MEMS sensor signal conversion method and system based on cycle-by-cycle noise reduction

Abstract

The invention discloses an MEMS sensor signal conversion method and system based on cycle-by-cycle noise reduction, and relates to the technical field of signal noise reduction, and the method comprises the steps: obtaining a first signal sequence of an MEMS sensor collection target device in a first sampling cycle, and carrying out the noise filtering according to a preset noise reduction parameter, and obtaining a first noise filtering signal sequence and a first noise signal sequence; collecting a first operation parameter sequence of the target equipment, and predicting a second prediction operation parameter sequence and a second prediction noise signal sequence in a second sampling period according to the first noise signal sequence; performing noise reduction parameter optimization in a second sampling period according to the second predicted noise signal sequence to obtain a second optimized noise reduction parameter; and performing noise reduction control in the second sampling period according to the second optimized noise reduction parameter, and continuing to perform period-by-period noise reduction signal conversion. According to the invention, the technical problem of poor signal noise reduction effect in the prior art is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of signal noise reduction, and in particular to a MEMS sensor signal conversion method and system based on cycle-by-cycle noise reduction. Background Art

[0002] In the industrial sector, equipment noise reduction is a common technical challenge faced by all equipment. Traditional noise reduction technologies often use fixed filtering or static parameter configurations, which are difficult to adapt to real-time changes in operating conditions. Fixed signal conversion schemes are prone to insufficient noise reduction or excessive noise suppression when operating conditions suddenly change, and they cannot meet the real-time requirements of highly dynamic scenarios. In addition, existing methods often rely on the statistical characteristics of historical data, with limited ability to track instantaneous burst noise and rapid operating condition transitions, resulting in potentially poor transient noise reduction effects. More importantly, noise characteristics are strongly coupled with the operating status of the equipment, requiring a dynamic balance between noise suppression and equipment operation. Summary of the Invention

[0003] The present application provides a MEMS sensor signal conversion method and system based on cycle-by-cycle noise reduction, which is used to solve the technical problem of poor signal noise reduction effect in the prior art.

[0004] In view of the above problems, the present application provides a MEMS sensor signal conversion method and system based on cycle-by-cycle noise reduction.

[0005] In a first aspect, the present application provides a MEMS sensor signal conversion method based on cycle-by-cycle noise reduction, the method comprising:

[0006] In a first sampling period, a first signal sequence of a MEMS sensor acquisition target device is acquired, and noise is filtered according to a preset noise reduction parameter to obtain a first noise-filtered signal sequence and a first noise signal sequence;

[0007] Collecting a first operating parameter sequence of the target device, and predicting a second predicted operating parameter sequence and a second predicted noise signal sequence within a second sampling period based on the first noise signal sequence;

[0008] performing noise reduction parameter optimization within a second sampling period according to the second predicted noise signal sequence to obtain second optimized noise reduction parameters, wherein the optimization parameters are configured according to a deviation between the second predicted operating parameter sequence and the second predicted noise signal sequence;

[0009] According to the second optimized noise reduction parameter, noise reduction control is performed in the second sampling period, and cycle-by-cycle noise reduction signal conversion is continued.

[0010] In a second aspect, the present application provides a MEMS sensor signal conversion system based on cycle-by-cycle noise reduction, comprising:

[0011] A first signal acquisition module is configured to acquire a first signal sequence from a MEMS sensor acquisition target device within a first sampling period, and perform noise filtering according to a preset noise reduction parameter to obtain a first noise-filtered signal sequence and a first noise signal sequence;

[0012] a second signal prediction module, configured to collect a first operating parameter sequence of the target device, and predict a second predicted operating parameter sequence and a second predicted noise signal sequence within a second sampling period based on the first noise signal sequence;

[0013] an optimization parameter acquisition module, configured to optimize the noise reduction parameters within a second sampling period according to the second predicted noise signal sequence to obtain second optimized noise reduction parameters, wherein the optimization parameters are configured according to a deviation between the second predicted operating parameter sequence and the second predicted noise signal sequence;

[0014] The noise reduction control module is configured to perform noise reduction control within a second sampling period according to the second optimized noise reduction parameter, and continue to perform cycle-by-cycle noise reduction signal conversion.

[0015] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0016] This application proposes a MEMS sensor signal conversion method and system based on cycle-by-cycle noise reduction. Through a closed-loop optimization mechanism that dynamically couples device operating status with noise prediction, the noise reduction effect under dynamic conditions is significantly improved. Compared with traditional methods, the technical solution provided by this application significantly overcomes the static limitations and response lag of fixed-parameter noise reduction, achieving the technical effect of low-latency, low-cost, and highly robust noise suppression in dynamic environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0018] Figure 1 A flow chart of a MEMS sensor signal conversion method based on cycle-by-cycle noise reduction provided in an embodiment of the present application.

