Component temperature regulation method, device, medium and air conditioner

Abstract

This invention discloses a method, device, medium, and air conditioner for regulating the temperature of components. First, temperature correlation data related to the future temperature of a target component in the air conditioner is acquired. A first frequency component is obtained through modal decomposition, and a second frequency component is obtained through component prediction. The second frequency component is then integrated and reconstructed to obtain the predicted temperature of the target component. Finally, appropriate target regulation parameters are matched to regulate the target circuit, thereby achieving temperature regulation of the target component. This scheme realizes the prediction and early regulation of component temperature, enabling targeted regulation measures to be taken before a significant increase in component temperature due to instantaneous large impacts such as power grid surges, effectively avoiding the risk of component damage due to high temperatures. Simultaneously, this framework significantly improves the accuracy of temperature prediction, allowing temperature rise intervention to occur before the actual temperature rise, fundamentally solving the lag problem of traditional feedback control.

Description

Technical Field

[0001] This invention relates to the field of air conditioner technology, and in particular to a method, apparatus, medium, and air conditioner for regulating the temperature of components. Background Technology

[0002] In the circuitry of an air conditioner, some components generate significant heat during actual operation, especially under high-power and harsh power grid conditions.

[0003] Existing technologies typically only execute corresponding response actions after the temperature of these components has risen significantly. This can lead to some components being easily damaged due to excessive temperature during sudden surges in power grids, which in turn can cause the air conditioner to malfunction. Summary of the Invention

[0004] Therefore, it is necessary to provide methods, devices, media, and air conditioners for regulating the temperature of components to solve the problem that components are prone to high-temperature damage when subjected to sudden large impacts due to the lag in temperature control response in the prior art.

[0005] In a first aspect, embodiments of this application provide a method for regulating the temperature of a component, the method comprising: Acquire temperature-related data; wherein, the temperature-related data represents data relating the current temperature to the future temperature of the target component; The temperature-correlated data is subjected to mode decomposition to obtain the first frequency component of the temperature-correlated data; Based on the first frequency component, component prediction is performed to obtain the second frequency component of the target component. The second frequency component is integrated and reconstructed to obtain the predicted temperature of the target component; Determine the target adjustment parameters that are compatible with the predicted temperature, and adjust the target circuit based on the target adjustment parameters to regulate the temperature of the target components.

[0006] In some embodiments of this application, performing mode decomposition processing on the temperature-correlated data to obtain a first frequency component of the temperature-correlated data includes: Based on the preset number of superpositions and the white noise amplitude coefficient, the temperature correlation data and white noise data are superimposed to obtain multiple sets of superimposed data. Each set of superimposed data is sieved and decomposed to obtain multiple intrinsic components and trend components of each set of superimposed data. The average value of multiple eigencomponents of the multiple sets of superimposed data is calculated within the same order eigencomponent to obtain the first target eigencomponent of each order. The average value of the trend components of the multiple sets of superimposed data is calculated to obtain the first target trend component; The first target intrinsic component and the first target trend component of each order are determined as the first frequency component of the temperature correlation data.

[0007] In some embodiments of this application, the first frequency component includes first target intrinsic components and first target trend components of each order, and the second frequency component obtained by component prediction based on the first frequency component includes: Based on the preset filtering rules, the first target intrinsic components and the first target trend components of each order are filtered to obtain the filtered first target intrinsic components and the first target trend components. The filtered first target intrinsic components are input into the corresponding first processing model for processing to obtain the second target intrinsic components. The filtered first target trend component is input into the second processing model for processing to obtain the second target trend component. The second target intrinsic component and the second target trend component are determined as the second frequency component.

[0008] In some embodiments of this application, the first processing model includes a gated loop layer, a first random deactivation layer, and a first fully connected layer connected in sequence; The gated recurrent layer is configured to perform feature filtering on the intrinsic components of the first target; The first random deactivation layer is configured to randomly deactivate the feature representation output by the gated recurrent layer; The first fully connected layer is configured to perform a weighted calculation on the feature representation output by the first randomly deactivated layer to obtain the second target intrinsic component; The second processing model includes multiple fully connected hidden layers, a second random deactivation layer, and a second fully connected layer connected in sequence; The plurality of fully connected hidden layers are configured to perform linear weighting calculations and nonlinear transformations on the first target trend component; The second random deactivation layer is configured to randomly deactivate the feature representations output by the plurality of fully connected hidden layers; The second fully connected layer is configured to perform weighted calculations on the feature representation output by the second random deactivation layer to obtain the second target trend component.

[0009] In some embodiments of this application, determining the target adjustment parameter adapted to the predicted temperature includes: The prediction confidence level is calculated based on the predicted temperature and the measured temperature at the corresponding time. When the prediction confidence level is greater than the confidence level threshold, the target cost value corresponding to each candidate adjustment parameter is determined based on the predicted temperature, and the target adjustment parameter is determined from the candidate adjustment parameters based on the target cost value. When the predicted confidence level is less than or equal to the confidence level threshold, the target adjustment parameter corresponding to the measured temperature is determined by PID control.

[0010] In some embodiments of this application, the step of determining the target cost value corresponding to each candidate adjustment parameter based on the predicted temperature, and determining the target adjustment parameter from the candidate adjustment parameters based on the target cost value, includes: Calculate the difference between the predicted temperature and the safe temperature corresponding to each candidate adjustment parameter to obtain the safety cost. Determine the power cost and comfort cost corresponding to each candidate adjustment parameter; The safety cost, power cost, and comfort cost are weighted and calculated to obtain the target cost for each candidate adjustment parameter; The candidate adjustment parameter corresponding to the minimum target cost is determined as the target adjustment parameter.

