An electrical signal optimization method for an air conditioning electronic control system

Abstract

The present invention proposes a method for optimizing the electrical signals of an air conditioner electronic control system, which relates to the technical field of air conditioner electrical signal optimization. The method comprises: obtaining the original air conditioner electrical control signal, decomposing the original electrical control signal into different frequency bands through wavelet packet decomposition to obtain multiple sub-band signals; obtaining the sliding energy entropy of each sub-band signal; comparing the sliding energy entropy of each sub-band signal with an entropy threshold to determine the corresponding denoising algorithm, and outputting the denoised sub-band signal; performing weighted reconstruction on the denoised sub-band signal, calculating a real-time residual evaluation value for the sub-band signal, comparing the real-time residual evaluation value with the residual threshold, and determining, based on the comparison result, whether to output the reconstructed sub-band signal as the output signal or to obtain correction parameters for analysis and optimization. The present invention addresses the inaccuracy and interference problems of existing methods in denoising air conditioner electrical signals through dynamic decomposition and feedback loops, mapping energy entropy to multimodal algorithms, and adaptive SNR adjustment through weighted fusion.

Description

Technical Field

[0001] The present invention belongs to the technical field of air-conditioning electric signal optimization, and in particular relates to an electric signal optimization method for an air-conditioning electric control system. Background Art

[0002] The intelligent electronic air conditioning control system is an air conditioning system based on electronics and intelligent control technologies. It achieves comfortable, energy-efficient, and environmentally friendly air conditioning control by intelligently adjusting parameters such as temperature, humidity, and wind speed. Whether it's the indoor or outdoor unit, the normal operation of the electronic control system and the coordinated control between the various air conditioning modules are achieved by transmitting electrical signals between the modules, such as those from temperature sensors, humidity sensors, wind speed sensors, power supplies, control signals, fault signals, and remote control signals.

[0003] However, due to electromagnetic interference between electronic components, sensor pulse fluctuations, and voltage instability, electrical signals are highly susceptible to noise, leading to errors in the intelligent control and regulation of air conditioners. Most existing technologies use traditional wavelet decomposition techniques to separate the signal into different components and then process only the parts identified as noise. However, the reality is that noise can permeate each component. In particular, high-frequency components may contain useful signals, while low-frequency components may contain residual noise. This makes it impossible to completely and accurately denoise the electrical signal, resulting in inaccurate denoised electrical signals and, in turn, inaccurate regulation of the air conditioner's intelligent electronic control system. Summary of the Invention

[0004] In response to the problems raised in the background technology, the present invention proposes an electrical signal optimization method for an air-conditioning electronic control system.

[0005] To achieve this object, the present invention adopts the following technical solutions:

[0006] A method for optimizing electrical signals of an air-conditioning electronic control system, comprising:

[0007] Step A: Obtain the original electric control signal of the air conditioner, and decompose the original electric control signal into different frequency bands through wavelet packet decomposition to obtain multiple sub-band signals;

[0008] Step B: Obtain the sliding energy entropy of each sub-band signal;

[0009] Step C: Compare the sliding energy entropy of each sub-band signal with the entropy value threshold to determine the corresponding denoising algorithm, and output each sub-band signal after denoising;

[0010] Step D: performing weighted reconstruction on the denoised sub-band signal to obtain a reconstructed sub-band signal;

[0011] Step E: Calculate a real-time residual evaluation value for the reconstructed subband signal, compare the real-time residual evaluation value with the residual threshold, and determine, based on the comparison result, whether to output the subband signal reconstructed in step D as the final output signal or obtain correction parameters and optimize steps A to C based on the correction parameters. The correction parameters include a new initial decomposition layer number for wavelet packet decomposition after optimization and a new entropy value threshold after optimization.

[0012] Preferably, the step A specifically includes:

[0013] Determine the initial wavelet packet decomposition layer number according to formula 1:

[0014] —Formula 1;

[0015] N represents the number of initial decomposition layers;

[0016] floor() represents the rounding down function;

[0017] fs represents the sampling frequency of the original electronic control signal;

[0018] Indicates the preset frequency resolution. The minimum bandwidth of the original electronic control signal obtained after N-layer decomposition must be greater than the preset frequency resolution.

[0019] Preferably, the step A further comprises:

[0020] A dynamic decomposition tree is constructed according to the initial decomposition layer number N. The bandwidth of the previous layer is divided into two at each decomposition layer, so that a corresponding number of frequency bands are obtained with the minimum bandwidth at the last decomposition layer, and the original electronic control signal is decomposed into sub-band signals of the corresponding number of frequency bands, including:

[0021] The first layer of decomposition: the sampling frequency of the original electric control signal is used as the initial bandwidth and divided by two, and the first layer The bandwidth of each sub-band is HZ;

[0022] The second layer decomposition: the first layer The bandwidth of each sub-band is divided into two equal parts, and the second layer is obtained. The bandwidth of each sub-band is HZ;

[0023] And so on, decompose in sequence until the Nth layer of decomposition;

[0024] Nth layer decomposition: decomposition of the N-1th layer The bandwidth of each sub-band is divided into two equal parts, and the Nth layer is obtained. The bandwidth of each sub-band is HZ, so that the original electric control signal is converted into the minimum frequency bandwidth HZ is decomposed into sub-band signals;

[0025] Get each subband signal at time The wavelet packet coefficient value is recorded as the sub-band signal amplitude of each sub-band signal .