[0019] Figure 2 This is a schematic diagram of the structure of a MEMS sensor signal conversion system based on cycle-by-cycle noise reduction provided in an embodiment of the present application.

[0020] In the accompanying drawings, the components represented by the reference numerals are described as follows:

[0021] A first signal acquisition module 100 , a second signal prediction module 200 , an optimization parameter acquisition module 300 , and a noise reduction control module 400 . DETAILED DESCRIPTION

[0022] The present application provides a MEMS sensor signal conversion method and system based on cycle-by-cycle noise reduction, which is used to solve the technical problem of poor signal noise reduction effect in the prior art.

[0023] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0024] It should be noted that the terms "including" and "having" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or are inherent to these processes, methods, products or devices.

[0025] Example 1, as Figure 1 As shown, the present application provides a MEMS sensor signal conversion method based on cycle-by-cycle noise reduction, wherein the method includes:

[0026] S10: In a first sampling period, a first signal sequence of a MEMS sensor acquisition target device is acquired, and noise is filtered according to a preset noise reduction parameter to obtain a first noise-filtered signal sequence and a first noise signal sequence.

[0027] In dynamic working conditions, traditional methods directly use fixed parameter noise reduction due to the lack of a reliable noise benchmark, resulting in the preset parameters being unable to adapt to the actual noise characteristics of the current equipment. Insufficient initial noise reduction may result in residual strong interference or excessive suppression, and the noise signal is not effectively separated from the useful signal, making it difficult to build a data foundation for subsequent optimization.

[0028] Step S10 in the method provided in the embodiment of the present application includes:

[0029] Acquire a first signal sequence of a MEMS sensor acquisition target device within a first sampling period;

[0030] Noise filtering is performed on the first signal sequence to obtain a first noise-filtered signal sequence and a first noise signal sequence.

[0031] In an embodiment of the present application, a MEMS sensor is used to collect a first signal sequence of a target device within a first sampling period. Exemplarily, the first sampling period is set to 60 minutes, and the first signal sequence of the target device is collected using a MEMS sensor within 30 minutes. The target device refers to a device that requires noise reduction optimization, such as a generator that generates noise during operation. MEMS sensors are functional devices that implement microelectromechanical system perception and signal processing, and have various structures and functions. Optionally, a MEMS sensor that can collect vibration signals is used to collect the vibration signals of the target device, which are integrated in chronological order to obtain a first signal sequence.

[0032] The first signal sequence is subjected to noise filtering to obtain a first noise-filtered signal sequence and a first noise signal sequence. Specifically, the first signal sequence is filtered using a filter. For example, the noise filtering parameter of the filter is set to 110% of the fundamental frequency of normal operation of the device to filter out high-frequency noise generated during device operation. The fundamental frequency of normal operation of the device can be obtained from specific design parameters of the device. For example, when the fundamental frequency of normal operation of the device is 1000 Hz, the noise filtering parameter is set to 1100 Hz, and the filtered first noise-filtered signal sequence and the filtered first noise signal sequence are output.

[0033] Initial noise filtering is completed by pre-setting noise reduction parameters, and the first noise signal sequence is simultaneously separated, providing accurate input for subsequent dynamic optimization. Initial noise filtering can ensure signal availability, and the separated noise sequence accurately characterizes the interference characteristics of the current working conditions, establishing a reliable benchmark for noise evolution prediction.

[0034] S20: Collect a first operating parameter sequence of the target device, and predict a second predicted operating parameter sequence and a second predicted noise signal sequence within a second sampling period according to the first noise signal sequence.

[0035] Existing technologies ignore the strong coupling relationship between equipment operating parameters and noise, and only rely on static predictions of historical noise data. Noise predictions that are not associated with changes in operating conditions may lag behind the actual evolution of interference, and sudden changes in noise characteristics caused by sudden operating conditions cannot be predicted, resulting in inaccurate configuration of noise reduction parameters for the next cycle.

[0036] Step S20 in the method provided in the embodiment of the present application includes:

[0037] Collecting a first operating parameter sequence of the target device within a first sampling period;

[0038] Inputting the first operating parameter sequence and the first noise signal sequence into a noise signal predictor, and predicting and outputting a second predicted operating parameter sequence and a second predicted noise signal sequence within a second sampling period;

[0039] The training step of the noise signal predictor includes:

[0040] According to the operation monitoring data of the same device, a sample first operation parameter sequence set and a sample first noise signal sequence set are collected, and a sample second operation parameter sequence set and a sample second noise signal sequence set in the next cycle are collected;

[0041] Build a noise signal predictor based on machine learning;

[0042] The sample first operating parameter sequence set, the sample first noise signal sequence set, the sample second operating parameter sequence set and the sample second noise signal sequence set are used to perform supervised training to adjust the parameters of the noise signal predictor until the test converges and the training is completed.

