cache controller is allocated a bandwidth of 16 GB / s.

[0111] Further, the bandwidth configuration parameter of each data interaction path is a guarantee for the smooth progress of the data interaction task, and the greater the allocated bandwidth, the higher the required power consumption. Therefore, based on the product of the bandwidth configuration parameter and a preset unit bandwidth power consumption coefficient, the unit time energy consumption value of each data interaction path is determined, wherein the unit bandwidth power consumption coefficient is used to represent the power consumption per unit data transmission bandwidth. For example, assuming that a certain data interaction path is allocated a bandwidth of 16 GB / s, and the unit bandwidth power consumption coefficient is that 0.5 W of power is consumed per 1 GB of data transmission per unit time, then the unit time energy consumption value of the data interaction path is 8 W. By counting the actual output power of each functional module and the unit time energy consumption value of the data interaction path between each power consumption allocation imbalance module, the current total chip power consumption can be determined, which is the power consumption value required for the normal operation of each functional module and the normal data interaction of each power consumption allocation imbalance module.

[0112] Further, based on the actual output power of each power consumption allocation imbalance module, the current total chip power consumption and the chip power consumption constraint condition, a genetic algorithm is used to optimize the energy allocation of each power consumption allocation imbalance module, which not only fully utilizes the capabilities of each power consumption allocation imbalance module, but also ensures that the power consumption value of each power consumption allocation imbalance module does not exceed its own power consumption constraint value, thereby solving the power consumption allocation imbalance problem while ensuring the normal data interaction of each power consumption allocation imbalance module.

[0113] As a preferred solution, the genetic algorithm is used to optimize the energy allocation of each power consumption allocation imbalance module according to the actual output power of each power consumption allocation imbalance module, the current total chip power consumption and the chip power consumption constraint condition, to determine the target energy allocation ratio of each power consumption allocation imbalance module, which specifically includes:

[0114] determining the initial energy allocation ratio of each power consumption allocation imbalance module according to the ratio between the actual output power of each power consumption allocation imbalance module and the current total chip power consumption;

[0115] determining the output power upper limit of each functional module and the total chip power consumption upper limit according to the chip power consumption constraint condition;

[0116] determining the current energy allocation constraint condition based on the output power upper limit and the total chip power consumption upper limit;

[0117] generate an initial population containing a plurality of chromosomes randomly in a preset initialization interval based on the initial energy distribution ratio; wherein the chromosomes contain a plurality of coding genes for representing the energy distribution ratio of each of the power consumption distribution imbalance modules and the energy distribution ratio of the remaining functional modules other than the power consumption distribution imbalance modules;

[0118] evaluate the fitness of each of the chromosomes in the initial population based on a fitness function to obtain a fitness evaluation result; wherein the fitness function is the reciprocal of the sum of squares of the difference between the energy distribution value of the functional module and the upper limit of its output power consumption; the energy distribution value is the product of the energy distribution ratio and the upper limit of the total power consumption of the chip;

[0119] update the initial population iteratively using a genetic algorithm based on the fitness evaluation result until the fitness evaluation result meets a preset iteration termination condition, and determine the target energy distribution ratio of each of the power consumption distribution imbalance modules according to the chromosome with the highest current fitness value.