[0026] Preferably, the step B specifically includes:

[0027] Calculate the energy integral of each subband according to Formula 2:

[0028] —Formula 2;

[0029] represents the energy of the Kth subband at time t;

[0030] K represents the subband number, which is used to identify the frequency band position being analyzed;

[0031] t represents the current time, which is used to define the end point of the sliding window, t>T;

[0032] T represents the sliding window width, which is used to determine the time range of energy statistics and is determined according to the air conditioning control cycle;

[0033] Indicates the integral variable from the time point in the interval [tT, t], which is used to traverse all time points in the window;

[0034] represents the sub-band signal amplitude, which is obtained by wavelet decomposition of the original electronic control signal in step A.

[0035] Preferably, the step B specifically includes:

[0036] The energy probability distribution of each subband is calculated according to Formula 3:

[0037] —Formula 3;

[0038] represents the energy probability distribution of the Kth sub-band, which is used to reflect the weight of the frequency band in the total energy;

[0039] M represents the total number of subbands, which is determined by the initial decomposition level N of step A. ;

[0040] j represents the subband index, which is used to traverse all subbands;

[0041] represents the energy of the jth subband at time t;

[0042] represents the energy of the Kth subband at time t.

[0043] Preferably, the step B specifically includes:

[0044] The sliding energy entropy of each subband is calculated according to formula 4:

[0045] —Formula 4;

[0046] represents the sliding energy entropy of the Kth subband, which is used to quantify the signal complexity. ;

[0047] when When it tends to 1, it means that the energy of each sub-band is evenly distributed, indicating that the original electronic control signal is a steady-state signal;

[0048] when When it tends to 0, it means that the energy is concentrated in a few sub-bands, indicating that there is pulse interference in the original electronic control signal.

[0049] Preferably, the step C comprises:

[0050] when When , the morphological filtering algorithm is used to denoise the current sub-band and output the denoised sub-band signal;

[0051] when When , the wavelet threshold algorithm is used to denoise the current sub-band and output the denoised sub-band signal;

[0052] when When , the LMS adaptive filtering algorithm is used to denoise the current sub-band and output the denoised sub-band signal;

[0053] and represents the entropy threshold, which is calibrated by the noise sample. , .

[0054] Preferably, the step D specifically includes:

[0055] This involves weighted reconstruction of each subband signal using Formula 5 and Formula 6:

[0056] —Formula 5;

[0057] —Formula 6;

[0058] represents the sub-band signal after denoising and reconstruction;

[0059] represents the Kth subband signal after denoising, obtained by denoising in step C using the corresponding denoising algorithm;

[0060] Represents the dynamic weight coefficient, which is used to reflect the credibility of the sub-band signal;

[0061] exp() represents the natural exponential function;

[0062] Represents the weight adjustment factor, which is used to reflect the balance between the number of sub-bands and sensitivity;

[0063] Represents the real-time sub-band signal-to-noise ratio, which is used to quantify the sub-band signal quality;

[0064] Indicates the signal-to-noise ratio baseline threshold, which is set according to the noise floor of the air conditioning system.

[0065] Preferably, the step E specifically includes:

[0066] Calculate the real-time residual evaluation value according to Formula 7:

[0067] —Formula 7;

[0068] Represents the real-time residual evaluation value;

[0069] Represents the amplitude error weight coefficient, which is used to control the contribution weight of the static deviation;

[0070] Represents the differential error weight coefficient, which is used to control the dynamically changing contribution weight;

[0071] Represents the original electric control signal;

[0072] represents the target signal after weighted reconstruction in step D;

[0073] It represents the first-order derivative of the original electronic control signal at time t, and is used to represent the instantaneous rate of change of the signal.

[0074] Preferably, the step D further comprises:

[0075] Compare the real-time residual evaluation value with the residual threshold:

[0076] when When , the subband signal reconstructed by weight in step D is output as the final output signal;

[0077] when When including:

[0078] According to formula 8, the initial decomposition layer number N is optimized. —Formula 8;

[0079] Indicates the number of decomposition layers after optimization;

[0080] sgn() represents the sign function. hour, ,when hour, ,when hour, ;

[0081] , represents the derivative of the real-time residual evaluation value R(t) at time t, which is used to reflect the changing trend of the real-time residual evaluation value;

[0082] According to formula nine, the entropy value threshold Optimize, —Formula 9;

[0083] Indicates the entropy threshold after optimization, the entropy threshold for the first optimization The entropy threshold value is calibrated by the noise sample at the initial time Based on the artificial calibration, the entropy threshold value after the first optimization is Calculated by formula nine;

[0084] represents the learning rate;

[0085] Indicates the difference in real-time residual evaluation values ​​before and after optimization;

[0086] Indicates the difference between the entropy thresholds before and after optimization;

[0087] The optimized decomposition layer number As the new decomposition level N, it is passed back to step A, and the dynamic decomposition tree is constructed using the new decomposition level N;

[0088] The optimized entropy threshold As the new entropy threshold Return to step C and use the new entropy threshold To expand the filtering range of the morphological filtering algorithm.

