A self-adaptive brightness intelligent control method for a sound-light integrated lamp and a sound-light integrated lamp thereof

Abstract

The present application relates to the technical field of lamp driving, and specifically discloses a sound-light integrated lamp adaptive brightness intelligent control method and a sound-light integrated lamp thereof, which comprises the following steps: acquiring sound signals and illumination signals in an environment, and constructing a sound signal time sequence and an illumination signal time sequence respectively; performing frequency spectrum conversion processing on the sound signal time sequence and the illumination signal time sequence to obtain a first sound-light environment state feature vector; performing state trend prediction on the first sound-light environment state feature vector to obtain a second sound-light environment state feature vector, or taking the first sound-light environment state feature vector as the second sound-light environment state feature vector; and acquiring a brightness parameter corresponding to an actual lighting demand level in a lookup table manner according to the second sound-light environment state feature vector. The sound-light integrated lamp adaptive brightness intelligent control method and the sound-light integrated lamp thereof solve the problem that a traditional lighting system cannot accurately reflect real environment dynamic changes.

Description

Technical Field

[0001] This invention relates to the field of lighting drive technology, and more specifically, to a method for adaptive brightness intelligent control of an integrated sound and light lamp and the integrated sound and light lamp thereof. Background Technology

[0002] Intelligent lighting technology plays a crucial role in modern living environments, serving as a core element in improving quality of life and reducing building energy consumption. Current lighting control schemes often rely on preset fixed values ​​to trigger switching or adjustments. This approach ignores the continuous fluctuations of environmental parameters over time, resulting in a significant lag and disconnect between device response and actual environmental changes. This neglect of environmental parameter fluctuations directly leads to the technical challenge of insufficient analysis of sound and light variation patterns. Because the system cannot accurately capture and analyze the real-time fluctuations of sound and light signals in the environment, it struggles to determine the actual lighting needs of the current scene. For example, in an office environment, when clouds suddenly block sunlight, causing indoor lighting to dim, and simultaneously, the noise from people walking and talking increases, the system cannot react promptly if it cannot recognize this synchronous change in sound and light.

[0003] This lack of understanding of environmental change patterns further hinders the balance between energy efficiency and comfort in lighting equipment operation. Because the system cannot accurately allocate energy consumption based on the dynamic needs of the actual scene, it often maintains high power output to ensure basic lighting, or sacrifices the user's visual experience to save electricity. This results in abrupt and inappropriate brightness adjustments, failing to cater to the personalized usage habits of different users. Therefore, it is difficult to achieve a precise balance between energy saving and user visual comfort in lighting equipment. Summary of the Invention

[0004] In order to overcome the defects of the existing technology, the present invention provides an adaptive brightness intelligent control method for an integrated sound and light lamp and an integrated sound and light lamp thereof, aiming to solve the problems in the above-mentioned existing technology.

[0005] The technical solution adopted by this invention to solve its technical problem is: a method for intelligent adaptive brightness control of an integrated sound and light lamp, comprising the following steps: S1: Acquire sound and light signals from the environment, and construct sound signal time sequence and light signal time sequence respectively; S2: The frequency domain conversion algorithm is used to perform spectral conversion processing on the acoustic signal time sequence and the optical signal time sequence respectively to obtain the acoustic signal frequency domain complex sequence and the optical signal frequency domain complex sequence; the acoustic signal frequency domain amplitude fluctuation characteristics corresponding to the acoustic signal frequency domain complex sequence and the optical signal frequency domain amplitude fluctuation characteristics corresponding to the optical signal frequency domain complex sequence are calculated and concatenated into a first acoustic-optical environment state feature vector; S3: Determine whether the frequency domain amplitude fluctuation feature of the acoustic signal and the frequency domain amplitude fluctuation feature of the optical signal in the first acoustic and optical environment state feature vector are both greater than the corresponding preset environmental change threshold. If so, input the first acoustic and optical environment state feature vector into the time series prediction model to predict the state trend and obtain the second acoustic and optical environment state feature vector. Otherwise, output the first acoustic and optical environment state feature vector as the second acoustic and optical environment state feature vector. S4: Based on the frequency domain amplitude fluctuation characteristics of the acoustic signal and the frequency domain amplitude fluctuation characteristics of the optical signal in the second acoustic and optical environment state feature vector, query the pre-established relational database, and determine the actual lighting requirement level of the current environment through mapping and matching; S5: Based on the actual lighting requirement level, obtain the brightness parameter corresponding to the actual lighting requirement level by looking up a table.