[0043] In an embodiment of the present application, a first operating parameter sequence of the target device within a first sampling period is collected, which may include operating parameters such as operating power (obtained by extracting the operating log of the target device) and operating temperature (obtained through a temperature sensor).

[0044] According to the operation monitoring data of the same device, a sample first operation parameter sequence set and a sample first noise signal sequence set are collected, and a sample second operation parameter sequence set and a sample second noise signal sequence set in the next cycle are collected. Wherein, one cycle is exemplarily set to 120 minutes. Wherein, the operation parameter sequence optionally adopts the operation power and operation temperature, and the noise signal sequence is obtained by collecting the signal sequence and filtering the noise using a filter. The operation parameter sequence is integrated into an operation parameter sequence set, and the noise signal parameter is integrated into a noise signal sequence set to obtain the sample first operation parameter sequence set, the sample first noise signal sequence set, the sample second operation parameter sequence set and the sample second noise signal sequence set.

[0045] Based on machine learning, a noise signal predictor is constructed. Preferably, a three-layer structure is used for construction, wherein the input layer is used to receive the first operating parameter sequence and the first noise signal sequence, the hidden layer contains 64 nodes and is activated using the ReLU function, and the output layer outputs the predicted second operating parameter sequence and the second noise signal sequence. Using a sample set of the first operating parameter sequence, a sample set of the first noise signal sequence, a sample set of the second operating parameter sequence, and a sample set of the second noise signal sequence, supervised training is performed to adjust the parameters of the noise signal predictor until the test converges. For example, if the error between the input operating parameter sequence, the output second predicted operating parameter sequence, and the second predicted noise signal is within a range of ±2%, the noise signal predictor training is completed.

[0046] The first operating parameter sequence and the first noise signal sequence are input into the noise signal predictor, and the prediction output obtains the second predicted operating parameter sequence and the second predicted noise signal sequence within the second sampling period.

[0047] By collaboratively predicting the equipment operating parameters and noise sequences, a dynamic correlation model of operating conditions and noise is constructed, in which the operating parameter sequence reflects the equipment state migration trend in real time, so that the noise prediction fits the actual operating conditions; the coupled prediction based on the current noise characteristics and operating conditions can significantly improve the foresight of sudden interference, providing a core basis for the forward-looking optimization of noise reduction parameters.

[0048] S30: Optimizing noise reduction parameters within a second sampling period according to the second predicted noise signal sequence to obtain second optimized noise reduction parameters, wherein the optimization parameters are configured according to a deviation between the second predicted operating parameter sequence and the second predicted noise signal sequence.

[0049] Traditional noise reduction optimization methods have limitations due to the lack of correlation between predicted working conditions and noise deviations: single-dimensional noise suppression can easily cause equipment operation problems, and static optimization objectives cannot adapt to the balance requirements between noise suppression and equipment operation in dynamic scenarios.

[0050] Step S30 in the method provided in the embodiment of the present application includes:

[0051] Randomly configure noise reduction parameters;

[0052] Calculating the operating parameter deviation between the second predicted operating parameter sequence and the first operating parameter sequence;

[0053] Calculating a noise signal deviation between the second predicted noise signal sequence and the first noise signal sequence;

[0054] Calculating a control deviation between the noise reduction parameter and a preset noise reduction parameter as a transient distortion degree;

[0055] Calculating the cost fitness according to the operating parameter deviation, the noise signal deviation and the transient distortion;

[0056] The cost fitness is calculated based on the operating parameter deviation, the noise signal deviation, and the transient distortion, including:

[0057] Calculating the sum of the operating parameter deviation and the noise signal deviation;

[0058] Calculating the similarity between the sum of the operating parameter deviation and the noise signal deviation and the transient distortion to obtain the cost fitness;

[0059] Analyzing and obtaining noise reduction fitness according to the second predicted noise signal sequence and noise reduction parameters;

[0060] The step of analyzing and obtaining noise reduction fitness based on the second predicted noise signal sequence and the noise reduction parameters includes:

[0061] Obtaining a noise reduction analyzer, wherein the noise reduction analyzer is trained using a sample noise signal sequence set, a sample noise reduction parameter set, and a labeled sample noise reduction fitness set, wherein each sample noise reduction fitness includes a ratio of a signal-to-noise ratio after noise reduction to a preset signal-to-noise ratio;