[0089] The beneficial effects of the present invention compared to the prior art are as follows:

[0090] 1. The present invention optimizes the number of decomposition layers in step A by dynamically adjusting the residual analysis in step E to solve the band aliasing caused by the fixed number of decomposition layers in the traditional wavelet packet decomposition method. It also optimizes the entropy threshold in step C and calibrates the decision threshold in real time to avoid overfitting or underfitting problems caused by fixed parameters, thereby achieving a closed loop of dynamic decomposition and feedback.

[0091] 2. The present invention quantifies the noise characteristics by the entropy value of step B, thereby achieving accurate matching between the noise characteristics and the denoising algorithm;

[0092] 3. The present invention ensures that the high signal-to-noise ratio subband dominates the reconstruction result through the exponential weight function in step D, and suppresses the noise influence of the low-quality subband. BRIEF DESCRIPTION OF THE DRAWINGS

[0093] Figure 1 The present invention is a flow chart of the method for optimizing the electrical signals of the air-conditioning electronic control system. DETAILED DESCRIPTION

[0094] The technical solution of the present invention will be further described below with reference to the accompanying drawings and through specific implementation methods.

[0095] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0096] The terms "first," "second," and so on, in the description and claims of the present invention and the accompanying drawings are used to distinguish between different items, not to describe a specific order. Furthermore, the terms "including," "having," and any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, apparatus, product, or end comprising a series of steps or elements is not limited to the listed steps or elements but may optionally include steps or elements not listed therein, or may optionally include other steps or elements inherent to such process, method, product, or end.

[0097] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present invention. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute a separate or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0098] This application proposes a method for optimizing electrical signals of an air conditioning electronic control system, comprising:

[0099] Step A: Obtain the original electric control signal of the air conditioner, and decompose the original electric control signal into different frequency bands through wavelet packet decomposition to obtain multiple sub-band signals;

[0100] Preferably, the step A specifically includes:

[0101] Determine the initial wavelet packet decomposition layer number according to formula 1:

[0102] —Formula 1;

[0103] N represents the number of initial decomposition layers;

[0104] floor() represents the rounding down function;

[0105] fs represents the sampling frequency of the original electronic control signal;

[0106] Indicates the preset frequency resolution. The minimum bandwidth of the original electronic control signal obtained after N-layer decomposition must be greater than the preset frequency resolution.

[0107] A dynamic decomposition tree is constructed according to the initial decomposition layer number N. The bandwidth of the previous layer is divided into two at each decomposition layer, so that a corresponding number of frequency bands are obtained with the minimum bandwidth at the last decomposition layer, and the original electronic control signal is decomposed into sub-band signals of the corresponding number of frequency bands, including:

[0108] The first layer of decomposition: the sampling frequency of the original electric control signal is used as the initial bandwidth and divided by two, and the first layer The bandwidth of each sub-band is HZ;

[0109] The second layer decomposition: the first layer The bandwidth of each sub-band is divided into two equal parts, and the second layer is obtained. The bandwidth of each sub-band is HZ;

[0110] And so on, decompose in sequence until the Nth layer of decomposition;

[0111] Nth layer decomposition: decomposition of the N-1th layer The bandwidth of each sub-band is divided into two equal parts, and the Nth layer is obtained. The bandwidth of each sub-band is HZ, so that the original electric control signal is converted into the minimum frequency bandwidth HZ is decomposed into sub-band signals;

[0112] Get each subband signal at time The wavelet packet coefficient value is recorded as the sub-band signal amplitude of each sub-band signal .

[0113] In this embodiment, the purpose of decomposing the original electronic control signal into different frequency bands is to provide an analysis basis for subsequent noise identification. For example, if the controller of the air conditioner indoor unit detects abnormal fluctuations in the temperature sensor signal, the following interference may exist:

[0114] Periodic electromagnetic interference, such as 50HZ power frequency noise;

[0115] Random pulse interference, such as caused by compressor start and stop;

[0116] High-frequency noise, such as inverter switching noise.

[0117] Then this method needs to perform wavelet packet decomposition on the original electronic control signal to obtain sub-band signals of different frequency bands, thereby providing a basis for the subsequent analysis of each sub-band signal; compared with the existing wavelet packet decomposition, the number of decomposition layers is a fixed number set manually. This method uses the sampling frequency and the preset frequency resolution to calculate the initial decomposition layer number, and at the same time, corrects the initial decomposition layer number in the dynamic adjustment of the residual analysis in step E to avoid frequency band aliasing caused by the artificially set fixed decomposition number. For example, the fixed decomposition layer number may cause the 50HZ power frequency interference to leak into the adjacent sub-band;

[0118] Specifically, taking the sampling frequency of fs as 200HZ as an example, the preset frequency resolution is If the accuracy requirement is 2HZ, the initial decomposition layer number is layer, so that at a sampling frequency of 200HZ, each layer decomposes the frequency band of the previous layer into two, and the minimum bandwidth obtained after the sixth layer decomposition is , meeting the requirement of greater than preset frequency resolution accuracy requirements.