[0006] Preferably, in step S1, the step of constructing the time sequence of the acoustic signal includes: An analog-to-digital converter is used to perform continuous sampling operations on the ambient sound signal to obtain an initial sound intensity sequence; The initial sound intensity sequence is processed using a maximum-minimum normalization function to unify the dimensions and construct a time sequence of the sound signal.

[0007] Optionally, in step S1, the step of constructing the optical signal timing sequence includes: An analog-to-digital converter is used to perform continuous sampling operations on the ambient light signal to obtain an initial light intensity sequence; The initial light intensity sequence is processed using a maximum-minimum normalization function to unify the dimensions and construct a time sequence of the optical signal.

[0008] Specifically, in step S2, the steps of obtaining the acoustic signal frequency domain signal and the optical signal frequency domain signal include: The acoustic signal time sequence and the optical signal time sequence are segmented using a Hamming window function to obtain the acoustic signal time-domain frame sequence and the optical signal time-domain frame sequence; The acoustic signal time-domain frame sequence and the optical signal time-domain frame sequence are subjected to a fast Fourier transform algorithm to perform a spectral transformation to obtain a complex frequency domain sequence of the acoustic signal and a complex frequency domain sequence of the optical signal.

[0009] It is worth noting that, in step S2, the step of concatenating the first acoustic-optical environment state feature vector includes: For each complex number in the frequency domain sequence of the acoustic signal, the complex modulus calculation formula is used to perform calculations to obtain the frequency domain amplitude sequence of the acoustic signal. For each optical signal frequency domain complex number in the optical signal frequency domain complex number sequence, the complex modulus calculation formula is used to perform calculation to obtain the optical signal frequency domain amplitude sequence; The standard deviation of the frequency domain amplitude sequence of the acoustic signal is calculated to obtain the frequency domain amplitude fluctuation characteristics of the acoustic signal, and the standard deviation of the frequency domain amplitude sequence of the optical signal is calculated to obtain the frequency domain amplitude fluctuation characteristics of the optical signal. The first acoustic-optical environment state feature vector is obtained by concatenating the frequency domain amplitude fluctuation characteristics of the acoustic signal and the frequency domain amplitude fluctuation characteristics of the optical signal.

[0010] Specifically, in step S3, the step of inputting the first acoustic-optical environment state feature vector into the time-series prediction model to predict the state trend and obtain the second acoustic-optical environment state feature vector includes: The first acoustic-optical environment state feature vector The input is fed into a Long Short-Term Memory (LSTM) network to perform state trend prediction, resulting in multiple initial prediction vectors. The prediction sequence is composed of b1, where b1 represents the frequency domain amplitude fluctuation characteristics of the acoustic signal, and b2 represents the frequency domain amplitude fluctuation characteristics of the optical signal. These are the characteristic values ​​of the acoustic signal. These are the characteristic values ​​of the optical signal; Get the preset sampling time interval ; Obtain the first initial prediction vector in the prediction sequence and the second initial prediction vector Difference of characteristic values ​​of mid-sound signal Calculate the slope of the trend evolution of the acoustic signal. ; Obtain the first initial prediction vector in the prediction sequence and the second initial prediction vector Difference of optical signal eigenvalues Calculate the slope of the trend evolution of optical signals ; The time decay weights of the acoustic signal and optical signal are calculated using an exponential decay function on the slopes of the acoustic signal and optical signal trends, respectively. The acoustic signal time decay weight... Optical signal time decay weight , where 'a' is the preset attenuation coefficient; Calculate the eigenvalues ​​of the first weighted acoustic signal Calculate the first weighted optical signal eigenvalue. ; Calculate the eigenvalues ​​of the second weighted acoustic signal Calculate the eigenvalues ​​of the second weighted optical signal. ; Calculate the frequency domain amplitude fluctuation characteristics of the updated acoustic signal Calculate the frequency domain amplitude fluctuation characteristics of the updated optical signal. The updated frequency domain amplitude fluctuation features of the acoustic signal and the updated frequency domain amplitude fluctuation features of the optical signal are concatenated to obtain the second acoustic-optical environment state feature vector. .