[0011] In some embodiments of this application, the second frequency component is predicted based on a first processing model and a second processing model. After determining the target adjustment parameter adapted to the predicted temperature and adjusting the target circuit based on the target adjustment parameter to adjust the temperature of the target component, the method further includes: The loss value is calculated based on the predicted temperature and the actual measured temperature at the corresponding time. When the loss value is greater than the loss value threshold and the air conditioner is in a stable operating state, the model parameters of the first processing model and the model parameters of the second processing model are adjusted based on the loss value.

[0012] Secondly, embodiments of this application also provide a component temperature regulating device, the component temperature regulating device comprising: The data acquisition module is used to acquire temperature-related data; wherein, the temperature-related data represents data that is currently associated with the future temperature of the target component; The decomposition processing module is used to perform modal decomposition processing on the temperature-related data to obtain the first frequency component of the temperature-related data. A component prediction module is used to perform component prediction based on the first frequency component to obtain the second frequency component of the target component. An integration and reconstruction module is used to integrate and reconstruct the second frequency component to obtain the predicted temperature of the target component; A temperature regulation module is used to determine a target regulation parameter that is compatible with the predicted temperature, and to adjust the target circuit based on the target regulation parameter in order to regulate the temperature of the target component.

[0013] Thirdly, this application also provides an air conditioner, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the steps in the above-described component temperature regulation method.

[0014] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps in the above-described component temperature regulation method.

[0015] Fifthly, embodiments of this application also provide a computer program product or computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the methods provided in the various optional implementations described in embodiments of this application.

[0016] This invention provides a method, device, medium, and air conditioner for regulating the temperature of components. The core of this invention achieves proactive temperature control through an innovative decomposition-prediction-reconstruction framework: first, temperature-related data concerning the future temperature of the target components in the air conditioner is acquired; then, a first frequency component is obtained through modal decomposition; a second frequency component is obtained through component prediction; finally, the second frequency component is integrated and reconstructed to obtain the predicted temperature of the target component; and finally, appropriate target regulation parameters are matched to adjust the target circuit, thereby regulating the temperature of the target component. This solution enables the prediction and early adjustment of component temperatures, allowing targeted adjustment measures to be taken before a significant increase in component temperature due to instantaneous large impacts such as power grid surges, effectively avoiding the risk of component damage due to high temperatures. Simultaneously, this framework significantly improves the accuracy of temperature prediction, ensuring that temperature rise intervention occurs before the actual temperature rise, fundamentally solving the lag problem of traditional feedback control, and significantly improving the stability and service life of the target components and even the entire air conditioner. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] in: Figure 1 This is a flowchart illustrating the method for regulating the temperature of components. Figure 2 This is a flowchart illustrating the modal decomposition process for temperature-related data. Figure 3 This is a flowchart illustrating the component prediction process based on the first frequency component. Figure 4 A flowchart illustrating the process of determining target adjustment parameters adapted to the predicted temperature; Figure 5 This is a schematic diagram of the temperature control device for components. Figure 6 This is a structural block diagram of an air conditioner. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] The terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0021] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0022] This invention provides a method, apparatus, medium, and air conditioner for regulating the temperature of electronic components. In some embodiments of this application, the provided method for regulating the temperature of electronic components can be applied to air conditioners. Specifically, the air conditioner can be applied to different scenarios, including but not limited to industrial air conditioners or household air conditioners. In some embodiments of this application, the air conditioner can be a single unit, such as a cabinet air conditioner or a wall-mounted air conditioner; in some embodiments of this application, the air conditioner can also be a central air conditioning system composed of multiple air conditioning units, such as a multi-split air conditioner, an air-cooled heat pump system, or an air conditioning system with heat recovery function.

[0023] Please see Figure 1 , Figure 1 This is a flowchart illustrating a component temperature regulation method provided in an embodiment of this application. Although the logical sequence is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than that shown in the figures. Specifically, the component temperature regulation method includes steps S101-S105, as follows: S101, acquire temperature-related data.

[0024] Among them, temperature-related data represents data relating the current temperature to the future temperature of the target component, which is a component in the target circuit within the air conditioner. The target circuit is a specific functional circuit that contains the target component, and whose operating state is directly related to the temperature of the target component.

[0025] Optionally, the target circuit can be any one of a single-phase active power factor correction circuit, a compressor frequency converter drive circuit, a fan speed control circuit, or a power supply voltage regulator circuit. The target component can be any one of the following: a common-mode inductor in a single-phase active power factor correction circuit, a MOSFET in a compressor drive circuit, a capacitor in a filter circuit, or a microcontroller chip in a control circuit.

[0026] Optionally, after the air conditioner is turned on, the control system begins to collect or calculate the following data in real time as temperature-related data, including but not limited to: load current. Circuit switching frequency Compressor operating frequency Current temperature of the target component Indoor temperature Outdoor condenser temperature Cooling fan speed Load current change rate Indoor temperature change rate and compressor predicted frequency change rate The sampling frequency can be set to 10Hz, and the sampling time window is 5 seconds.

[0027] Optionally, the compressor predicts the rate of change of frequency. This can be determined in the following way: if ,So .if ,So .if ,So Among them, a, b, c, d, and , , Specifically, it can be set as a=2, b=-0.1, c=0.5, d=35. =1, =-1, =0.5.

[0028] By acquiring the aforementioned multi-dimensional temperature correlation data, the temperature variation pattern of the target component can be fully characterized, providing sufficient feature basis for subsequent frequency component decomposition and dimensional prediction.

[0029] S102, perform mode decomposition on the temperature-correlated data to obtain the first frequency component of the temperature-correlated data.

[0030] The first frequency component refers to the characteristic components of the temperature-related data under different frequency dimensions after modal decomposition.

[0031] Optionally, temperature-related data can be decomposed into high-frequency detail components and low-frequency approximation components at different scales based on a preset wavelet basis function.