[0119] Furthermore, the db4 wavelet basis is selected to construct a dynamic decomposition tree. The bandwidth of the previous layer is divided into two when each layer is decomposed, so that the corresponding number of frequency bands is obtained with the minimum bandwidth when the last layer is decomposed, and the original electronic control signal is decomposed into sub-band signals of the corresponding number of frequency bands. For example:

[0120] First-level decomposition: The sampling frequency 200 Hz is used as the initial bandwidth and divided into two equal parts to obtain two sub-band bandwidths. The bandwidth of the first sub-band is 0-100 Hz, and the bandwidth of the second sub-band is 100-200 Hz.

[0121] Second-layer decomposition: The bandwidth of the first sub-band of the first layer (0-100 Hz) and the bandwidth of the second sub-band of the first layer are divided into two equal parts to obtain four sub-band bandwidths: the bandwidth of the first sub-band is 0-50 Hz, the bandwidth of the second sub-band is 50-100 Hz, the bandwidth of the third sub-band is 100-150 Hz, and the bandwidth of the fourth sub-band is 150-200 Hz.

[0122] And so on, until the 6th layer of decomposition, we get a total of Sub-band signals, the bandwidth of the first sub-band signal is 0-3.125HZ, the bandwidth of the second sub-band is 3.125-6.25HZ, ..., the bandwidth of the 31st sub-band is 93.75-96.875HZ... the bandwidth of the 64th sub-band is 196.875-200HZ; then we decompose the original electric control signal S(t) into 64 sub-band signals according to the bandwidth of different sub-bands. Different sub-bands may contain interference signals and real signals. For example, the first sub-band 0-3.125HZ may contain the real temperature change low-frequency signal 0.23HZ, and the 31st sub-band 93.75-96.875HZ may contain the compressor interference signal 95.67HZ.

[0123] Get each subband signal at time The wavelet packet coefficient value is recorded as the sub-band signal amplitude of each sub-band signal For example, for a signal with a sampling frequency of 200 Hz, after 6 layers of wavelet packet decomposition, 64 sub-bands are obtained. Each sub-band contains a set of coefficients. The number of coefficients depends on the length of the original electric control signal. For example, if the original electric control signal has 1000 sampling points, then the number of coefficients in each sub-band is about 1000. The actual wavelet packet decomposition will reduce the number of points due to downsampling, but usually the number of points can be kept consistent through appropriate processing, such as the existing technology of maximum overlap decomposition. Then the sub-band signal amplitude is the i-th subband at time The wavelet packet coefficient value at .

[0124] Step B: Obtain the sliding energy entropy of each subband signal, specifically including:

[0125] Calculate the energy integral of each subband according to Formula 2:

[0126] --Formula 2;

[0127] represents the energy of the Kth subband at time t;

[0128] K represents the subband number, which is used to identify the frequency band position being analyzed;

[0129] t represents the current time, which is used to define the end point of the sliding window, t>T;

[0130] T represents the sliding window width, which is used to determine the time range of energy statistics and is determined according to the air conditioning control cycle;

[0131] Indicates the integral variable from the time point in the interval [tT, t], which is used to traverse all time points in the window;

[0132] represents the sub-band signal amplitude, which is obtained by wavelet decomposition of the original electronic control signal in step A.

[0133] In this embodiment, taking the 31st sub-band 93.75-96.875 Hz as an example, the energy integral of the 31st sub-band is calculated. If the signal amplitude of the 31st sub-band in the window [t-0.5, t] is: [0.1, -0.2, -0.15, 0.3, -0.25] V, where V represents the unit volt and the sampling interval is 0.1 s, then the energy integral

[0134] Here, we choose a sliding window width of 0.5s. This is because the air conditioning control cycle is usually 200ms to 1s, similar to the 20ms period of industrial frequency noise. Therefore, a window width of 0.5s can cover 25 periods, satisfying statistical stability. By calculating the energy integral of the 31st sub-band within the 0.5-second window, if there is a compressor pulse interference signal, the energy here will increase sharply.

[0135] Preferably, the step B specifically includes:

[0136] The energy probability distribution of each subband is calculated according to Formula 3:

[0137] —Formula 3;

[0138] represents the energy probability distribution of the Kth sub-band, which is used to reflect the weight of the frequency band in the total energy;

[0139] M represents the total number of subbands, which is determined by the initial decomposition level N of step A. ;

[0140] j represents the subband index, which is used to traverse all subbands;

[0141] represents the energy of the jth subband at time t;

[0142] represents the energy of the Kth subband at time t.

[0143] In one embodiment, if the total energy of all sub-bands at the current time t is , then the energy probability of the 31st sub-band is .

[0144] Preferably, the step B specifically includes:

[0145] The sliding energy entropy of each subband is calculated according to formula 4:

[0146] —Formula 4;

[0147] represents the sliding energy entropy of the Kth subband, which is used to quantify the signal complexity. ;

[0148] when When it tends to 1, it means that the energy of each sub-band is evenly distributed, indicating that the original electronic control signal is a steady-state signal;

[0149] when When it tends to 0, it means that the energy is concentrated in a few sub-bands, indicating that there is pulse interference in the original electronic control signal.