[0062] Inputting the second predicted noise signal sequence and the noise reduction parameters into the noise reduction analyzer, and outputting the obtained noise reduction fitness;

[0063] Configuring a noise reduction weight and a cost weight according to the operating parameter deviation and the noise signal deviation, and performing weighted calculation on the noise reduction fitness and the cost fitness to obtain the noise reduction fitness;

[0064] The noise reduction weight and the cost weight are configured according to the operating parameter deviation and the noise signal deviation, and the noise reduction fitness and the cost fitness are weightedly calculated to obtain the noise reduction fitness, including:

[0065] Calculating the cost weight according to the operating parameter deviation and the noise signal deviation, and obtaining the noise reduction weight;

[0066] The cost weight is calculated according to the operating parameter deviation and the noise signal deviation, and the noise reduction weight is obtained by calculation, including:

[0067] Calculating the ratio of the operating parameter deviation, the noise signal deviation, and the maximum value of the historical operating parameter deviation and the maximum value of the noise signal deviation, and calculating the mean to obtain the deviation coefficient as the cost weight;

[0068] Calculating a noise reduction weight according to the cost weight;

[0069] Using the noise reduction weight and the cost weight, weighted calculation is performed on the noise reduction fitness and the cost fitness to obtain the noise reduction fitness;

[0070] Iteratively optimize the noise reduction parameters to obtain converged second optimized noise reduction parameters, wherein the second optimized noise reduction parameters have the maximum noise reduction fitness during the optimization.

[0071] In an embodiment of the present application, a random number generator is used to randomly generate and configure noise reduction parameters. For example, the noise reduction parameters generated are frequency-based noise reduction, and the parameters are 108% of the basic frequency of normal operation of the device. For example, if the basic frequency of normal operation of the device is 1000Hz, the noise reduction parameter is 1080Hz.

[0072] Calculate the arithmetic mean of the operating parameters of the second predicted operating parameter sequence and the first operating parameter sequence, and calculate the degree of deviation based on the arithmetic mean of the operating parameters of the second predicted operating parameter sequence and the first operating parameter sequence. Specifically, the degree of deviation of the operating parameter = (arithmetic mean of the second predicted operating parameter sequence - arithmetic mean of the first operating parameter sequence) ÷ the arithmetic mean of the first operating parameter sequence, expressed in percentages. For example, if the arithmetic mean of the operating power in the second predicted operating parameter sequence is 120W and the arithmetic mean of the operating power in the first operating parameter sequence is 150W, then the degree of deviation of the operating power in the operating parameter = (120-150) ÷ 120 = 25%. The degree of deviation of the operating parameter is equal to the arithmetic mean of the degree of deviation of multiple operating parameters in the operating parameter. For example, the degree of deviation of the operating parameter = (the degree of deviation of the operating power in the operating parameter + the degree of deviation of the operating temperature in the operating parameter) ÷ 2. The degree of deviation of the operating parameter is a value that indicates the accuracy of the predicted operating parameter. The smaller the degree of deviation of the operating parameter, the higher the accuracy and credibility of the second predicted operating parameter sequence.

[0073] Calculate the arithmetic mean of the noise signals of the second predicted noise signal sequence and the first noise signal sequence, and then calculate the deviation based on the arithmetic mean of the noise signals of the second predicted noise signal sequence and the first noise signal sequence. Specifically, the noise signal deviation = (arithmetic mean of the second predicted noise signal sequence - arithmetic mean of the first noise signal sequence) ÷ arithmetic mean of the first noise signal sequence, expressed in percentages. For example, if the arithmetic mean of the second predicted noise signal sequence is 1100 Hz and the arithmetic mean of the first noise signal sequence is 1200 Hz, then the noise signal deviation = (1100 - 1200) ÷ 1200 = 8.3%. Similarly, a smaller noise deviation indicates a higher accuracy of the predicted noise signal sequence.

[0074] Calculate the control deviation between the noise reduction parameter and the preset noise reduction parameter as transient distortion. Transient distortion = (Noise reduction parameter - Preset noise reduction parameter) ÷ Preset noise reduction parameter, expressed in percentages. For example, if the noise reduction parameter is 1080Hz and the preset noise reduction parameter is 1100Hz, then transient distortion = (1080-1100) ÷ 1100 = 1.8%. Transient distortion reflects the deviation between the noise reduction parameter and the preset noise reduction parameter. A higher transient distortion indicates that the noise reduction parameter is deviating and is more likely to be inconsistent with the current operating conditions.

[0075] The cost fitness is calculated based on the operating parameter deviation, noise signal deviation and transient distortion.

[0076] Calculate the sum of the operating parameter deviation and the noise signal deviation.