[0032] In some embodiments of this application, such as Figure 2 As shown, step S201-S205 involves performing mode decomposition on the temperature-related data in step S102 to obtain the first frequency component of the temperature-related data. The specific steps are as follows: S201, based on the preset number of superpositions and white noise amplitude coefficient, superimposes the temperature correlation data and white noise data to obtain multiple sets of superimposed data.

[0033] Optionally, considering that air conditioning load fluctuations are more severe than in typical industrial applications, the white noise amplitude factor is set to 0.15. The number of superpositions is set to 150 to improve decomposition stability.

[0034] For example, assuming the original temperature correlation data is D=[d1, d2, ..., dn], and each time a set of random white noise Nk=[n1, n2, ..., nm] (k=1 to 150) is generated, then the kth set of superimposed data can be Dk=D+0.15×Nk, and finally 150 sets of superimposed data D1~D150 are obtained.

[0035] S202, perform sieving and decomposition processing on each set of superimposed data to obtain multiple intrinsic components and trend components of each set of superimposed data.

[0036] Optionally, for each set of superimposed data (such as D1) generated by S201, a sieve decomposition process based on "Empirical Mode Decomposition (EMD)" is performed to finally separate the "multiple intrinsic components (IMF)" and "1 trend component" of the data set.

[0037] Optionally, the screening and decomposition process includes: identifying all local maxima and minima for a set of superimposed data Dk; fitting the maxima with cubic spline interpolation to obtain the upper envelope, and fitting the minima to obtain the lower envelope; calculating the mean of the upper and lower envelopes, and subtracting the mean from Dk to obtain a candidate component; repeating the above steps until the candidate component satisfies the IMF condition, which is an IMF for the set of data; subtracting the IMF from Dk, and repeating the screening for the remaining signal until no more IMFs can be extracted, the remaining signal being the trend component. Ultimately, each set of superimposed data will be decomposed into m IMFs (e.g., IMF1-IMF6) and 1 trend component (R1), resulting in 150×m IMFs and 150 trend components for 150 sets of data.

[0038] Optionally, the IMF conditions can be: 1. The number of extreme points and the number of zero crossings of the component are equal or differ by no more than 1 throughout the entire time series; 2. At any time, the mean of the upper envelope formed by the local maximum and the lower envelope formed by the local minimum is 0.

[0039] S203, calculate the average value of multiple eigencomponents of multiple sets of superimposed data within the same order eigencomponent to obtain the first target eigencomponent of each order.

[0040] Optionally, the first-order IMFs of the 150 sets of data are IMF1_1, IMF1_2, ..., IMF1_150, and their mean IMF1_avg = (IMF1_1 + IMF1_2 + ... + IMF1_150) / 150 is calculated. This mean is the "first-order first target eigencomponent". Similarly, the mean of the second to m-order IMFs is calculated to obtain the first target eigencomponent of each order (such as IMF1_avg ~ IMFm_avg).

[0041] S204, calculate the average value of the trend components of multiple sets of superimposed data to obtain the first target trend component.

[0042] Optionally, the first target trend component R_avg = (R1 + R2 + ... + R150) / 150.

[0043] S205, the first target intrinsic component and the first target trend component of each order are determined as the first frequency component of the temperature correlation data.

[0044] In other words, the first frequency component of the temperature-related data = [first-order first target eigencomponent, second-order first target eigencomponent, ..., m-th-order first target eigencomponent, first target trend component].

[0045] By superimposing and merging the means in the above embodiments, the stability and accuracy of the decomposition results can be improved, providing high-quality input data for subsequent dimensionality prediction.

[0046] S103, perform component prediction based on the first frequency component to obtain the second frequency component of the target component.

[0047] The second frequency component refers to the frequency characteristic component that corresponds to the future temperature change law of the target component, calculated by the prediction algorithm.

[0048] Optionally, the second frequency component of the target component can be obtained by predicting the components based on the first frequency component using models such as autoregressive moving average models, long short-term memory neural networks, and linear regression models.

[0049] In some embodiments of this application, the first frequency component includes first target intrinsic components and first target trend components of various orders, such as... Figure 3 As shown, S103 involves component prediction based on the first frequency component to obtain the second frequency component, which includes steps S301-S304, as follows: S301, based on the preset filtering rules, the first target intrinsic components and the first target trend components of each order are filtered to obtain the filtered first target intrinsic components and the first target trend components.

[0050] Optionally, the screening rule can be based on "energy percentage", "correlation coefficient", or "human preset", mainly to retain components that have a significant impact on the temperature change of the target component through screening.

[0051] For example, filtering can be based on energy percentage: calculate the proportion of the energy (such as signal variance) of each intrinsic component to the total energy, and retain components with a percentage higher than a threshold (such as 5%). This is because low-energy components are mostly noise, and removing them can reduce the amount of computation.

[0052] Alternatively, a correlation-based screening method can be used: calculate the correlation coefficient between each intrinsic component and the historical temperature of the target component, and retain components with an absolute correlation coefficient value higher than a threshold (e.g., 0.3). This is because components with low correlation contribute less to temperature prediction.

[0053] Alternatively, filtering can be based on pre-defined rules: components with different frequency characteristics are categorized and their corresponding valid data features are retained. For example, for high-frequency components (IMF1-IMF3), I... load f sw f comp ΔI load ΔT room， Δf comp_pred The relevant first target intrinsic components; for mid-to-low frequency components (IMF4-IMF6), retain I load f comp T room T cond , Δf comp_pred， Fan speed The relevant first objective eigencomponent; for the trend component, retain The corresponding first target trend component. This ensures that the filtered data can accurately match the input requirements of different subsequent prediction models.