[0120] Specifically, the embodiment first determines the initial energy distribution proportion of each power consumption distribution imbalance module according to the ratio between the actual output power consumption of each power consumption distribution imbalance module and the current total chip power consumption. It can be understood that although the actual output power consumption of each power consumption distribution imbalance module is imbalanced, the actual output power consumption of each power consumption distribution imbalance module generally does not deviate too much from the rated output power consumption. However, if the initial population of each chromosome is simply randomly generated in the process of executing the genetic algorithm, there may be many chromosomes in which the energy distribution proportion represented by each gene code is unreasonable, resulting in many invalid solutions. Therefore, the embodiment generates an initial population containing a plurality of chromosomes in a preset initialization interval based on the initial energy distribution proportion. The initialization interval can be set according to actual needs. For example, taking the initial energy distribution proportion as the reference proportion, introducing a proportion adjustment coefficient, and defining an initialization interval in which the ratio between the absolute value of the difference between the proportion value in the initialization interval and the initial energy distribution proportion and the initial energy distribution proportion is less than or equal to the proportion adjustment coefficient. For example, assuming that the initial energy distribution proportion is [0.5, 0.3, 0.2], wherein 0.5 and 0.3 represent the initial energy distribution proportions of two power consumption distribution imbalance modules, and 0.2 represents the initial energy distribution proportion of the remaining functional modules other than the two power consumption distribution imbalance modules, and the current proportion adjustment coefficient is 10%, one of the initialized chromosomes can be coded as [0.48, 0.32, 0.2], wherein 0.48 and 0.32 represent the energy distribution proportions of the two power consumption distribution imbalance modules, and 0.2 represents the energy distribution proportion of the remaining functional modules other than the two power consumption distribution imbalance modules. The variation values of the three energy distribution proportions compared with the initial energy distribution proportion all satisfy the current proportion adjustment coefficient, so as to narrow the search range of the genetic algorithm, avoid wasting the iteration number in the invalid area, and improve the optimization efficiency.

[0121] Further, the fitness function in the embodiment is the reciprocal of the sum of squares of the difference between the energy allocation value of a functional module and its upper limit of output power consumption. When the energy allocation value of a certain functional module approaches its upper limit of output power consumption, the fitness value decreases, thereby guiding the genetic algorithm to find a more balanced allocation scheme. After fitness evaluation of each chromosome in the initial population, the initial population is iteratively updated based on the fitness evaluation results using the genetic algorithm, which involves selection operation, crossover operation and mutation operation. Specifically, the selection operation adopts the roulette wheel selection method, that is, parent chromosomes are selected according to the fitness values using the roulette wheel selection method. Chromosomes with high fitness values have a higher probability of being selected. The specific steps are as follows: calculate the fitness value of each chromosome; calculate the cumulative fitness value of each chromosome; generate a random number and select a chromosome according to the cumulative fitness value. Through the crossover operation, two parent chromosomes are randomly selected, and the encoding genes are exchanged according to the preset crossover probability (such as 80%) to generate a new allocation combination, i.e. a child chromosome. Through the mutation operation, a chromosome that needs to be mutated is selected according to the preset mutation probability (such as 5%), and then a random disturbance is added to a certain encoding gene of the selected chromosome to increase the diversity of the population and avoid falling into a local optimum, and it is necessary to ensure that the adjusted value still satisfies the constraint condition of energy allocation ratio, i.e. the sum of the values of all encoding gene bits is 1. Then the fitness value of each chromosome in the current new population is calculated, and the chromosomes with high fitness values are selected into the next generation population according to the fitness values, until the fitness evaluation results meet the preset iteration termination condition, for example, when the change rate of the optimal fitness value of 10 consecutive generations is less than 0.01, it is considered that the genetic algorithm has converged. After the iteration is terminated, the values of the encoding gene bits in the chromosome with the highest current fitness value determine the target energy allocation ratio of each power consumption allocation imbalance module.

[0122] As a preferred solution, the power consumption limit value of each power consumption allocation imbalance module is determined according to the target energy allocation ratio, specifically including:

[0123] The power consumption limit value of each power consumption allocation imbalance module is determined according to the product of the upper limit of the total chip power consumption in the chip power consumption constraint condition and each target energy allocation ratio.

[0124] Specifically, assuming that the upper limit of the total chip power consumption in the chip power consumption constraint condition is 150W, there are currently three power consumption allocation imbalance modules, namely GPU, CPU and cache controller, the target energy allocation ratio of GPU is 0.4, the target energy allocation ratio of CPU is 0.3, and the target energy allocation ratio of the cache controller is 0.2, then the power consumption limit value corresponding to the GPU is 60W, the power consumption limit value corresponding to the CPU is 45W, and the power consumption limit value corresponding to the cache controller is 30W.