[0011] Preferably, in step S4, the step of determining the actual lighting requirement level of the current environment includes: Extract the frequency domain amplitude fluctuation features of the acoustic signal and the frequency domain amplitude fluctuation features of the optical signal from the second acoustic-optical environment state feature vector; Input the frequency domain amplitude fluctuation characteristics of the acoustic signal and the frequency domain amplitude fluctuation characteristics of the optical signal into a pre-established relational database; In a pre-established relational database, the frequency domain amplitude fluctuation characteristics of the acoustic signal are numerically compared with the preset acoustic signal feature interval to obtain the target acoustic signal feature interval where the frequency domain amplitude fluctuation characteristics of the acoustic signal are located. In a pre-established relational database, the frequency domain amplitude fluctuation characteristics of the optical signal are numerically compared with the preset optical signal characteristic range to obtain the target optical signal characteristic range in which the frequency domain amplitude fluctuation characteristics of the optical signal are located. Based on the target acoustic signal characteristic range, the corresponding first ambient lighting demand level value is obtained from the pre-established relational database; based on the target optical signal characteristic range, the corresponding second ambient lighting demand level value is obtained from the pre-established relational database. If the first ambient lighting demand level value is the same as the second ambient lighting demand level value, then the first ambient lighting demand level value is taken as the actual lighting demand level of the current environment; if the first ambient lighting demand level value is different from the second ambient lighting demand level value, then the maximum value between the first ambient lighting demand level value and the second ambient lighting demand level value is obtained, and the maximum value is taken as the actual lighting demand level of the current environment.

[0012] A sound and light integrated lamp, using the aforementioned sound and light integrated lamp adaptive brightness intelligent control method.

[0013] The beneficial effects of this invention are as follows: In the integrated sound and light lamp adaptive brightness intelligent control method, sound signals and light signals in the environment are collected simultaneously. A frequency domain conversion algorithm is used to extract the frequency domain amplitude fluctuation characteristics of the two types of signals, and these are then concatenated to form a sound and light environment state feature vector. When the fluctuation characteristics of both sound and light signals exceed a preset environmental change threshold, it indicates that the environment is undergoing significant change. At this point, a time-series prediction model is activated to predict the future environmental state, thereby achieving a forward-looking judgment of lighting needs. Based on the current or predicted sound and light environment state feature vector, the system accurately determines the actual lighting demand level and obtains the corresponding brightness parameters by querying a pre-established environmental demand mapping table. This solution, through multi-dimensional fusion analysis of sound and light signals and an intelligent prediction mechanism, effectively improves the accuracy and timeliness of the lighting system's perception of environmental changes, achieving a more intelligent and user-friendly lighting control effect. Attached Figure Description

[0014] Figure 1 The flowchart shows the adaptive brightness intelligent control method for an integrated sound and light lamp.

[0015] Figure 2 This is a flowchart of the steps in step S2.

[0016] Figure 3 This is a schematic diagram of the structure of an integrated sound and light lamp.

[0017] In the picture: 1. Sound and light integrated lamp, 2. Light assembly, 3. Housing, 4. Microphone. Detailed Implementation

[0018] The specific embodiments of the present invention will be further described below with reference to the accompanying drawings. It should be noted that these descriptions are for the purpose of aiding understanding the present invention, but do not constitute a limitation thereof. Furthermore, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0019] Combination Figures 1 to 3 The method for adaptive brightness intelligent control of an integrated sound and light lamp, as shown, includes the following steps: S1: Acquire sound and light signals from the environment, and construct sound signal time sequence and light signal time sequence respectively; S2: The frequency domain conversion algorithm is used to perform spectral conversion processing on the acoustic signal time sequence and the optical signal time sequence respectively to obtain the acoustic signal frequency domain complex sequence and the optical signal frequency domain complex sequence; the acoustic signal frequency domain amplitude fluctuation characteristics corresponding to the acoustic signal frequency domain complex sequence and the optical signal frequency domain amplitude fluctuation characteristics corresponding to the optical signal frequency domain complex sequence are calculated and concatenated into a first acoustic-optical environment state feature vector; S3: Determine whether the frequency domain amplitude fluctuation feature of the acoustic signal and the frequency domain amplitude fluctuation feature of the optical signal in the first acoustic and optical environment state feature vector are both greater than the corresponding preset environmental change threshold. If so, input the first acoustic and optical environment state feature vector into the time series prediction model to predict the state trend and obtain the second acoustic and optical environment state feature vector. Otherwise, output the first acoustic and optical environment state feature vector as the second acoustic and optical environment state feature vector. S4: Based on the frequency domain amplitude fluctuation characteristics of the acoustic signal and the frequency domain amplitude fluctuation characteristics of the optical signal in the second acoustic and optical environment state feature vector, query the pre-established relational database, and determine the actual lighting requirement level of the current environment through mapping and matching; S5: Based on the actual lighting requirement level, obtain the brightness parameter corresponding to the actual lighting requirement level by looking up a table.