[0054] S302, the filtered first target intrinsic component is input into the corresponding first processing model for processing to obtain the second target intrinsic component.

[0055] Optionally, all input first target intrinsic components need to be standardized before being input into the first processing model (z=(x-μ) / σ), where μ and σ are the mean and standard deviation of each first target intrinsic component.

[0056] In some embodiments of this application, the first processing model includes a gated recurrent layer, a first random deactivation layer, and a first fully connected layer connected in sequence. The gated recurrent layer is configured to perform feature filtering on the first target intrinsic components, thereby filtering out invalid features of the first target intrinsic components and extracting valid temporal fluctuation features. The first random deactivation layer is configured to perform random deactivation processing on the feature representation output by the gated recurrent layer, thereby avoiding model overfitting. The first fully connected layer is configured to perform weighted calculation on the feature representation output by the first random deactivation layer to obtain the second target intrinsic components.

[0057] S303, input the filtered first target trend component into the second processing model for processing to obtain the second target trend component.

[0058] In some embodiments of this application, the second processing model includes a plurality of fully connected hidden layers, a second random deactivation layer, and a second fully connected layer connected in sequence. The plurality of fully connected hidden layers are configured to perform linear weighting and nonlinear transformation on the first target trend component, thereby capturing long-term variation patterns. The second random deactivation layer is configured to perform random deactivation processing on the feature representations output by the plurality of fully connected hidden layers, thereby avoiding model overfitting. The second fully connected layer is configured to perform weighted calculation on the feature representations output by the second random deactivation layer to obtain the second target trend component.

[0059] S304, the second target intrinsic component and the second target trend component are determined as the second frequency component.

[0060] In other words, the second frequency component = [second target intrinsic component 1, second target intrinsic component 2, ..., second target intrinsic component m, second target trend component], where m is the order of the intrinsic components retained after screening, such as 6.

[0061] The above embodiments selectively screen effective features that are key to temperature prediction in the first frequency component, and then use a first processing model adapted to the temporal fluctuation characteristics of the intrinsic component and a second processing model adapted to the smoothing law of the trend component to make predictions separately. Finally, the second frequency component is obtained by integration. This not only eliminates redundant noise and improves prediction efficiency, but also ensures prediction accuracy through multi-dimensional accurate modeling.

[0062] S104 integrates and reconstructs the second frequency component to obtain the predicted temperature of the target component.

[0063] Optionally, the formula for integrating and reconstructing the second frequency component can be expressed as:

[0064] In the above formula, Let be the predicted temperature at time t+1, and m be the order of the eigencomponents retained after filtering. For time t+1, the j-th order second objective eigencomponent is... This represents the second objective trend component at time t+1.

[0065] S105, determine the target adjustment parameters that are adapted to the predicted temperature, and adjust the target circuit based on the target adjustment parameters to regulate the temperature of the target components.

[0066] Among them, the target adjustment parameters refer to the parameters that can be used to adjust the working state of the target circuit, thereby achieving temperature regulation of the target components, such as the switching frequency of the target circuit, the amplitude of the supply voltage, the duty cycle of the pulse width modulation signal, and the fan speed.

[0067] In some embodiments of this application, such as Figure 4 As shown, the determination of the target adjustment parameters for temperature adaptation in S105 includes steps S401-S403, as detailed below: S401, calculate the prediction confidence level based on the predicted temperature and the measured temperature at the corresponding time.

[0068] Among them, prediction confidence is an indicator that quantifies the reliability of temperature prediction results. The higher the prediction confidence, the higher the reliability of the predicted temperature.

[0069] Alternatively, the prediction confidence level can be calculated using the following formula:

[0070] In the above formula, To predict temperature, This represents the measured temperature at the corresponding moment. and This represents the upper and lower limits of the safe temperature for the target component. This formula maps the temperature deviation to the interval [0, 1], and the closer the value is to 1, the higher the confidence level.

[0071] In some embodiments of this application, S401, which calculates the prediction confidence level based on the predicted temperature and the measured temperature at the corresponding time, specifically includes the following steps: subtracting the measured temperature at the corresponding time from the predicted temperature to obtain a temperature prediction difference; dividing the temperature prediction difference by the standard deviation of the difference to obtain a prediction deviation ratio; and subtracting the prediction deviation ratio from the first parameter to obtain the prediction confidence level.

[0072] Understandably, calculating the prediction bias ratio can adapt to the recent prediction accuracy of the system (for example, if the recent prediction error fluctuates greatly, the standard deviation will be large, and the corresponding bias ratio will be corrected). Subtracting the prediction bias ratio from the first parameter ensures that the confidence level is a positively vectorized indicator and is inversely proportional to the bias.

[0073] Optionally, the formula associated with the calculation process in the above embodiments may be:

[0074] In the above formula, 'a' is the first parameter, for example, a=1. It is the standard deviation of the temperature prediction differences of the first N (e.g., N=50).

[0075] S402, when the prediction confidence is greater than the confidence threshold, determine the target substitution value corresponding to each candidate regulation parameter based on the predicted temperature, and determine the target regulation parameter from the candidate regulation parameters based on the target substitution value.

[0076] Candidate adjustment parameters refer to a set of multiple alternative parameters that can be used to adjust the operating state of the target circuit to change the temperature of the target components. Target cost value is a quantitative indicator used to evaluate the merits of each candidate adjustment parameter; the smaller the target cost value, the better the corresponding candidate adjustment parameter.

[0077] Understandably, if the prediction confidence level is greater than the confidence level threshold (e.g., 0.7), it indicates that the predicted temperature is reliable. The target adjustment parameter can be determined based on the prediction result, and then optimized adjustment can be carried out.