[0150] In one embodiment, if the energy distribution at a certain moment is:

[0151] Subband 31: ;

[0152] The energy probability of the other 63 subbands is approximately 0.0154;

[0153] Then the real-time energy entropy of the 31st sub-band is

[0154] ; The energy entropy value of 0.72 tends to 1, indicating the presence of moderate noise interference. Ideally, the entropy value of a noise-free signal should be greater than 0.9.

[0155] From the above example, we can see that when a sub-band signal has continuous pulse interference, for example, its energy proportion will increase significantly, causing the real-time energy entropy to decrease, that is, approaching 0. If it is a stationary signal, the energy of each sub-band is evenly distributed, and the real-time energy entropy value will be close to 1.

[0156] Step C: Compare the sliding energy entropy of each sub-band signal with the entropy value threshold to determine the corresponding denoising algorithm, and output each sub-band signal after denoising;

[0157] Preferably, the step C comprises:

[0158] when When , the morphological filtering algorithm is used to denoise the current sub-band and output the denoised sub-band signal;

[0159] when When , the wavelet threshold algorithm is used to denoise the current sub-band and output the denoised sub-band signal;

[0160] when When , the LMS adaptive filtering algorithm is used to denoise the current sub-band and output the denoised sub-band signal;

[0161] and represents the entropy threshold, which is calibrated by the noise sample. , .

[0162] In this embodiment, when When , it means that it is in a high entropy value scene, and the morphological filtering algorithm is used for denoising. The morphological filtering algorithm is a prior art. This application proposes an existing example for explanation. For example, the opening operation formula of the morphological filtering algorithm is , the closing operation formula of the morphological filtering algorithm is , represents the denoised subband signal amplitude obtained after the opening operation, represents the denoised subband signal amplitude obtained after the closing operation, represents the dilation operation using the structuring element B, represents the erosion operation using the structural element B;

[0163] when When , it means that it is in a medium entropy value scenario, and the wavelet threshold algorithm is used for denoising. The wavelet threshold algorithm is a prior art. This application proposes an existing example for explanation. For example, for the sub-band signal amplitude , use the soft threshold processing formula to perform denoising, the soft threshold processing formula is: ; Represents the subband signal amplitude after denoising by wavelet threshold algorithm, sign() represents the sign function, when , ,when , ,when , , represents the soft threshold, , It represents the estimate of the noise standard deviation, which is usually obtained by dividing the median of the absolute value of the sub-band coefficient by the constant 0.6745, that is, , N represents the number of decomposition layers;

[0164] when When , it means that it is in a low entropy value scenario, and the LMS adaptive filtering algorithm is used for denoising. The LMS adaptive filtering algorithm is a prior art. This application proposes an existing example for explanation: The core of the LMS algorithm is the adaptive update formula of the filter coefficient. Assume:

[0165] n is the current moment;

[0166] is the current filter coefficient vector;

[0167] is the input vector (reference noise signal);

[0168] is the expected response (i.e., noisy signal);

[0169] The output of the filter, ,T is transpose;

[0170] is the error signal, ;

[0171] Then the update formula of the filter coefficient is:

[0172] ;

[0173] in, is the step size factor, i.e. the convergence factor.

[0174] Furthermore, for the entropy threshold , higher than this value indicates that the signal structure is stable and suitable for morphological filtering algorithm. , which is lower than this value, indicates strong randomness and requires LMS adaptive filtering algorithm; for the entropy threshold To determine the optimal entropy threshold, we collected experimental data and proved that 0.5 and 0.85 are the optimal entropy thresholds, which can better distinguish slowly varying noise. The experimental data table is as follows:

[0175]

[0176] As shown in the table above, setting a buffer zone between the distribution intervals, that is, the entropy threshold (0.5-0.85), can avoid critical value oscillation.

[0177] Furthermore, as shown in the complementary relationship table of the three algorithms:

[0178]

[0179] The algorithm complementarity table shows that for different frequency bands and different entropy value scenarios, the use of corresponding denoising algorithms can help improve defect compensation in noise detection, solve the adaptability defects of traditional methods in complex noise scenarios, and provide highly robust signal preprocessing guarantees for intelligent air conditioning control.

[0180] Step D: performing weighted reconstruction on the denoised sub-band signal to obtain a reconstructed sub-band signal;

[0181] Preferably, the step D specifically includes:

[0182] This involves weighted reconstruction of each subband signal using Formula 5 and Formula 6:

[0183] —Formula 5;

[0184] —Formula 6;

[0185] represents the sub-band signal after denoising and reconstruction;

[0186] represents the Kth subband signal after denoising, obtained by denoising in step C using the corresponding denoising algorithm;

[0187] Represents the dynamic weight coefficient, which is used to reflect the credibility of the sub-band signal;

[0188] exp() represents the natural exponential function;

[0189] Represents the weight adjustment factor, which is used to reflect the balance between the number of sub-bands and sensitivity;

[0190] Represents the real-time sub-band signal-to-noise ratio, which is used to quantify the sub-band signal quality;

[0191] Indicates the signal-to-noise ratio baseline threshold, which is set according to the noise floor of the air conditioning system.