[0077] Calculate the similarity between the sum of the operating parameter deviation and the noise signal deviation and the transient distortion to obtain the cost fitness. Optionally, the cost fitness = 1 ÷ |the sum of the operating parameter deviation and the noise signal deviation - the transient distortion|. When the sum of the operating parameter deviation and the noise signal deviation - the transient distortion = 0, the cost fitness is 1. A higher cost fitness reduces the cost of adjusting the noise reduction parameters for transient distortion, and thus allows for greater adaptability to the target device and its current operating conditions while minimizing the cost of operating parameter changes.

[0078] Obtain a noise reduction analyzer. For example, machine learning is used to construct the noise reduction analyzer, for example, a three-layer structure is adopted, the input layer receives the noise signal sequence and the noise reduction parameters, the hidden layer adopts 16 nodes, the ReLU function is used for activation, and the output layer outputs the noise reduction fitness of the analysis.

[0079] The noise reduction analyzer is trained using a set of sample noise signal sequences, a set of sample noise reduction parameters, and a set of annotated sample noise reduction fitness values. Each sample's noise reduction fitness value consists of the ratio of the signal-to-noise ratio after noise reduction to the preset signal-to-noise ratio. For example, the preset signal-to-noise ratio can be set to 20dB. The noise reduction fitness value is calculated as the signal-to-noise ratio after noise reduction divided by the preset signal-to-noise ratio. A higher noise reduction fitness value indicates a higher signal-to-noise ratio after noise reduction, a stronger signal relative to the noise, and better noise reduction results.

[0080] The training is completed until the noise reduction analyzer converges, for example, when the noise signal sequence, noise reduction parameters, and the output noise reduction fitness accuracy are above 90%.

[0081] The second predicted noise signal sequence and the noise reduction parameters are input into the noise reduction analyzer, and the noise reduction fitness is obtained as an output.

[0082] Calculate the ratios of the operating parameter deviation, noise signal deviation, and the historical maximum operating parameter deviation and noise signal deviation, and average these to obtain the deviation coefficient, which serves as the cost weight. Cost weight = (operating parameter deviation ÷ historical maximum operating parameter deviation + noise deviation ÷ historical maximum noise deviation) ÷ 2. The cost weight is the weight of the cost fitness. A larger cost weight indicates a higher requirement for operating parameter stability and a greater sensitivity to noise reduction optimization at the expense of operating parameters, thus assigning a greater weight to the cost fitness.

[0083] The noise reduction weight is calculated based on the cost weight: Noise reduction weight = 1 - cost weight.

[0084] Using the noise reduction weight and the cost weight, the noise reduction fitness and the cost fitness are weighted and calculated to obtain the noise reduction fitness. Noise reduction fitness = noise reduction fitness × noise reduction weight + cost fitness × cost weight.

[0085] Repeatedly randomly configure the noise reduction parameters and iteratively optimize the noise reduction parameters to obtain the converged second optimized noise reduction parameters, where the second optimized noise reduction parameters have the highest noise reduction fitness during the optimization. For example, the convergence condition is set to be that if no better noise reduction fitness is obtained after 20 iterations, the noise reduction parameters corresponding to the current best noise reduction fitness are output as the second optimized noise reduction parameters.

[0086] The deviation between the predicted operating parameters and the noise sequence is introduced as the basis for optimizing parameter configuration to achieve multi-objective dynamic balance. By integrating multi-dimensional parameters such as noise deviation, operating condition deviation and transient distortion, the noise reduction parameters are adaptively tuned in real time between the interference suppression intensity and the equipment operating status, ensuring the optimal coordination between the noise reduction effect and equipment operation under complex working conditions.

[0087] S40: performing noise reduction control in a second sampling period according to the second optimized noise reduction parameter, and continuing to perform cycle-by-cycle noise reduction signal conversion.

[0088] In the embodiment of the present application, according to the second optimized noise reduction parameter, the noise reduction signal conversion is performed in the second sampling period, and the noise reduction signal conversion is continued period by period, for example, in the third sampling period, the noise reduction signal conversion is performed according to the same method.

[0089] By executing the prediction-optimization-control process cycle by cycle, a continuously adaptive dynamic noise reduction system is built. The noise reduction parameters are updated in each cycle based on the latest working conditions and noise predictions, so that the noise reduction process always fits the real-time status of the equipment. At the same time, the closed-loop iterative mechanism can accumulate errors.