[0125] It is worth mentioning that when adjusting the operating parameters of the power consumption distribution imbalance module to meet the power consumption limit value, it can be achieved by dynamically adjusting the working frequency or voltage. For example, assuming that the power consumption limit value of the GPU is 60W, but its current actual operating power consumption is 70W, the power consumption output of the GPU can be reduced by reducing its working frequency, such as from 1.5GHz to 1.2GHz. Such adjustment is usually completed by the power management unit inside the chip, which monitors and dynamically adjusts the parameters of each module in real time to ensure that the power consumption is stable within the limit range. For example, the current sensor and voltage sensor inside the chip can collect data multiple times per second to form a high-precision operating state data set. Assuming that the current value of the GPU is 5A and the voltage value is 12V, its real-time operating power consumption is 60W.

[0126] Please refer to Figure 2 The second aspect of the embodiment of the application provides a power consumption adjustment system of an RFID tag chip, which comprises:

[0127] A thermal effect interference distribution analysis module 11 is configured to determine thermal effect interference distribution information of a target RFID tag chip according to structure parameters and surface temperature distribution data of the target RFID tag chip.

[0128] A power consumption distribution imbalance module determination module 12 is configured to determine a plurality of power consumption distribution imbalance modules based on the thermal effect interference distribution information and temperature detection data of each functional module in the target RFID tag chip.

[0129] A data interaction configuration module 13 is configured to determine a data interaction configuration scheme of each power consumption distribution imbalance module according to power consumption data of each power consumption distribution imbalance module and preset chip configuration parameters.

[0130] An energy distribution optimization module 14 is configured to perform energy distribution optimization on each power consumption distribution imbalance module according to the data interaction configuration scheme and a preset chip power consumption constraint condition, and determine a target energy distribution ratio of each power consumption distribution imbalance module.

[0131] An output power consumption adjustment module 15 is configured to determine a power consumption limit value of each power consumption distribution imbalance module according to the target energy distribution ratio, and adjust the output power consumption of each power consumption distribution imbalance module according to the power consumption limit value.

[0132] As a preferred solution, the thermal effect interference distribution analysis module 11 is configured to determine thermal effect interference distribution information of a target RFID tag chip according to structure parameters and surface temperature distribution data of the target RFID tag chip, and specifically comprises:

[0133] determine assembly positions, heat generation powers, surface areas, heat transfer distances and heat transfer coefficients of each of the functional modules in the target RFID tag chip according to the structure parameters;

[0134] determine heat generation influence radii of each of the functional modules according to the heat generation powers, the surface areas, the heat transfer distances and the heat transfer coefficients;

[0135] determine heat generation influence areas of each of the functional modules according to the assembly positions and the heat generation influence radii, and determine at least one heat effect superposition area in the target RFID tag chip based on overlapping areas between each of the heat generation influence areas;

[0136] determine a plurality of surface temperature collection points in the heat effect superposition area and surface temperature values corresponding to each of the surface temperature collection points based on the surface temperature distribution data;

[0137] determine whether each of the heat effect superposition areas is a heat effect interference area according to the surface temperature values and a preset surface temperature threshold value;

[0138] when any one of the heat effect superposition areas is the heat effect interference area, determine the heat effect interference distribution information according to position information of the heat effect interference area and a plurality of target functional modules corresponding to the heat effect interference area.

[0139] As a preferred solution, the power consumption allocation imbalance module determination module 12 is configured to determine a plurality of power consumption allocation imbalance modules based on the heat effect interference distribution information and temperature detection data of each functional module in the target RFID tag chip, and specifically includes:

[0140] determine a plurality of target functional modules corresponding to each of the heat effect interference areas according to the heat effect interference distribution information;

[0141] obtain a target temperature detection value sequence of each of the target functional modules in a preset temperature detection period from the temperature detection data of each of the functional modules;

[0142] input the target temperature detection value sequence into a preset autoregressive moving average model to obtain a power consumption value sequence corresponding to the target temperature detection value sequence of each of the target functional modules; wherein the autoregressive moving average model is obtained based on historical temperature detection data and historical power consumption data of each of the functional modules in a preset historical time period;

[0143] determine whether the output power distribution of the target function modules corresponding to each of the thermal effect interference regions is unbalanced according to the real-time total chip power consumption at each time point in the power consumption value sequence, the preset upper limit of power consumption proportion, the preset lower limit of power consumption proportion of each of the target function modules, and the power consumption value sequence;

[0144] When the output power distribution of the target function modules corresponding to any one of the thermal effect interference regions is unbalanced, the target function modules corresponding to the any one of the thermal effect interference regions are determined as the power consumption distribution imbalance modules.