[0192] In this embodiment, It represents the signal-to-noise ratio threshold, which is set according to the noise floor of the air conditioning system. For example, when the air conditioning is in standby mode, the average SNR of the system floor noise is 16.2dB, and the standard deviation of the SNR fluctuation is 1.2dB. , then it means that when When the dynamic weight coefficient Tends to 1, indicating that the kth subband can be fully trusted. When the dynamic weight coefficient tends to 0, indicating that the kth subband will be suppressed;

[0193] Furthermore, for the weight adjustment factor , we usually take a value of 0.5, as shown in the following table:

[0194] Corresponding to the above table, when When the real-time sub-band SNR is combined with the calculation of the SNR reference threshold, the inferior sub-band can be quickly suppressed in the event of sudden interference. When , a steady-state smooth transition can be achieved.

[0195] Specifically, the following three sub-band embodiments are provided for the instantaneous start of air conditioning cooling:

[0196]

[0197] Reconstructed output:

[0198] , achieving the dominant output of low-frequency effective signals; therefore, through real-time signal-to-noise ratio evaluation, dynamically assigning differentiated weights to sub-bands, breaking through the limitations of traditional fixed weights or average fusion; by setting the signal-to-noise ratio benchmark threshold based on the system background noise statistics, the weight adjustment factor is dynamically configured; ensuring a balance between noise suppression and signal retention: for the dominant output of high-SNR sub-bands, the effective signal can be retained, and for the low-SNR sub-bands, strong suppression can eliminate residual noise, fundamentally solving the problem of effective signal contamination under strong interference in traditional reconstruction methods, providing high-fidelity signal recovery capabilities for air-conditioning intelligent electronic control systems.

[0199] Step E: Calculate a real-time residual evaluation value for the reconstructed subband signal, compare the real-time residual evaluation value with the residual threshold, and determine, based on the comparison result, whether to output the subband signal reconstructed in step D as the final output signal or obtain correction parameters and optimize steps A to C based on the correction parameters. The correction parameters include a new initial decomposition layer number for wavelet packet decomposition after optimization and a new entropy value threshold after optimization.

[0200] Preferably, the step E specifically includes:

[0201] Calculate the real-time residual evaluation value according to Formula 7:

[0202] —Formula 7;

[0203] Represents the real-time residual evaluation value, which is used to quantify the deviation between the denoised signal and the ideal signal;

[0204] Represents the amplitude error weight coefficient, which is used to control the contribution weight of the static deviation;

[0205] Represents the differential error weight coefficient, which is used to control the dynamically changing contribution weight;

[0206] Represents the original electric control signal;

[0207] represents the subband signal after weighted reconstruction in step D;

[0208] It represents the first-order derivative of the original electronic control signal at time t, and is used to represent the instantaneous rate of change of the signal.

[0209] In this embodiment, the real-time residual evaluation value can be calculated by formula 7. The real-time residual evaluation value is used to quantify the deviation between the denoised signal and the ideal signal, that is, the deviation between the denoised signal and the ideal signal we envision is quantified by the real-time residual evaluation value, which is used to subsequently determine whether the parameters of step A and step C need to be optimized. In one example, if the signal range of the air conditioner temperature sensor is 0-5V, corresponding to 0-50°C, the frequency is 200HZ, and the sampling interval is 0.005s;

[0210] The real-time residual evaluation value calculation is shown in the following table:

[0211]

[0212] As can be seen from the table above, when the signal changes rapidly ( ), even if the amplitude error is very small, the real-time residual evaluation value is still greater than 1, indicating that the system is extremely sensitive to the dynamic process.

[0213] Preferably, the step D further comprises:

[0214] Compare the real-time residual evaluation value with the residual threshold:

[0215] Residual Threshold In this embodiment, the noise standard deviation is calculated by collecting the air conditioner standby state signal. For example, a noise signal of 1000 sampling points is used to calculate the noise standard deviation σ = np.std(noise). np.std() represents the function used in the NumPy library to calculate the standard deviation of an array. The experimental data value of the noise standard deviation is 0.0167V. Then, the 3σ criterion covers 99.7% of the noise, and the residual threshold is .

[0216] when When , the subband signal reconstructed by weight in step D is output as the final output signal;

[0217] When the real-time residual evaluation value is less than or equal to the residual threshold, we believe that under the existing parameter calculation of steps A and C, the denoised subband signal can be used as the final output signal after weighted reconstruction; when the real-time residual evaluation value is greater than the residual threshold, we believe that it is necessary to optimize the number of decomposition layers of step A and the entropy threshold of step C.

[0218] when When including:

[0219] According to formula 8, the initial decomposition layer number N is optimized. —Formula 8;

[0220] Indicates the number of decomposition layers after optimization;

[0221] sgn() represents the sign function. hour, ,when hour, ,when hour, ;

[0222] , represents the derivative of the real-time residual evaluation value R(t) at time t, which is used to reflect the changing trend of the real-time residual evaluation value;

[0223] For example, in one embodiment, the initial decomposition level N=6, and the detected , which means that the residual continues to increase, then the number of decomposition layers after optimization , increase the number of initial decomposition layers to achieve finer frequency band division.