[0020] In the proposed integrated sound and light lighting adaptive brightness intelligent control method, sound and light signals from the environment are simultaneously collected. A frequency domain conversion algorithm is used to extract the frequency domain amplitude fluctuation characteristics of both signals, and these are then concatenated to form a sound and light environment state feature vector. When the fluctuation characteristics of both sound and light signals exceed a preset environmental change threshold, it indicates that the environment is undergoing significant change. At this point, a time-series prediction model is activated to predict future environmental conditions, thereby achieving a proactive assessment of lighting needs. Based on the current or predicted sound and light environment state feature vector, the system accurately determines the actual lighting demand level and obtains the corresponding brightness parameters by querying a pre-established environmental demand mapping table. This solution, through multi-dimensional fusion analysis of sound and light signals and an intelligent prediction mechanism, effectively improves the accuracy and timeliness of the lighting system's perception of environmental changes, achieving a more intelligent and user-friendly lighting control effect.

[0021] It is worth noting that, in step S1, the steps of constructing the time sequence of the acoustic signal include: An analog-to-digital converter is used to perform continuous sampling operations on the ambient sound signal to obtain an initial sound intensity sequence; The initial sound intensity sequence is processed using a maximum-minimum normalization function to unify the dimensions and construct a time sequence of the sound signal.

[0022] In this embodiment, ambient sound signals are acquired using a microphone, and then continuous sampling is performed on each ambient sound signal using an analog-to-digital converter. The analog-to-digital converter converts the continuous analog signal into a discrete digital signal, and periodically captures the sound wave amplitude at a preset sampling time interval to obtain an initial sound intensity sequence.

[0023] Preferably, in step S1, the step of constructing the optical signal timing sequence includes: An analog-to-digital converter is used to perform continuous sampling operations on the ambient light signal to obtain an initial light intensity sequence; The initial light intensity sequence is processed using a maximum-minimum normalization function to unify the dimensions and construct a time sequence of the optical signal.

[0024] In this embodiment, ambient light signals are acquired using photodiodes, and then continuous sampling is performed on these signals using analog-to-digital converters (ADCs). The ADCs convert continuous analog signals into discrete digital signals, and periodically capture the light intensity at preset sampling time intervals to obtain an initial light intensity sequence.

[0025] It should be noted that since the initial sound intensity sequence is in decibels (dB) and the initial light intensity sequence is in lux (lux), their numerical ranges differ significantly. Direct joint analysis would cause the larger signal to mask the smaller signal's characteristics. Therefore, a maximum-minimum normalization function is used to unify the dimensions of the initial sound intensity and initial light intensity sequences. Maximum-minimum normalization is a linear transformation technique. The process involves first traversing the initial sound intensity sequence to find the maximum and minimum sound intensity values. Then, the minimum sound intensity value is subtracted from each current sound intensity value in the sequence to obtain the difference. Simultaneously, the range between the maximum and minimum sound intensity values ​​is calculated. Finally, the difference is divided by this range, thus proportionally mapping each original sound intensity value to a dimensionless interval of zero to one. The processing for the initial light intensity sequence is completely consistent, extracting the light intensity extrema and applying the same linear scaling to eliminate the original unit limitations. After this dimension unification process, the system successfully constructs sound and light signal time sequences at the same numerical scale. This process effectively eliminates dimensional barriers between data from multi-source heterogeneous sensors.

[0026] Optionally, in step S2, the steps of obtaining the acoustic signal frequency domain signal and the optical signal frequency domain signal include: The acoustic signal time sequence and the optical signal time sequence are segmented using a Hamming window function to obtain the acoustic signal time-domain frame sequence and the optical signal time-domain frame sequence; The acoustic signal time-domain frame sequence and the optical signal time-domain frame sequence are subjected to a fast Fourier transform algorithm to perform a spectral transformation to obtain a complex frequency domain sequence of the acoustic signal and a complex frequency domain sequence of the optical signal.

[0027] For example, the system acquires the acoustic signal time sequence and optical signal time sequence output in the previous step, and then uses a Hamming window function to segment the acoustic signal time sequence and optical signal time sequence to obtain the acoustic signal time-domain frame sequence and optical signal time-domain frame sequence. It should be noted that the Hamming window function is a smooth weighting function. The system sets a fixed time window length and sliding step size to divide the continuous sequence into multiple overlapping data segments. For each data segment, the values ​​of its internal sampling points are multiplied one by one by the weight coefficients corresponding to the Hamming window function, so that the middle sampling points retain a larger weight, and the weights of the edge sampling points gradually decrease, thus completing the construction of the time-domain frame sequence. For the acoustic signal time sequence, the last N most recent acoustic signals captured in the current window are extracted as the acoustic signal time-domain frame sequence; for the optical signal time sequence, the last N most recent optical signals captured in the current window are extracted as the optical signal time-domain frame sequence. Finally, the fast Fourier transform algorithm is used to perform a spectral transformation on the time-domain frame sequences to obtain the corresponding frequency domain complex sequence. The Fast Fourier Transform algorithm can decompose a time-domain signal into components of different frequencies and output a complex number containing both real and imaginary parts.