[0090] Example 2, as Figure 2 As shown, based on the same inventive concept as the MEMS sensor signal conversion method based on cycle-by-cycle noise reduction provided in the first embodiment, an embodiment of the present invention further provides a MEMS sensor signal conversion system based on cycle-by-cycle noise reduction, including:

[0091] The first signal acquisition module 100 is configured to acquire a first signal sequence from a MEMS sensor acquisition target device within a first sampling period, and perform noise filtering according to preset noise reduction parameters to obtain a first noise-filtered signal sequence and a first noise signal sequence;

[0092] The second signal prediction module 200 is configured to collect a first operating parameter sequence of the target device and predict a second predicted operating parameter sequence and a second predicted noise signal sequence within a second sampling period based on the first noise signal sequence;

[0093] an optimization parameter acquisition module 300, configured to optimize noise reduction parameters within a second sampling period based on the second predicted noise signal sequence to obtain second optimized noise reduction parameters, wherein the optimization parameters are configured based on a deviation between the second predicted operating parameter sequence and the second predicted noise signal sequence;

[0094] The noise reduction control module 400 is configured to perform noise reduction control within a second sampling period according to the second optimized noise reduction parameter, and continue to perform cycle-by-cycle noise reduction signal conversion.

[0095] In one embodiment, the first signal acquisition module 100 is further configured to:

[0096] Acquire a first signal sequence of a MEMS sensor acquisition target device within a first sampling period;

[0097] Noise filtering is performed on the first signal sequence to obtain a first noise-filtered signal sequence and a first noise signal sequence.

[0098] In one embodiment, the second signal prediction module 200 is further configured to:

[0099] Collecting a first operating parameter sequence of the target device within a first sampling period;

[0100] Inputting the first operating parameter sequence and the first noise signal sequence into a noise signal predictor, and predicting and outputting a second predicted operating parameter sequence and a second predicted noise signal sequence within a second sampling period;

[0101] The training step of the noise signal predictor includes:

[0102] According to the operation monitoring data of the same device, a sample first operation parameter sequence set and a sample first noise signal sequence set are collected, and a sample second operation parameter sequence set and a sample second noise signal sequence set in the next cycle are collected;

[0103] Build a noise signal predictor based on machine learning;

[0104] The sample first operating parameter sequence set, the sample first noise signal sequence set, the sample second operating parameter sequence set and the sample second noise signal sequence set are used to perform supervised training to adjust the parameters of the noise signal predictor until the test converges and the training is completed.

[0105] In one embodiment, the optimization parameter acquisition module 300 is further configured to:

[0106] Randomly configure noise reduction parameters;

[0107] Calculating the operating parameter deviation between the second predicted operating parameter sequence and the first operating parameter sequence;

[0108] Calculating a noise signal deviation between the second predicted noise signal sequence and the first noise signal sequence;

[0109] Calculating a control deviation between the noise reduction parameter and a preset noise reduction parameter as a transient distortion degree;

[0110] Calculating the cost fitness according to the operating parameter deviation, the noise signal deviation and the transient distortion;

[0111] The cost fitness is calculated based on the operating parameter deviation, the noise signal deviation, and the transient distortion, including:

[0112] Calculating the sum of the operating parameter deviation and the noise signal deviation;

[0113] Calculating the similarity between the sum of the operating parameter deviation and the noise signal deviation and the transient distortion to obtain the cost fitness;

[0114] Analyzing and obtaining noise reduction fitness according to the second predicted noise signal sequence and noise reduction parameters;

[0115] The step of analyzing and obtaining noise reduction fitness based on the second predicted noise signal sequence and the noise reduction parameters includes:

[0116] Obtaining a noise reduction analyzer, wherein the noise reduction analyzer is trained using a sample noise signal sequence set, a sample noise reduction parameter set, and a labeled sample noise reduction fitness set, wherein each sample noise reduction fitness includes a ratio of a signal-to-noise ratio after noise reduction to a preset signal-to-noise ratio;

[0117] Inputting the second predicted noise signal sequence and the noise reduction parameters into the noise reduction analyzer, and outputting the obtained noise reduction fitness;

[0118] Configuring a noise reduction weight and a cost weight according to the operating parameter deviation and the noise signal deviation, and performing weighted calculation on the noise reduction fitness and the cost fitness to obtain the noise reduction fitness;

[0119] The noise reduction weight and the cost weight are configured according to the operating parameter deviation and the noise signal deviation, and the noise reduction fitness and the cost fitness are weightedly calculated to obtain the noise reduction fitness, including:

[0120] Calculating the cost weight according to the operating parameter deviation and the noise signal deviation, and obtaining the noise reduction weight;

[0121] The cost weight is calculated according to the operating parameter deviation and the noise signal deviation, and the noise reduction weight is obtained by calculation, including:

[0122] Calculating the ratio of the operating parameter deviation, the noise signal deviation, and the maximum value of the historical operating parameter deviation and the maximum value of the noise signal deviation, and calculating the mean to obtain the deviation coefficient as the cost weight;

[0123] Calculating a noise reduction weight according to the cost weight;

[0124] Using the noise reduction weight and the cost weight, weighted calculation is performed on the noise reduction fitness and the cost fitness to obtain the noise reduction fitness;

[0125] Iteratively optimize the noise reduction parameters to obtain converged second optimized noise reduction parameters, wherein the second optimized noise reduction parameters have the maximum noise reduction fitness during the optimization.