[0224] According to formula nine, the entropy value threshold Optimize, —Formula 9;

[0225] Indicates the entropy threshold after optimization, the entropy threshold for the first optimization The entropy threshold value is calibrated by the noise sample at the initial time Based on the artificial calibration, the entropy threshold value after the first optimization is Calculated by formula nine;

[0226] Indicates the learning rate; in this embodiment, it is 0.01;

[0227] Indicates the difference in real-time residual evaluation values ​​before and after optimization;

[0228] Indicates the difference between the entropy thresholds before and after optimization;

[0229] Examples are shown in the following table:

[0230]

[0231] As shown in the table above, the entropy threshold for the first optimization 0.87 is the entropy threshold value calibrated by the initial noise sample =0.85, because it is learned by the ratio of the difference between the residual before and after optimization and the difference between the entropy thresholds before and after optimization, so the entropy threshold of the first optimization needs to be manually calibrated through experience to form a difference before subsequent automatic optimization. Therefore, the entropy threshold obtained after the second optimization is =0.885 is calculated by Formula 9 based on the entropy threshold manually calibrated in the first optimization. That is, starting from the second optimization, the entropy thresholds of subsequent optimizations can be corrected by Formula 9.

[0232] For example, the entropy threshold value of 0.885 obtained by the second optimization is calculated first. , that is, the difference between the residual of the first optimization and the residual of the 0th optimization, , that is, the difference between the entropy threshold of the first optimization and the entropy threshold of the 0th optimization, , substituting it into Formula 9, the entropy threshold obtained by the second optimization is: .

[0233] For example, in one embodiment, when , it means that the entropy threshold needs to be increased , indicating that the entropy threshold of the current decision is too low, and it is necessary to continue to increase the entropy threshold to expand the morphological filtering range.

[0234] The optimized decomposition layer number As the new decomposition level number N, it is returned to step A and a dynamic decomposition tree is constructed using the new decomposition level number N;

[0235] The optimized entropy threshold As the new entropy threshold Return to step C and use the new entropy threshold To expand the filtering range of the morphological filtering algorithm.

[0236] In step E, by constructing the two-dimensional residual, that is, the static amplitude error To ensure the control accuracy, the dynamic differential error terms to ensure response accuracy; by optimizing the number of decomposition layers N and the entropy value threshold, it can fundamentally solve the defect that the traditional open-loop denoising system cannot adapt to complex working conditions, and provide the air-conditioning electronic control system with dual guarantees of anti-interference and control accuracy.

[0237] The technical principles of the present invention have been described above with reference to specific embodiments. These descriptions are intended solely to illustrate the principles of the present invention and are not to be construed in any way as limiting the scope of protection of the present invention. Based on the explanations herein, those skilled in the art will readily conceive of other specific embodiments of the present invention without inventive effort, and such embodiments will fall within the scope of protection of the present invention.

Claims

1. A method for optimizing electrical signals of an air-conditioning electronic control system, characterized in that: Step A: Obtain the original electric control signal of the air conditioner, and decompose the original electric control signal into different frequency bands through wavelet packet decomposition to obtain multiple sub-band signals; Step B: Obtain the sliding energy entropy of each sub-band signal; Step C: Compare the sliding energy entropy of each sub-band signal with the entropy value threshold to determine the corresponding denoising algorithm, and output each sub-band signal after denoising; Step D: performing weighted reconstruction on the denoised sub-band signal to obtain a reconstructed sub-band signal; Step E: Calculate the real-time residual evaluation value of the reconstructed subband signal , the real-time residual evaluation value and residual threshold Comparisons include: when When , the subband signal reconstructed by weight in step D is output as the final output signal; when When including: According to formula 8, the initial decomposition layer number N is optimized. —Formula 8; Indicates the number of decomposition layers after optimization; sgn() represents the sign function. hour, ,when hour, ,when hour, ; , represents the derivative of the real-time residual evaluation value R(t) at time t, which is used to reflect the changing trend of the real-time residual evaluation value; According to formula nine, the entropy value threshold Optimize, —Formula 9; Indicates the entropy threshold after optimization, the entropy threshold for the first optimization The entropy threshold value is calibrated by the noise sample at the initial time Based on the artificial calibration, the entropy threshold value after the first optimization is Calculated by formula nine; represents the learning rate; Indicates the difference in real-time residual evaluation values ​​before and after optimization; Indicates the difference between the entropy thresholds before and after optimization; The optimized decomposition layer number As the new decomposition level N, it is passed back to step A, and the dynamic decomposition tree is constructed using the new decomposition level N; The optimized entropy threshold As the new entropy threshold Return to step C and use the new entropy threshold To expand the filtering range of the morphological filtering algorithm.