[0028] Specifically, in step S2, the step of concatenating the first acoustic-optical environment state feature vector includes: For each complex number in the frequency domain sequence of the acoustic signal, the complex modulus calculation formula is used to perform calculations to obtain the frequency domain amplitude sequence of the acoustic signal. For each optical signal frequency domain complex number in the optical signal frequency domain complex number sequence, the complex modulus calculation formula is used to perform calculation to obtain the optical signal frequency domain amplitude sequence; The standard deviation of the frequency domain amplitude sequence of the acoustic signal is calculated to obtain the frequency domain amplitude fluctuation characteristics of the acoustic signal, and the standard deviation of the frequency domain amplitude sequence of the optical signal is calculated to obtain the frequency domain amplitude fluctuation characteristics of the optical signal. The first acoustic-optical environment state feature vector is obtained by concatenating the frequency domain amplitude fluctuation characteristics of the acoustic signal and the frequency domain amplitude fluctuation characteristics of the optical signal.

[0029] In this embodiment, for each frequency domain complex number, the complex modulus calculation formula is used to perform calculations, extracting the real and imaginary parts, calculating the squares of the real and imaginary parts respectively, adding them together and taking the square root to obtain the true amplitude, thus obtaining the frequency domain amplitude sequence of the acoustic signal and the frequency domain amplitude sequence of the optical signal. When calculating the standard deviation of the amplitude sequence, the square root of the mean of the squares of the differences between each amplitude and the average value is taken to obtain the corresponding frequency domain amplitude fluctuation feature. Finally, when concatenating the frequency domain amplitude fluctuation features of the acoustic signal and the frequency domain amplitude fluctuation features of the optical signal to obtain the first acoustic-optical environment state feature vector, the frequency domain amplitude fluctuation feature of the acoustic signal is used as the first element of the first acoustic-optical environment state feature vector, and the frequency domain amplitude fluctuation of the optical signal is used as the second element of the first acoustic-optical environment state feature vector.

[0030] It is worth noting that, in step S3, the step of inputting the first acoustic-optical environment state feature vector into the time-series prediction model to predict the state trend and obtain the second acoustic-optical environment state feature vector includes: The first acoustic-optical environment state feature vector The input is fed into a Long Short-Term Memory (LSTM) network to perform state trend prediction, resulting in multiple initial prediction vectors. The prediction sequence is composed of b1, where b1 represents the frequency domain amplitude fluctuation characteristics of the acoustic signal, and b2 represents the frequency domain amplitude fluctuation characteristics of the optical signal. These are the characteristic values ​​of the acoustic signal. These are the characteristic values ​​of the optical signal; Get the preset sampling time interval ; Obtain the first initial prediction vector in the prediction sequence and the second initial prediction vector Difference of characteristic values ​​of mid-sound signal Calculate the slope of the trend evolution of the acoustic signal. ; Obtain the first initial prediction vector in the prediction sequence and the second initial prediction vector Difference of optical signal eigenvalues Calculate the slope of the trend evolution of optical signals ; The time decay weights of the acoustic signal and optical signal are calculated using an exponential decay function on the slopes of the acoustic signal and optical signal trends, respectively. The acoustic signal time decay weight... Optical signal time decay weight , where 'a' is the preset attenuation coefficient; Calculate the eigenvalues ​​of the first weighted acoustic signal Calculate the first weighted optical signal eigenvalue. ; Calculate the eigenvalues ​​of the second weighted acoustic signal Calculate the eigenvalues ​​of the second weighted optical signal. ; Calculate the frequency domain amplitude fluctuation characteristics of the updated acoustic signal Calculate the frequency domain amplitude fluctuation characteristics of the updated optical signal. The updated frequency domain amplitude fluctuation features of the acoustic signal and the updated frequency domain amplitude fluctuation features of the optical signal are concatenated to obtain the second acoustic-optical environment state feature vector. .