[0145] As a preferred solution, the data interaction configuration module 13 is configured to determine a data interaction configuration scheme of each of the power consumption distribution imbalance modules according to the power consumption data of each of the power consumption distribution imbalance modules and preset chip configuration parameters, and specifically includes:

[0146] determine a data interaction priority coefficient of each of the power consumption distribution imbalance modules according to the power consumption data of each of the power consumption distribution imbalance modules;

[0147] determine the data interaction configuration scheme of each of the power consumption distribution imbalance modules according to the data interaction priority coefficient and the chip configuration parameters.

[0148] As a preferred solution, the data interaction configuration module 13 is configured to determine a data interaction priority coefficient of each of the power consumption distribution imbalance modules according to the power consumption data of each of the power consumption distribution imbalance modules, and specifically includes:

[0149] determine actual output power consumption and rated output power consumption of each of the power consumption distribution imbalance modules according to the power consumption data;

[0150] determine the data interaction priority coefficient of each of the power consumption distribution imbalance modules according to the ratio between the actual output power consumption and the rated output power consumption.

[0151] As a preferred solution, the data interaction configuration module 13 is configured to determine a data interaction configuration scheme of each of the power consumption distribution imbalance modules according to the data interaction priority coefficient and the chip configuration parameters, and specifically includes:

[0152] According to the chip configuration parameter, determine the interconnection bandwidth parameter between each of the power consumption allocation imbalance modules, the task emergency coefficient to be processed and the data interaction coefficient; wherein, the task emergency coefficient to be processed is determined based on the difference between the preset processing deadline of the current task to be processed of the power consumption allocation imbalance module and the current time; the data interaction coefficient is used to represent the data interaction probability between each of the power consumption allocation imbalance modules, and the data interaction coefficient is determined based on the historical data interaction record of the power consumption allocation imbalance module;

[0153] According to the interconnection bandwidth parameter, the task emergency coefficient to be processed, the data interaction coefficient and the data interaction priority coefficient, determine the data interaction demand weight between each of the power consumption allocation imbalance modules;

[0154] According to the data interaction demand weight and the current residual bandwidth resource, determine the bandwidth configuration parameter of the data interaction path between each of the power consumption allocation imbalance modules;

[0155] According to the bandwidth configuration parameter of each of the data interaction paths, determine the data interaction configuration scheme of each of the power consumption allocation imbalance modules.

[0156] As a preferred scheme, the energy allocation optimization module 14 is used to perform energy allocation optimization on each of the power consumption allocation imbalance modules according to the data interaction configuration scheme and the preset chip power consumption constraint condition, and determine the target energy allocation proportion of each of the power consumption allocation imbalance modules, specifically including:

[0157] According to the data interaction configuration scheme, determine the bandwidth configuration parameter of the data interaction path between each of the power consumption allocation imbalance modules;

[0158] Based on the bandwidth configuration parameter and the preset unit bandwidth power consumption coefficient, determine the unit time energy consumption value of each of the data interaction paths;

[0159] According to the actual output power consumption of each of the functional modules and the unit time energy consumption value, determine the current total chip power consumption of the target RFID tag chip;

[0160] According to the actual output power consumption of each of the power consumption allocation imbalance modules, the current total chip power consumption and the chip power consumption constraint condition, use genetic algorithm to perform energy allocation optimization on each of the power consumption allocation imbalance modules, and determine the target energy allocation proportion of each of the power consumption allocation imbalance modules.