[0126] In summary, the embodiments of the present application have at least the following technical effects:

[0127] This application proposes a MEMS sensor signal conversion method and system based on cycle-by-cycle noise reduction. Through a closed-loop optimization mechanism that dynamically couples the device operating state and noise prediction, the noise reduction effect under dynamic working conditions is significantly improved. Specifically, by collecting the device operating parameters cycle by cycle and performing collaborative prediction in conjunction with the noise signal, a dynamic correlation between the working condition and noise is constructed, so that the optimization of the noise reduction parameters is always synchronized with the evolution of the actual state of the device, avoiding insufficient noise reduction or excessive signal suppression due to sudden changes in the working condition; secondly, an online optimization mechanism for noise reduction parameters based on the predicted noise sequence is introduced to weigh the noise suppression intensity and the device operating state in real time, and maintain a stable signal-to-noise ratio in a complex interference environment; finally, the noise reduction parameters are intelligently evaluated by the noise reduction analyzer, and the dynamic ratio of the cost weight and the noise reduction weight is combined to ensure that the optimization direction always fits the core requirements of the current working condition. Compared with traditional methods, the technical solution provided by this application significantly overcomes the static limitations and response lag defects of fixed noise reduction parameters, and achieves the technical effect of low latency, low cost, and high robustness noise suppression in a dynamic environment.

[0128] It should be noted that the order in which the embodiments of the present application are presented is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. Furthermore, the foregoing descriptions of specific embodiments of this specification are provided. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order or sequential sequence shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0129] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application.

[0130] This specification and drawings are merely illustrative of the present application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Obviously, those skilled in the art may make various modifications and variations to this application without departing from the scope of this application. Thus, this application is intended to include such modifications and variations as fall within the scope of this application and its equivalents.

Claims

1. A MEMS sensor signal conversion method based on cycle-by-cycle noise reduction, characterized in that: The method comprises: In a first sampling period, a first signal sequence of a MEMS sensor acquisition target device is acquired, and noise is filtered according to a preset noise reduction parameter to obtain a first noise-filtered signal sequence and a first noise signal sequence; Collecting a first operating parameter sequence of the target device, and predicting a second predicted operating parameter sequence and a second predicted noise signal sequence within a second sampling period based on the first noise signal sequence; performing noise reduction parameter optimization within a second sampling period according to the second predicted noise signal sequence to obtain second optimized noise reduction parameters, wherein the optimization parameters are configured according to a deviation between the second predicted operating parameter sequence and the second predicted noise signal sequence; According to the second optimized noise reduction parameter, noise reduction control is performed in the second sampling period, and cycle-by-cycle noise reduction signal conversion is continued.

2. The MEMS sensor signal conversion method based on cycle-by-cycle noise reduction according to claim 1, characterized in that: In a first sampling period, a first signal sequence of a MEMS sensor acquisition target device is acquired, and noise is filtered to obtain a first noise-filtered signal sequence and a first noise signal sequence, including: Acquire a first signal sequence of a MEMS sensor acquisition target device within a first sampling period; Noise filtering is performed on the first signal sequence to obtain a first noise-filtered signal sequence and a first noise signal sequence.

3. The MEMS sensor signal conversion method based on cycle-by-cycle noise reduction according to claim 1, characterized in that: Acquiring a first operating parameter sequence of the target device and predicting a second predicted operating parameter sequence and a second predicted noise signal sequence within a second sampling period based on the first noise signal sequence includes: Collecting a first operating parameter sequence of the target device within a first sampling period; The first operating parameter sequence and the first noise signal sequence are input into a noise signal predictor, and a second predicted operating parameter sequence and a second predicted noise signal sequence within a second sampling period are obtained through prediction output.

4. The MEMS sensor signal conversion method based on cycle-by-cycle noise reduction according to claim 3, characterized in that: The training step of the noise signal predictor comprises: According to the operation monitoring data of the same device, a sample first operation parameter sequence set and a sample first noise signal sequence set are collected, and a sample second operation parameter sequence set and a sample second noise signal sequence set in the next cycle are collected; Build a noise signal predictor based on machine learning; The sample first operating parameter sequence set, the sample first noise signal sequence set, the sample second operating parameter sequence set and the sample second noise signal sequence set are used to perform supervised training to adjust the parameters of the noise signal predictor until the test converges and the training is completed.