[0031] In this embodiment, if the thresholds are not simultaneously exceeded, the system directly outputs the first acoustic-optical environment state feature vector as the second acoustic-optical environment state feature vector. If both exceed the thresholds, the system inputs the first acoustic-optical environment state feature vector into a long short-term memory network to perform state trend prediction. The long short-term memory network includes a forget gate, an input gate, and an output gate. It filters information through internal cell state transfer and gating mechanisms, and outputs an initial prediction vector for the future environmental state.

[0032] Specifically, the preset environmental change threshold is a quantitative critical value used to distinguish between normal minor environmental disturbances and substantial synchronous environmental abrupt changes. When both the frequency domain amplitude fluctuation characteristics of the acoustic signal and the frequency domain amplitude fluctuation characteristics of the optical signal are greater than the corresponding preset environmental change threshold, it indicates that a substantial synchronous environmental abrupt change has occurred, with noise spikes such as pulse interference or electromagnetic noise present. The remaining cases indicate normal minor environmental disturbances. In the case of a substantial synchronous environmental abrupt change, replacing the original first acoustic-optical environmental state feature vector with a fused state vector can take into account both the actual environmental change trend and real-time measured data, suppressing random instantaneous spikes.

[0033] Preferably, in step S4, the step of determining the actual lighting requirement level of the current environment includes: Extract the frequency domain amplitude fluctuation features of the acoustic signal and the frequency domain amplitude fluctuation features of the optical signal from the second acoustic-optical environment state feature vector; Input the frequency domain amplitude fluctuation characteristics of the acoustic signal and the frequency domain amplitude fluctuation characteristics of the optical signal into a pre-established relational database; In a pre-established relational database, the frequency domain amplitude fluctuation characteristics of the acoustic signal are numerically compared with the preset acoustic signal feature interval to obtain the target acoustic signal feature interval where the frequency domain amplitude fluctuation characteristics of the acoustic signal are located. In a pre-established relational database, the frequency domain amplitude fluctuation characteristics of the optical signal are numerically compared with the preset optical signal characteristic range to obtain the target optical signal characteristic range in which the frequency domain amplitude fluctuation characteristics of the optical signal are located. Based on the target acoustic signal characteristic range, the corresponding first ambient lighting demand level value is obtained from the pre-established relational database; based on the target optical signal characteristic range, the corresponding second ambient lighting demand level value is obtained from the pre-established relational database. If the first ambient lighting demand level value is the same as the second ambient lighting demand level value, then the first ambient lighting demand level value is taken as the actual lighting demand level of the current environment; if the first ambient lighting demand level value is different from the second ambient lighting demand level value, then the maximum value between the first ambient lighting demand level value and the second ambient lighting demand level value is obtained, and the maximum value is taken as the actual lighting demand level of the current environment.

[0034] Specifically, a pre-established relational database stores multiple preset acoustic signal characteristic intervals and preset optical signal characteristic intervals, as well as the corresponding environmental lighting requirement levels for each interval. The preset characteristic intervals are continuous numerical ranges divided based on the intensity of signal fluctuations in historical environmental data.

[0035] The system performs numerical comparison operations in a relational database, comparing the extracted frequency domain amplitude fluctuation characteristics of the acoustic signal with the upper and lower limits of various preset acoustic signal feature intervals to determine the specific range in which it falls, thereby obtaining the target acoustic signal feature interval. Similarly, the system compares the frequency domain amplitude fluctuation characteristics of the optical signal with preset optical signal feature intervals to locate the target optical signal feature interval. Subsequently, the system uses the key-value mapping relationship of the relational database to query the corresponding first ambient lighting demand level value based on the target acoustic signal feature interval, and the corresponding second ambient lighting demand level value based on the target optical signal feature interval.

[0036] In this embodiment, the ambient lighting demand level is typically set to an integer from one to five, with a higher value indicating a higher demand for lighting intensity in the current environment. It should be noted that if the first and second ambient lighting demand level values ​​obtained are completely identical, the system directly uses this value as the actual lighting demand level for the current environment. If the first and second ambient lighting demand level values ​​are inconsistent, it indicates a difference in lighting demands reflected by the acoustic and light environments. In this case, the system takes the larger of the two values ​​as the actual lighting demand level for the current environment. This strategy of taking the maximum value is based on a safety redundancy business rule setting. When the acoustic environment indicates dense human activity while the light environment indicates sufficient natural light, taking the maximum value covers the highest level of lighting conditions.