2. The method for optimizing electrical signals of an air-conditioning electronic control system according to claim 1, characterized in that: The step A specifically includes: Determine the initial wavelet packet decomposition layer number according to formula 1: —Formula 1; N represents the number of initial decomposition layers; floor() represents the rounding down function; fs represents the sampling frequency of the original electronic control signal; Indicates the preset frequency resolution. The minimum bandwidth of the original electronic control signal obtained after N-layer decomposition must be greater than the preset frequency resolution.

3. The method for optimizing electrical signals of an air-conditioning electronic control system according to claim 2, characterized in that: The step A further comprises: A dynamic decomposition tree is constructed according to the initial decomposition layer number N. The bandwidth of the previous layer is divided into two at each decomposition layer, so that a corresponding number of frequency bands are obtained with the minimum bandwidth at the last decomposition layer, and the original electronic control signal is decomposed into sub-band signals of the corresponding number of frequency bands, including: The first layer of decomposition: the sampling frequency of the original electric control signal is used as the initial bandwidth and divided by two, and the first layer The bandwidth of each sub-band is HZ; The second layer decomposition: the first layer The bandwidth of each sub-band is divided into two equal parts, and the second layer is obtained. The bandwidth of each sub-band is HZ; And so on, decompose in sequence until the Nth layer of decomposition; Nth layer decomposition: decomposition of the N-1th layer The bandwidth of each sub-band is divided into two equal parts, and the Nth layer is obtained. The bandwidth of each sub-band is HZ, so that the original electric control signal is converted into the minimum frequency bandwidth HZ is decomposed into sub-band signals; Get each subband signal at time The wavelet packet coefficient value is recorded as the sub-band signal amplitude of each sub-band signal .

4. The method for optimizing electrical signals of an air-conditioning electronic control system according to claim 3, characterized in that: The step B specifically includes: Calculate the energy integral of each subband according to Formula 2: —Formula 2; represents the energy of the Kth subband at time t; K represents the subband number, which is used to identify the frequency band position being analyzed; t represents the current time, which is used to define the end point of the sliding window, t>T; T represents the sliding window width, which is used to determine the time range of energy statistics and is determined according to the air conditioning control cycle; Indicates the integral variable from the time point in the interval [tT, t], which is used to traverse all time points in the window; represents the sub-band signal amplitude, which is obtained by wavelet decomposition of the original electronic control signal in step A.

5. The method for optimizing electrical signals of an air-conditioning electronic control system according to claim 4, characterized in that: The step B specifically includes: The energy probability distribution of each subband is calculated according to Formula 3: —Formula 3; represents the energy probability distribution of the Kth sub-band, which is used to reflect the weight of the frequency band in the total energy; M represents the total number of subbands, which is determined by the initial decomposition level N of step A. ; j represents the subband index, which is used to traverse all subbands; represents the energy of the jth subband at time t; represents the energy of the Kth subband at time t.

6. The method for optimizing electrical signals of an air-conditioning electronic control system according to claim 5, characterized in that: The step B specifically includes: The sliding energy entropy of each subband is calculated according to formula 4: —Formula 4; represents the sliding energy entropy of the Kth subband, which is used to quantify the signal complexity. ; when When it tends to 1, it means that the energy of each sub-band is evenly distributed, indicating that the original electronic control signal is a steady-state signal; when When it tends to 0, it means that the energy is concentrated in a few sub-bands, indicating that there is pulse interference in the original electronic control signal.

7. The method for optimizing electrical signals of an air-conditioning electronic control system according to claim 6, characterized in that: The step C comprises: when When , the morphological filtering algorithm is used to denoise the current sub-band and output the denoised sub-band signal; when When , the wavelet threshold algorithm is used to denoise the current sub-band and output the denoised sub-band signal; when When , the LMS adaptive filtering algorithm is used to denoise the current sub-band and output the denoised sub-band signal; and represents the entropy threshold, which is calibrated by the noise sample. , .

8. The method for optimizing electrical signals of an air-conditioning electronic control system according to claim 7, characterized in that: The step D specifically includes: This involves weighted reconstruction of each subband signal using Formula 5 and Formula 6: —Formula 5; —Formula 6; represents the sub-band signal after denoising and reconstruction; represents the Kth subband signal after denoising, obtained by denoising in step C using the corresponding denoising algorithm; Represents the dynamic weight coefficient, which is used to reflect the credibility of the sub-band signal; exp() represents the natural exponential function; Represents the weight adjustment factor, which is used to reflect the balance between the number of sub-bands and sensitivity; Represents the real-time sub-band signal-to-noise ratio, which is used to quantify the sub-band signal quality; Indicates the signal-to-noise ratio baseline threshold, which is set according to the noise floor of the air conditioning system.

9. The method for optimizing electrical signals of an air-conditioning electronic control system according to claim 8, characterized in that: The step E specifically includes: Calculate the real-time residual evaluation value according to Formula 7: —Formula 7; Represents the real-time residual evaluation value; Represents the amplitude error weight coefficient, which is used to control the contribution weight of the static deviation; Represents the differential error weight coefficient, which is used to control the dynamically changing contribution weight; Represents the original electric control signal; represents the subband signal after weighted reconstruction in step D; It represents the first-order derivative of the original electronic control signal at time t, and is used to represent the instantaneous rate of change of the signal.

Images