[0161] As a preferred solution, the energy distribution optimization module 14 is configured to determine the target energy distribution ratio of each of the power distribution imbalance modules by using a genetic algorithm to perform energy distribution optimization on each of the power distribution imbalance modules according to the actual output power consumption of each of the power distribution imbalance modules, the current total chip power consumption, and the chip power consumption constraint condition, and specifically includes the following steps:

[0162] determining an initial energy distribution ratio of each of the power distribution imbalance modules according to a ratio between the actual output power consumption of each of the power distribution imbalance modules and the current total chip power consumption;

[0163] determining an upper limit of output power consumption of each of the functional modules and an upper limit of total chip power consumption according to the chip power consumption constraint condition;

[0164] determining a current energy distribution constraint condition based on the upper limit of output power consumption and the upper limit of total chip power consumption;

[0165] generating an initial population containing a plurality of chromosomes in a preset initialization interval based on the initial energy distribution ratio; wherein the chromosomes contain a plurality of coding genes for representing the energy distribution ratio of each of the power distribution imbalance modules and the energy distribution ratio of the remaining functional modules other than the power distribution imbalance modules;

[0166] performing fitness evaluation on each of the chromosomes in the initial population based on a fitness function to obtain a fitness evaluation result; wherein the fitness function is an inverse of a sum of squares of differences between energy distribution values of the functional modules and the upper limit of output power consumption thereof; the energy distribution value is a product of the energy distribution ratio and the upper limit of total chip power consumption;

[0167] iteratively updating the initial population by using a genetic algorithm based on the fitness evaluation result until the fitness evaluation result meets a preset iteration termination condition, and determining the target energy distribution ratio of each of the power distribution imbalance modules according to a chromosome with the highest current fitness value.

[0168] As a preferred solution, the output power adjustment module 15 is configured to determine a power consumption limit value of each of the power distribution imbalance modules according to the target energy distribution ratio, and specifically includes the following steps:

[0169] determining the power consumption limit value of each of the power distribution imbalance modules according to a product of the upper limit of total chip power consumption in the chip power consumption constraint condition and each of the target energy distribution ratios.

[0170] The power consumption adjustment system of the RFID tag chip provided by the embodiment of the present application can accurately and effectively perform power consumption distribution optimization on the power consumption distribution imbalance modules in the presence of thermal effect interference, while ensuring smooth data interaction of each power consumption distribution imbalance module, thereby ensuring the stability and performance of the target RFID tag chip, and helping to meet the demand for coexistence of high performance and miniaturization of the RFID tag chip.

[0171] The above is the preferred embodiment of the present application. It should be noted that those skilled in the art can make several improvements and refinements without departing from the principles of the present application, and these improvements and refinements are also considered within the scope of protection of the present application.

Claims

1. A method for adjusting the power consumption of an RFID tag chip, characterized in that, include: Based on the structural parameters and surface temperature distribution data of the target RFID tag chip, the thermal effect interference distribution information of the target RFID tag chip is determined; Based on the thermal effect interference distribution information and the temperature detection data of each functional module in the target RFID tag chip, several power consumption imbalance modules are identified. Based on the power consumption data of each of the power consumption imbalance modules and the preset chip configuration parameters, determine the data interaction configuration scheme of each of the power consumption imbalance modules; Based on the data interaction configuration scheme and the preset chip power consumption constraints, energy allocation optimization is performed on each of the power distribution imbalance modules to determine the target energy allocation ratio of each of the power distribution imbalance modules. Based on the target energy allocation ratio, the power consumption limit value of each of the power consumption allocation imbalance modules is determined, and the output power consumption of each of the power consumption allocation imbalance modules is adjusted according to the power consumption limit value.