5. The MEMS sensor signal conversion method based on cycle-by-cycle noise reduction according to claim 1, characterized in that: Optimizing noise reduction parameters within a second sampling period according to the second predicted noise signal sequence to obtain second optimized noise reduction parameters includes: Randomly configure noise reduction parameters; Calculating the operating parameter deviation between the second predicted operating parameter sequence and the first operating parameter sequence; Calculating a noise signal deviation between the second predicted noise signal sequence and the first noise signal sequence; Calculating a control deviation between the noise reduction parameter and a preset noise reduction parameter as a transient distortion degree; Calculating the cost fitness according to the operating parameter deviation, the noise signal deviation and the transient distortion; Analyzing and obtaining noise reduction fitness according to the second predicted noise signal sequence and noise reduction parameters; Configuring a noise reduction weight and a cost weight according to the operating parameter deviation and the noise signal deviation, and performing weighted calculation on the noise reduction fitness and the cost fitness to obtain the noise reduction fitness; Iteratively optimize the noise reduction parameters to obtain converged second optimized noise reduction parameters, wherein the second optimized noise reduction parameters have the maximum noise reduction fitness during the optimization.

6. The MEMS sensor signal conversion method based on cycle-by-cycle noise reduction according to claim 5, characterized in that: Calculating the cost fitness based on the operating parameter deviation, the noise signal deviation, and the transient distortion includes: Calculating the sum of the operating parameter deviation and the noise signal deviation; The similarity between the sum of the operating parameter deviation and the noise signal deviation and the transient distortion is calculated to obtain the cost fitness.

7. The MEMS sensor signal conversion method based on cycle-by-cycle noise reduction according to claim 5, characterized in that: Analyzing and obtaining noise reduction fitness according to the second predicted noise signal sequence and the noise reduction parameters includes: Obtaining a noise reduction analyzer, wherein the noise reduction analyzer is trained using a sample noise signal sequence set, a sample noise reduction parameter set, and a labeled sample noise reduction fitness set, wherein each sample noise reduction fitness includes a ratio of a signal-to-noise ratio after noise reduction to a preset signal-to-noise ratio; The second predicted noise signal sequence and the noise reduction parameters are input into the noise reduction analyzer, and the noise reduction fitness is obtained as an output.

8. The MEMS sensor signal conversion method based on cycle-by-cycle noise reduction according to claim 5, characterized in that: Configuring a noise reduction weight and a cost weight according to the operating parameter deviation and the noise signal deviation, and performing weighted calculation on the noise reduction fitness and the cost fitness to obtain the noise reduction fitness, including: Calculating the cost weight according to the operating parameter deviation and the noise signal deviation, and obtaining the noise reduction weight; The noise reduction weight and the cost weight are used to perform weighted calculation on the noise reduction fitness and the cost fitness to obtain the noise reduction fitness.

9. The MEMS sensor signal conversion method based on cycle-by-cycle noise reduction according to claim 8, characterized in that: Calculating the cost weight according to the operating parameter deviation and the noise signal deviation, and calculating the noise reduction weight, including: Calculating the ratio of the operating parameter deviation, the noise signal deviation, and the maximum value of the historical operating parameter deviation and the maximum value of the noise signal deviation, and calculating the mean to obtain the deviation coefficient as the cost weight; The noise reduction weight is calculated according to the cost weight.

10. A MEMS sensor signal conversion system based on cycle-by-cycle noise reduction, characterized in that: A system for implementing the MEMS sensor signal conversion method based on cycle-by-cycle noise reduction according to any one of claims 1 to 9, comprising: A first signal acquisition module is configured to acquire a first signal sequence from a MEMS sensor acquisition target device within a first sampling period, and perform noise filtering according to a preset noise reduction parameter to obtain a first noise-filtered signal sequence and a first noise signal sequence; a second signal prediction module, configured to collect a first operating parameter sequence of the target device, and predict a second predicted operating parameter sequence and a second predicted noise signal sequence within a second sampling period based on the first noise signal sequence; an optimization parameter acquisition module, configured to optimize the noise reduction parameters within a second sampling period according to the second predicted noise signal sequence to obtain second optimized noise reduction parameters, wherein the optimization parameters are configured according to a deviation between the second predicted operating parameter sequence and the second predicted noise signal sequence; The noise reduction control module is configured to perform noise reduction control within a second sampling period according to the second optimized noise reduction parameter, and continue to perform cycle-by-cycle noise reduction signal conversion.

Images