[0078] In some embodiments of this application, step S402, which determines the target cost value corresponding to each candidate adjustment parameter based on the predicted temperature and determines the target adjustment parameter from the candidate adjustment parameters based on the target cost value, specifically includes the following steps: calculating the difference between the predicted temperature and the safe temperature corresponding to each candidate adjustment parameter to obtain the safety cost value; determining the power cost value and comfort cost value corresponding to each candidate adjustment parameter; performing a weighted calculation on the safety cost value, power cost value, and comfort cost value to obtain the target cost value of each candidate adjustment parameter; and determining the candidate adjustment parameter corresponding to the minimum target cost value as the target adjustment parameter.

[0079] The safe temperature refers to the upper limit of the temperature range within which the target component can operate stably for a long period without performance degradation or damage. Different candidate adjustment parameters correspond to different safe temperatures. For example, the safe temperature for a switching frequency of 70kHz is 90℃, while the safe temperature for 80kHz is 85℃.

[0080] Optionally, the formula for calculating the target cost value of each candidate adjustment parameter is as follows:

[0081] In the above formula, Candidate adjustment parameters; Sacrificing value for safety; To predict temperature; For safe temperature; The power cost corresponding to the candidate adjustment parameters; This refers to the compressor power. This refers to the fan power. This refers to the circuit power. The comfort cost corresponding to the candidate adjustment parameters; Indoor temperature; To set the temperature. , , These are the weights corresponding to the cost values.

[0082] because The value of is discrete, so an enumeration method can be used to solve it. The controller calculates the target cost value corresponding to each candidate regulation parameter in the feasible candidate regulation parameters (e.g., the switching frequency set [65, 70, 75, ..., 110] kHz), and then selects the candidate regulation parameter that minimizes the target cost value as the target regulation parameter.

[0083] The calculation of the target cost value allows for a comprehensive quantification of the performance of candidate adjustment parameters in various dimensions, including component temperature rise suppression, overall air conditioner energy efficiency control, and user comfort assurance. This effectively resolves the conflict between temperature rise suppression, overall energy efficiency, and user experience, enabling the system to intelligently balance and precisely select among conflicting objectives. The target adjustment parameters selected based on the minimum target cost value can take into account component safety, overall operating efficiency, and user experience, truly achieving system-level optimal control of air conditioner component temperature regulation. This makes the temperature regulation strategy more scientific, comprehensive, and aligned with actual operating needs.

[0084] S403 determines the target adjustment parameter corresponding to the measured temperature through PID control when the prediction confidence level is less than or equal to the confidence level threshold.

[0085] Understandably, if the prediction confidence level is less than or equal to the confidence level threshold, it indicates that the reliability of the predicted temperature is insufficient, and a stable PID closed-loop regulation is required to ensure the control effect.

[0086] Optionally, the safe operating temperature or a manually set temperature of the target component is used as the setpoint Tset, the current measured temperature Tact of the component is collected, and the deviation between the two is calculated as e=Tset. Tact. The deviation value 'e' is input into the PID controller, which calculates the proportional, integral, and derivative terms respectively. The three results are then weighted and summed to obtain the PID control output value. Finally, the control output value is converted into the target control parameters that the target circuit can execute, according to a preset output value-control parameter mapping relationship.

[0087] Furthermore, after determining the target adjustment parameters that match the predicted temperature, the determined target adjustment parameters (such as a switching frequency of 70kHz, a duty cycle of 40%, etc.) are sent to the control unit of the target circuit. The control unit then adjusts according to the parameter instructions. For example, adjusting the switching frequency can be achieved by changing the on / off frequency of power devices (such as MOSFETs). Lowering the frequency reduces switching losses and reduces component heat generation; increasing the frequency may improve heat dissipation efficiency. Alternatively, adjusting the supply voltage / duty cycle can reduce component energy consumption and heat generation, while increasing it increases energy consumption and heat generation. Another approach is to adjust the fan speed related to the heat dissipation of the target circuit. Increasing the fan speed accelerates airflow over the component surface, enhancing heat dissipation; conversely, decreasing the fan speed weakens heat dissipation.

[0088] The above embodiments quantify the reliability of temperature prediction by calculating the prediction confidence level. When the confidence level is high, the optimal adjustment parameters that take into account safety, energy consumption and comfort are selected based on multi-objective cost value. When the confidence level is insufficient, the mature PID closed-loop regulation is switched. This achieves both precise optimization and regulation and ensures the stability of regulation under extreme conditions.

[0089] Furthermore, the measured temperature of the target component is collected in real time. If the temperature does not reach the safe range, the process returns to execute S101 and subsequent steps until the component temperature stabilizes within the safe range. The control unit maintains the current target adjustment parameters to ensure temperature stability.

[0090] The above embodiment achieves proactive temperature control through an innovative decomposition-prediction-reconstruction framework: First, temperature-related data concerning the future temperature of the target components in the air conditioner is acquired. A first frequency component is obtained through modal decomposition, and a second frequency component is predicted from this component. The second frequency component is then integrated and reconstructed to obtain the predicted temperature of the target components. Finally, appropriate target adjustment parameters are matched to adjust the target circuit, thereby regulating the temperature of the target components. This solution enables the prediction and advance adjustment of component temperatures, allowing for targeted adjustments before a significant increase in component temperature due to sudden surges in power grids, effectively mitigating the risk of component damage from high temperatures. Simultaneously, this framework significantly improves temperature prediction accuracy, ensuring that temperature rise intervention occurs before the actual temperature rise, fundamentally solving the lag problem of traditional feedback control and significantly improving the stability and lifespan of the target components and the entire air conditioner.

[0091] In some embodiments of this application, the second frequency component is predicted based on the first processing model and the second processing model. After determining the target adjustment parameter adapted to the predicted temperature in step S105, and adjusting the target circuit based on the target adjustment parameter to adjust the temperature of the target component, the following steps are further performed: calculating the loss value based on the predicted temperature and the measured temperature at the corresponding time. When the loss value is greater than the loss value threshold and the air conditioner is in a stable operating state, adjusting the model parameters of the first processing model and the model parameters of the second processing model based on the loss value.