2. The power consumption adjustment method for the RFID tag chip as described in claim 1, characterized in that, The step of determining the thermal interference distribution information of the target RFID tag chip based on its structural parameters and surface temperature distribution data specifically includes: Based on the structural parameters, determine the assembly position, heating power, surface area, heat transfer distance, and thermal conductivity of each functional module within the target RFID tag chip; The heating influence radius of each functional module is determined based on the heating power, the surface area, the heat transfer distance, and the thermal conductivity. Based on the assembly position and the heat-affected radius, the heat-affected area of ​​each functional module is determined, and based on the overlapping area between the heat-affected areas, at least one thermal effect superposition area within the target RFID tag chip is determined. Based on the surface temperature distribution data, several surface temperature acquisition points in the superimposed thermal effect area and the surface temperature value corresponding to each surface temperature acquisition point are determined. Based on the surface temperature value and the preset surface temperature threshold, it is determined whether each of the thermal effect superposition regions is a thermal effect interference region; When any thermal effect superposition region exists as the thermal effect interference region, the thermal effect interference distribution information is determined based on the location information of the thermal effect interference region and the several target functional modules corresponding to the thermal effect interference region.

3. The power consumption adjustment method for the RFID tag chip as described in claim 2, characterized in that, Based on the thermal effect interference distribution information and the temperature detection data of each functional module within the target RFID tag chip, several power consumption imbalance modules are identified, specifically including: Based on the thermal effect interference distribution information, determine a number of target functional modules corresponding to each thermal effect interference region; Obtain the target temperature detection value sequence of each target functional module within a preset temperature detection period from the temperature detection data of each functional module. The target temperature detection value sequence is input into a preset autoregressive moving average model to obtain the power consumption value sequence corresponding to the target temperature detection value sequence of each target functional module; wherein, the autoregressive moving average model is obtained by training on the historical temperature detection data and historical power consumption data of each functional module within a preset historical time period; Based on the real-time total chip power consumption at each moment in the power consumption value sequence, the preset upper limit of the power consumption ratio of each target functional module, the preset lower limit of the power consumption ratio, and the power consumption value sequence, it is determined whether the output power consumption distribution of several target functional modules corresponding to each thermal effect interference region is unbalanced. When there is an imbalance in the output power distribution of several target functional modules corresponding to any thermal effect interference region, the several target functional modules corresponding to any thermal effect interference region are identified as the power distribution imbalance module.

4. The power consumption adjustment method for the RFID tag chip as described in claim 1, characterized in that, The step of determining the data interaction configuration scheme for each of the power distribution imbalance modules based on the power consumption data of each module and preset chip configuration parameters specifically includes: Based on the power consumption data of each of the power consumption imbalance modules, determine the data interaction priority coefficient of each of the power consumption imbalance modules; Based on the data interaction priority coefficient and the chip configuration parameters, the data interaction configuration scheme for each of the power consumption imbalance modules is determined.

5. The power consumption adjustment method for the RFID tag chip as described in claim 4, characterized in that, The step of determining the data interaction priority coefficient of each power allocation imbalance module based on the power consumption data of each power allocation imbalance module specifically includes: Based on the power consumption data, determine the actual output power consumption and rated output power consumption of each of the power distribution imbalance modules; The data interaction priority coefficient of each of the power distribution imbalance modules is determined based on the ratio between the actual output power consumption and the rated output power consumption.

6. The power consumption adjustment method for the RFID tag chip as described in claim 4, characterized in that, The step of determining the data interaction configuration scheme for each of the power consumption imbalance modules based on the data interaction priority coefficient and the chip configuration parameters specifically includes: Based on the chip configuration parameters, the interconnection bandwidth parameters, the urgency coefficient of the pending tasks, and the data interaction coefficient between each of the power allocation imbalance modules are determined; wherein, the urgency coefficient of the pending tasks is determined based on the difference between the preset processing deadline time and the current time of the current pending task of the power allocation imbalance module; the data interaction coefficient is used to represent the data interaction probability between each of the power allocation imbalance modules, and the data interaction coefficient is determined based on the historical data interaction records of the power allocation imbalance modules. Based on the interconnection bandwidth parameters, the urgency coefficient of the task to be processed, the data interaction coefficient, and the data interaction priority coefficient, the weight of the data interaction demand between each of the power consumption distribution imbalance modules is determined. Based on the data interaction demand weight and the current remaining bandwidth resources, determine the bandwidth configuration parameters of the data interaction path between each of the power consumption imbalance modules; Based on the bandwidth configuration parameters of each of the data interaction paths, the data interaction configuration scheme of each of the power consumption distribution imbalance modules is determined.