[0037] Finally, the system receives the actual lighting demand level and performs a key-value pair matching operation in a preset parameter mapping table to extract the corresponding target brightness parameter. Specifically, the parameter mapping table is stored in memory, with the lighting demand level as the key and the target brightness parameter as the value, which is typically the duty cycle setting of a pulse width modulation (PWM). The system sends the duty cycle setting to the PWM controller, which generates a digital square wave signal with a specific high-low level time ratio. This signal serves as the lighting adjustment command, which drives the lighting equipment to adjust its brightness.

[0038] Optionally, a sound and light integrated lamp uses the aforementioned sound and light integrated lamp adaptive brightness intelligent control method.

[0039] like Figure 3As shown, the integrated sound and light lamp 1 includes a lampshade 2, a housing 3, a microphone 4, and a photodiode disposed outside the housing and away from the lampshade 2. The lampshade 2 is connected to the housing 3, and the microphone 4 is disposed on the lampshade 2. After the integrated sound and light lamp 1 is installed on the ceiling, the lampshade 2 faces downwards, so that the microphone 4 can collect the sound generated below the integrated sound and light lamp 1. The photodiode is disposed away from the lampshade 2 to avoid the light generated by the lamp assembly inside the lampshade 2 affecting the ambient brightness collected by the photodiode.

[0040] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings, but the present invention is not limited to the described embodiments. For those skilled in the art, various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and these variations still fall within the protection scope of the present invention.

Claims

1. A method for adaptive brightness intelligent control of an integrated sound and light lamp, characterized in that, Includes the following steps: S1: Acquire sound and light signals from the environment, and construct sound signal time sequence and light signal time sequence respectively; S2: The acoustic signal time sequence and the optical signal time sequence are processed by frequency domain conversion algorithm to obtain the acoustic signal frequency domain complex sequence and the optical signal frequency domain complex sequence respectively; Calculate the frequency domain amplitude fluctuation characteristics of the acoustic signal corresponding to the frequency domain complex sequence and the frequency domain amplitude fluctuation characteristics of the optical signal corresponding to the frequency domain complex sequence, and concatenate them into a first acoustic-optical environment state feature vector; S3: Determine whether the frequency domain amplitude fluctuation feature of the acoustic signal and the frequency domain amplitude fluctuation feature of the optical signal in the first acoustic and optical environment state feature vector are both greater than the corresponding preset environmental change threshold. If so, input the first acoustic and optical environment state feature vector into the time series prediction model to predict the state trend and obtain the second acoustic and optical environment state feature vector. Otherwise, output the first acoustic and optical environment state feature vector as the second acoustic and optical environment state feature vector. S4: Based on the frequency domain amplitude fluctuation characteristics of the acoustic signal and the frequency domain amplitude fluctuation characteristics of the optical signal in the second acoustic and optical environment state feature vector, query the pre-established relational database, and determine the actual lighting requirement level of the current environment through mapping and matching; S5: Based on the actual lighting requirement level, obtain the brightness parameter corresponding to the actual lighting requirement level by looking up a table.

2. The method for adaptive brightness intelligent control of an integrated sound and light lamp according to claim 1, characterized in that, In step S1, the steps of constructing the time sequence of the acoustic signal include: An analog-to-digital converter is used to perform continuous sampling operations on the ambient sound signal to obtain an initial sound intensity sequence; The initial sound intensity sequence is processed using a maximum-minimum normalization function to unify the dimensions and construct a time sequence of the sound signal.

3. The method for adaptive brightness intelligent control of an integrated sound and light lamp according to claim 2, characterized in that, In step S1, the step of constructing the optical signal timing sequence includes: An analog-to-digital converter is used to perform continuous sampling operations on the ambient light signal to obtain an initial light intensity sequence; The initial light intensity sequence is processed using a maximum-minimum normalization function to unify the dimensions and construct a time sequence of the optical signal.

4. The method for adaptive brightness intelligent control of an integrated sound and light lamp according to claim 1, characterized in that, In step S2, the steps of obtaining the frequency domain signals of the acoustic signal and the optical signal include: The acoustic signal time sequence and the optical signal time sequence are segmented using a Hamming window function to obtain the acoustic signal time-domain frame sequence and the optical signal time-domain frame sequence; The acoustic signal time-domain frame sequence and the optical signal time-domain frame sequence are subjected to a fast Fourier transform algorithm to perform a spectral transformation to obtain a complex frequency domain sequence of the acoustic signal and a complex frequency domain sequence of the optical signal.