[0092] Understandably, the first and second processing models are the core of temperature prediction. As the operating conditions of the air conditioner change (such as seasonal changes or component aging), the initial model parameters may no longer be suitable. Therefore, continuous optimization through feedback adjustment is necessary to ensure the accuracy of temperature prediction and thus improve the reliability of temperature regulation.

[0093] Optionally, the loss value can be the difference between the predicted temperature and the measured temperature at the corresponding time, or it can be the mean absolute error of the difference, or it can be the mean square error of the difference; there are no limitations on this.

[0094] Optionally, model parameter adjustment should only be initiated if both of the following conditions are met simultaneously to avoid blind optimization under unstable operating conditions: Condition 1: Loss value > loss value threshold (e.g., MAE > 2℃). This indicates that the current model's prediction accuracy is insufficient and parameter adjustment is necessary. Condition 2: The air conditioner is in a stable operating state. For example, it should be in "standby mode" or "continuously running at low load for more than 5 minutes" to ensure a stable data environment during adjustment.

[0095] Optionally, during tuning, the model parameters of the first processing model and the second processing model are adjusted based on the loss value and a set learning rate. The training rounds are limited to 5 epochs, and a subset of new data is used for validation to prevent model overfitting.

[0096] To facilitate better implementation of the component temperature regulation method, a component temperature regulation device based on the above-described component temperature regulation method is also provided. The meanings of the terms used are the same as in the above-described component temperature regulation method, and specific implementation details can be found in the description of the method embodiments.

[0097] Please see Figure 5 , Figure 5 This is a schematic diagram of the component temperature regulation device provided in the embodiment, which may specifically include: The data acquisition module 501 is used to acquire temperature-related data; wherein, the temperature-related data represents the data that is related to the current future temperature of the target component; The decomposition processing module 502 is used to perform modal decomposition processing on the temperature-related data to obtain the first frequency component of the temperature-related data. The component prediction module 503 is used to perform component prediction based on the first frequency component to obtain the second frequency component of the target component. The integration and reconstruction module 504 is used to integrate and reconstruct the second frequency component to obtain the predicted temperature of the target component. The temperature regulation module 5005 is used to determine the target regulation parameters that are adapted to the predicted temperature, and to adjust the target circuit based on the target regulation parameters in order to regulate the temperature of the target components.

[0098] In the above embodiment, the data acquisition module 501 is used to acquire temperature-related data related to the future temperature of the target component of the air conditioner; the decomposition processing module 502 is used to decompose the temperature-related data to obtain a first frequency component; the component prediction module 503 is used to predict a second frequency component based on the first frequency component; the integration and reconstruction module 504 is used to integrate and reconstruct the second frequency component to obtain the predicted temperature of the target component; and the temperature adjustment module 5005 is used to finally match and adapt the target adjustment parameters to adjust the target circuit. This scheme realizes the forward prediction and early adjustment of the temperature of the target component, and can take targeted adjustment measures before the component temperature rises significantly when a sudden large impact arrives, effectively avoiding the risk of component damage due to excessive temperature under a sudden large impact, and improving the stability and service life of the target component and even the whole unit of the air conditioner.

[0099] In some embodiments, the decomposition processing module 502 performs mode decomposition processing on the temperature-related data to obtain a first frequency component of the temperature-related data, including: Based on the preset number of superpositions and white noise amplitude coefficient, temperature correlation data and white noise data are superimposed to obtain multiple sets of superimposed data. Each set of superimposed data is sieved and decomposed to obtain multiple intrinsic components and trend components of each set of superimposed data. The average value of multiple eigencomponents of multiple sets of superimposed data is calculated within the same order eigencomponent to obtain the first target eigencomponent of each order. The first target trend component is obtained by averaging the trend components of multiple sets of superimposed data. The first target intrinsic component and the first target trend component of each order are determined as the first frequency component of the temperature correlation data.

[0100] In some embodiments, the first frequency component includes first target intrinsic components and first target trend components of each order. The component prediction module 503 performs component prediction based on the first frequency component, and the resulting second frequency component includes: Based on the preset filtering rules, the first target intrinsic components and the first target trend components of each order are filtered to obtain the filtered first target intrinsic components and the first target trend components. The filtered first target intrinsic components are input into the corresponding first processing model for processing to obtain the second target intrinsic components. The filtered first target trend component is input into the second processing model for processing to obtain the second target trend component. The intrinsic component and trend component of the second objective are defined as the second frequency component.

[0101] In some embodiments, the first processing model includes a gated recurrent layer, a first random deactivation layer, and a first fully connected layer connected in sequence. The gated recurrent layer is configured to perform feature filtering on the intrinsic components of the first target; The first random deactivation layer is configured to randomly deactivate the feature representation output by the gated recurrent layer. The first fully connected layer is configured to perform weighted computation on the feature representation output by the first randomly deactivated layer to obtain the second target intrinsic component; The second processing model includes multiple fully connected hidden layers, a second random deactivation layer, and a second fully connected layer connected in sequence. Multiple fully connected hidden layers are configured to perform linear weighting and nonlinear transformation on the first target trend component; The second random deactivation layer is configured to randomly deactivate the feature representations output by multiple fully connected hidden layers; The second fully connected layer is configured to perform weighted computation on the feature representation output of the second random deactivated layer to obtain the second target trend component.