7. The power consumption adjustment method for the RFID tag chip as described in claim 6, characterized in that, The step of optimizing energy allocation for each of the power imbalance modules based on the data interaction configuration scheme and preset chip power consumption constraints, and determining the target energy allocation ratio for each of the power imbalance modules, specifically includes: Based on the data interaction configuration scheme, determine the bandwidth configuration parameters of the data interaction path between each of the power consumption distribution imbalance modules; Based on the bandwidth configuration parameters and the preset unit bandwidth power consumption coefficient, the unit time energy consumption value of each data interaction path is determined. The current total power consumption of the target RFID tag chip is determined based on the actual output power consumption of each functional module and the energy consumption value per unit time. Based on the actual output power of each of the power distribution imbalance modules, the current total power consumption of the chip, and the chip power consumption constraints, a genetic algorithm is used to optimize the energy allocation of each of the power distribution imbalance modules to determine the target energy allocation ratio of each of the power distribution imbalance modules.

8. The power consumption adjustment method for an RFID tag chip as described in claim 7, characterized in that, The step involves using a genetic algorithm to optimize the energy allocation of each power imbalance module based on its actual output power consumption, the current total chip power consumption, and the chip power consumption constraints, thereby determining the target energy allocation ratio for each power imbalance module. Specifically, this includes: The initial energy allocation ratio of each of the power allocation imbalance modules is determined based on the ratio between the actual output power of each module and the current total power consumption of the chip. Based on the chip power consumption constraints, determine the upper limit of output power consumption for each functional module and the upper limit of total chip power consumption; Based on the upper limit of output power consumption and the upper limit of total chip power consumption, the current energy allocation constraints are determined; Based on the initial energy allocation ratio, an initial population containing several chromosomes is randomly generated within a preset initialization interval; wherein, the chromosome contains several coding genes for representing the energy allocation ratio of each of the power allocation imbalance modules and the energy allocation ratio of the other functional modules besides the power allocation imbalance modules. Based on the fitness function, the fitness of each chromosome in the initial population is evaluated to obtain the fitness evaluation result; wherein, the fitness function is the reciprocal of the sum of the squares of the differences between the energy allocation value of the functional module and its output power consumption limit; the energy allocation value is the product of the energy allocation ratio and the chip's total power consumption limit; Based on the fitness evaluation results, the initial population is iteratively updated using a genetic algorithm until the fitness evaluation results meet the preset iteration termination conditions. Based on the chromosome with the highest current fitness value, the target energy allocation ratio of each of the power consumption distribution imbalance modules is determined.

9. The power consumption adjustment method for an RFID tag chip as described in claim 1, characterized in that, The step of determining the power consumption limit value for each of the power consumption imbalance modules based on the target energy allocation ratio specifically includes: The power consumption limit value of each of the power consumption imbalance modules is determined by multiplying the upper limit of the total power consumption of the chip in the chip power consumption constraint conditions and the product of each of the target energy allocation ratios.

10. A power consumption adjustment system for an RFID tag chip, characterized in that, include: The thermal effect interference distribution analysis module is used to determine the thermal effect interference distribution information of the target RFID tag chip based on the structural parameters and surface temperature distribution data of the target RFID tag chip. The power distribution imbalance module determination module is used to determine several power distribution imbalance modules based on the thermal effect interference distribution information and the temperature detection data of each functional module in the target RFID tag chip. The data interaction configuration module is used to determine the data interaction configuration scheme of each of the power distribution imbalance modules based on the power consumption data of each of the power distribution imbalance modules and the preset chip configuration parameters. An energy allocation optimization module is used to optimize the energy allocation of each of the power allocation imbalance modules according to the data interaction configuration scheme and the preset chip power consumption constraints, and to determine the target energy allocation ratio of each of the power allocation imbalance modules. The output power consumption adjustment module is used to determine the power consumption limit value of each of the power consumption imbalance modules according to the target energy distribution ratio, and adjust the output power consumption of each of the power consumption imbalance modules according to the power consumption limit value.

Images