5. The method for adaptive brightness intelligent control of an integrated sound and light lamp according to claim 4, characterized in that, In step S2, the step of concatenating the first acoustic-optical environment state feature vector includes: For each complex number in the frequency domain sequence of the acoustic signal, the complex modulus calculation formula is used to perform calculations to obtain the frequency domain amplitude sequence of the acoustic signal. For each optical signal frequency domain complex number in the optical signal frequency domain complex number sequence, the complex modulus calculation formula is used to perform calculation to obtain the optical signal frequency domain amplitude sequence; The standard deviation of the frequency domain amplitude sequence of the acoustic signal is calculated to obtain the frequency domain amplitude fluctuation characteristics of the acoustic signal, and the standard deviation of the frequency domain amplitude sequence of the optical signal is calculated to obtain the frequency domain amplitude fluctuation characteristics of the optical signal. The first acoustic-optical environment state feature vector is obtained by concatenating the frequency domain amplitude fluctuation characteristics of the acoustic signal and the frequency domain amplitude fluctuation characteristics of the optical signal.

6. The method for adaptive brightness intelligent control of an integrated sound and light lamp according to claim 1, characterized in that, In step S3, the step of inputting the first acoustic-optical environment state feature vector into the time-series prediction model to predict the state trend and obtain the second acoustic-optical environment state feature vector includes: The first acoustic-optical environment state feature vector The input is fed into a Long Short-Term Memory (LSTM) network to perform state trend prediction, resulting in multiple initial prediction vectors. The prediction sequence is composed of b1, where b1 represents the frequency domain amplitude fluctuation characteristics of the acoustic signal, and b2 represents the frequency domain amplitude fluctuation characteristics of the optical signal. These are the characteristic values ​​of the acoustic signal. These are the characteristic values ​​of the optical signal; Get the preset sampling time interval ; Obtain the first initial prediction vector in the prediction sequence and the second initial prediction vector Difference of characteristic values ​​of mid-sound signal Calculate the slope of the trend evolution of the acoustic signal. ; Obtain the first initial prediction vector in the prediction sequence and the second initial prediction vector Difference of optical signal eigenvalues Calculate the slope of the trend evolution of optical signals ; The time decay weights of the acoustic signal and optical signal are calculated using an exponential decay function on the slopes of the acoustic signal and optical signal trends, respectively. The acoustic signal time decay weight... Optical signal time decay weight , where 'a' is the preset attenuation coefficient; Calculate the eigenvalues ​​of the first weighted acoustic signal Calculate the first weighted optical signal eigenvalue. ; Calculate the eigenvalues ​​of the second weighted acoustic signal Calculate the eigenvalues ​​of the second weighted optical signal. ; Calculate the frequency domain amplitude fluctuation characteristics of the updated acoustic signal Calculate the frequency domain amplitude fluctuation characteristics of the updated optical signal. The updated frequency domain amplitude fluctuation features of the acoustic signal and the updated frequency domain amplitude fluctuation features of the optical signal are concatenated to obtain the second acoustic-optical environment state feature vector. .

7. The method for adaptive brightness intelligent control of an integrated sound and light lamp according to claim 1, characterized in that, In step S4, the step of determining the actual lighting requirement level of the current environment includes: Extract the frequency domain amplitude fluctuation features of the acoustic signal and the frequency domain amplitude fluctuation features of the optical signal from the second acoustic-optical environment state feature vector; Input the frequency domain amplitude fluctuation characteristics of the acoustic signal and the frequency domain amplitude fluctuation characteristics of the optical signal into a pre-established relational database; In a pre-established relational database, the frequency domain amplitude fluctuation characteristics of the acoustic signal are numerically compared with the preset acoustic signal feature interval to obtain the target acoustic signal feature interval where the frequency domain amplitude fluctuation characteristics of the acoustic signal are located. In a pre-established relational database, the frequency domain amplitude fluctuation characteristics of the optical signal are numerically compared with the preset optical signal characteristic range to obtain the target optical signal characteristic range in which the frequency domain amplitude fluctuation characteristics of the optical signal are located. Based on the target acoustic signal characteristic range, the corresponding first ambient lighting demand level value is obtained from the pre-established relational database; based on the target optical signal characteristic range, the corresponding second ambient lighting demand level value is obtained from the pre-established relational database. If the first ambient lighting demand level value is the same as the second ambient lighting demand level value, then the first ambient lighting demand level value is taken as the actual lighting demand level of the current environment; if the first ambient lighting demand level value is different from the second ambient lighting demand level value, then the maximum value between the first ambient lighting demand level value and the second ambient lighting demand level value is obtained, and the maximum value is taken as the actual lighting demand level of the current environment.

8. A sound and light integrated lamp, characterized in that, The method for adaptive brightness control of an integrated sound and light lamp as described in any one of claims 1-7.

Images