[0102] In some embodiments, the temperature regulation module 5005 determines target regulation parameters adapted to the predicted temperature, including: The prediction confidence level is calculated based on the predicted temperature and the measured temperature at the corresponding time. When the prediction confidence is greater than the confidence threshold, the target cost value corresponding to each candidate adjustment parameter is determined based on the predicted temperature, and the target adjustment parameter is determined from the candidate adjustment parameters based on the target cost value. When the prediction confidence level is less than or equal to the confidence level threshold, the target adjustment parameter corresponding to the measured temperature is determined by PID control.

[0103] In some embodiments, the temperature regulation module 5005 determines the target cost value corresponding to each candidate regulation parameter based on the predicted temperature, and determines the target regulation parameter from the candidate regulation parameters based on the target cost value, including: Calculate the difference between the predicted temperature and the safe temperature corresponding to each candidate adjustment parameter to obtain the safety cost. Determine the power cost and comfort cost corresponding to each candidate adjustment parameter; The safety cost, power cost, and comfort cost are weighted and calculated to obtain the target cost for each candidate adjustment parameter; The candidate adjustment parameter that minimizes the target cost is determined as the target adjustment parameter.

[0104] In some embodiments, the second frequency component is predicted based on the first processing model and the second processing model. The temperature adjustment module 5005 determines a target adjustment parameter that matches the predicted temperature, and adjusts the target circuit based on the target adjustment parameter to adjust the temperature of the target component. It is then further used for: The loss value is calculated based on the predicted temperature and the actual measured temperature at the corresponding time. When the loss value is greater than the loss value threshold and the air conditioner is in a stable operating state, the model parameters of the first processing model and the model parameters of the second processing model are adjusted based on the loss value.

[0105] In addition, an air conditioner is also provided, such as Figure 6 As shown, it illustrates the structural diagram of the air conditioner involved, specifically: The air conditioner may include components such as a processor 601 with one or more processing cores, a memory 602 with one or more computer-readable storage media, a power supply 603, and an input unit 604. Those skilled in the art will understand that... Figure 6 The air conditioner structure shown does not constitute a limitation on the air conditioner and may include more or fewer components than shown, or combine certain components, or have different component arrangements. Wherein: The processor 601 is the control center of the air conditioner. It connects to various parts of the air conditioner via various interfaces and lines. By running or executing software programs and / or modules stored in the memory 602, and by calling data stored in the memory 602, it performs various functions and processes data, thereby providing overall monitoring of the air conditioner. Optionally, the processor 601 may include one or more processing cores; preferably, the processor 601 may integrate an application processor and a modem processor, wherein the application processor mainly handles the operating system, user interface, and applications, and the modem processor mainly handles wireless communication. It is understood that the modem processor may not be integrated into the processor 601.

[0106] The memory 602 can be used to store software programs and modules. The processor 601 executes various functional applications and data processing by running the software programs and modules stored in the memory 602. The memory 602 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function, etc.; the data storage area may store data created based on the use of the air conditioner, etc. In addition, the memory 602 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory 602 may also include a memory controller to provide the processor 601 with access to the memory 602.

[0107] The air conditioner also includes a power supply 603 that supplies power to the various components. Preferably, the power supply 603 can be logically connected to the processor 601 through a power management system, thereby enabling functions such as charging, discharging, and power consumption management through the power management system. The power supply 603 may also include one or more DC or AC power supplies, recharging systems, power equipment debugging circuits, power converters or inverters, power status indicators, and other arbitrary components.

[0108] The air conditioner may also include an input unit 604, which can be used to receive input digital or character information, and generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function control.

[0109] Although not shown, the air conditioner may also include a display unit, etc., which will not be described in detail here. Specifically, in this embodiment, the processor 601 in the air conditioner loads the executable files corresponding to the processes of one or more application programs into the memory 602 according to the following instructions, and the processor 601 runs the application programs stored in the memory 602, thereby implementing the steps in any of the component temperature regulation methods provided in the embodiment: acquiring temperature-related data; wherein, the temperature-related data represents data currently associated with the future temperature of the target component; performing mode decomposition processing on the temperature-related data to obtain a first frequency component of the temperature-related data; performing component prediction based on the first frequency component to obtain a second frequency component of the target component; integrating and reconstructing the second frequency component to obtain the predicted temperature of the target component; determining a target adjustment parameter adapted to the predicted temperature, and adjusting the target circuit based on the target adjustment parameter to regulate the temperature of the target component.

[0110] The above embodiment achieves proactive temperature control through an innovative decomposition-prediction-reconstruction framework: First, temperature-related data concerning the future temperature of the target components in the air conditioner is acquired. A first frequency component is obtained through modal decomposition, and a second frequency component is predicted from this component. The second frequency component is then integrated and reconstructed to obtain the predicted temperature of the target components. Finally, appropriate target adjustment parameters are matched to adjust the target circuit, thereby regulating the temperature of the target components. This solution enables the prediction and advance adjustment of component temperatures, allowing for targeted adjustments before a significant increase in component temperature due to sudden surges in power grids, effectively mitigating the risk of component damage from high temperatures. Simultaneously, this framework significantly improves temperature prediction accuracy, ensuring that temperature rise intervention occurs before the actual temperature rise, fundamentally solving the lag problem of traditional feedback control and significantly improving the stability and lifespan of the target components and the entire air conditioner.

[0111] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.

[0112] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be performed by instructions, or by instructions controlling related hardware. These instructions can be stored in a computer-readable storage medium and loaded and executed by a processor.

[0113] To this end, a computer-readable storage medium is provided, on which a computer program is stored, which can be loaded by a processor to execute the steps in any of the provided component temperature regulation methods.

[0114] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.

[0115] The computer-readable storage medium may include: read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0116] Since the instructions stored in the computer-readable storage medium can execute the steps in any of the provided component temperature regulation methods, the beneficial effects that any of the provided component temperature regulation methods can achieve can be realized, as detailed in the preceding embodiments, and will not be repeated here.

[0117] The above provides a detailed description of a component temperature regulation method, device, air conditioner, and computer-readable storage medium. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for regulating the temperature of a component, characterized in that, The method includes: Acquire temperature-related data; wherein, the temperature-related data represents data relating the current temperature to the future temperature of the target component; The temperature-correlated data is subjected to mode decomposition to obtain the first frequency component of the temperature-correlated data; Based on the first frequency component, component prediction is performed to obtain the second frequency component of the target component. The second frequency component is integrated and reconstructed to obtain the predicted temperature of the target component; Determine the target adjustment parameters that are compatible with the predicted temperature, and adjust the target circuit based on the target adjustment parameters to regulate the temperature of the target components.

2. The component temperature regulation method according to claim 1, characterized in that, The modal decomposition process of the temperature-correlated data to obtain the first frequency component of the temperature-correlated data includes: Based on the preset number of superpositions and the white noise amplitude coefficient, the temperature correlation data and white noise data are superimposed to obtain multiple sets of superimposed data. Each set of superimposed data is sieved and decomposed to obtain multiple intrinsic components and trend components of each set of superimposed data. The average value of multiple eigencomponents of the multiple sets of superimposed data is calculated within the same order eigencomponent to obtain the first target eigencomponent of each order. The average value of the trend components of the multiple sets of superimposed data is calculated to obtain the first target trend component; The first target intrinsic component and the first target trend component of each order are determined as the first frequency component of the temperature correlation data.

3. The component temperature regulation method according to claim 1, characterized in that, The first frequency component includes first target intrinsic components and first target trend components of each order. The second frequency component obtained by component prediction based on the first frequency component includes: Based on the preset filtering rules, the first target intrinsic components and the first target trend components of each order are filtered to obtain the filtered first target intrinsic components and the first target trend components. The filtered first target intrinsic components are input into the corresponding first processing model for processing to obtain the second target intrinsic components. The filtered first target trend component is input into the second processing model for processing to obtain the second target trend component. The second target intrinsic component and the second target trend component are determined as the second frequency component.

4. The component temperature regulation method according to claim 3, characterized in that, The first processing model includes a gated recurrent layer, a first random deactivation layer, and a first fully connected layer connected in sequence; The gated recurrent layer is configured to perform feature filtering on the intrinsic components of the first target; The first random deactivation layer is configured to randomly deactivate the feature representation output by the gated recurrent layer; The first fully connected layer is configured to perform a weighted calculation on the feature representation output by the first randomly deactivated layer to obtain the second target intrinsic component; The second processing model includes multiple fully connected hidden layers, a second random deactivation layer, and a second fully connected layer connected in sequence; The plurality of fully connected hidden layers are configured to perform linear weighting calculations and nonlinear transformations on the first target trend component; The second random deactivation layer is configured to randomly deactivate the feature representations output by the plurality of fully connected hidden layers; The second fully connected layer is configured to perform weighted calculations on the feature representation output by the second random deactivation layer to obtain the second target trend component.

5. The component temperature regulation method according to claim 1, characterized in that, The determination of the target adjustment parameter adapted to the predicted temperature includes: The prediction confidence level is calculated based on the predicted temperature and the measured temperature at the corresponding time. When the prediction confidence level is greater than the confidence level threshold, the target cost value corresponding to each candidate adjustment parameter is determined based on the predicted temperature, and the target adjustment parameter is determined from the candidate adjustment parameters based on the target cost value. When the predicted confidence level is less than or equal to the confidence level threshold, the target adjustment parameter corresponding to the measured temperature is determined by PID control.

6. The component temperature regulation method according to claim 5, characterized in that, The step of determining the target substitution value corresponding to each candidate adjustment parameter based on the predicted temperature, and determining the target adjustment parameter from the candidate adjustment parameters based on the target substitution value, includes: Calculate the difference between the predicted temperature and the safe temperature corresponding to each candidate adjustment parameter to obtain the safety cost. Determine the power cost and comfort cost corresponding to each candidate adjustment parameter; The safety cost, power cost, and comfort cost are weighted and calculated to obtain the target cost for each candidate adjustment parameter; The candidate adjustment parameter corresponding to the minimum target cost is determined as the target adjustment parameter.

7. The component temperature regulation method according to claim 1, characterized in that, The second frequency component is predicted based on the first processing model and the second processing model. After determining the target adjustment parameter adapted to the predicted temperature and adjusting the target circuit based on the target adjustment parameter to adjust the temperature of the target component, the method further includes: The loss value is calculated based on the predicted temperature and the actual measured temperature at the corresponding time. When the loss value is greater than the loss value threshold and the air conditioner is in a stable operating state, the model parameters of the first processing model and the model parameters of the second processing model are adjusted based on the loss value.

8. A component temperature regulating device, characterized in that, The component temperature control device includes: The data acquisition module is used to acquire temperature-related data; wherein, the temperature-related data represents data that is currently associated with the future temperature of the target component; The decomposition processing module is used to perform modal decomposition processing on the temperature-related data to obtain the first frequency component of the temperature-related data. A component prediction module is used to perform component prediction based on the first frequency component to obtain the second frequency component of the target component. An integration and reconstruction module is used to integrate and reconstruct the second frequency component to obtain the predicted temperature of the target component; A temperature regulation module is used to determine a target regulation parameter that is compatible with the predicted temperature, and to adjust the target circuit based on the target regulation parameter in order to regulate the temperature of the target component.

9. A computer-readable storage medium, characterized in that, The device stores a computer program that, when executed by a processor, causes the processor to perform the steps of the method as described in any one of claims 1 to 7.

10. An air conditioner, characterized in that, It includes a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the steps of the method as described in any one of claims 1 to 7.

Images