Noise processing method, device, electronic device, and storage medium

Abstract

The present disclosure provides a noise processing method, device, electronic device, and storage medium. [Solution] The method includes obtaining a first noise signal at the location of a target staff member in a target workplace within a first period; predicting overall control parameters of a plurality of signal emitters within a second period, which is a future period of the first period, based on the first noise signal, and obtaining a current parameter prediction result; determining one or more target emitters that need to operate within the second period from the plurality of signal emitters based on the current parameter prediction result, and obtaining a parameter prediction value for each target emitter within the second period; and controlling each target emitter to emit a noise interference signal within the second period according to the corresponding parameter prediction value so as to weaken the second noise signal at the location of the target staff member in the target workplace within the second period.

Description

[Technical field]

[0001] The present disclosure relates to the field of data processing, and in particular to a noise processing method, device, electronic device, and storage medium. [Background technology]

[0002] In the industrial environment of the spinning process, the complexity and length of the process flow means that it is necessary to set up multiple production workshops with different functions, and in addition, a huge number of process equipments are installed in each production workshop, which generates significant noise pollution when they are operating. Summary of the Invention [Problem to be solved by the invention]

[0003] The present disclosure provides a noise processing method, device, electronic device, and storage medium to solve or alleviate one or more technical problems in the prior art. [Means for solving the problem]

[0004] In a first aspect, the present disclosure provides a noise processing method applied to an electronic device, the electronic device being in communication with a plurality of signal generators arranged in a target workplace, the arrangement of the plurality of signal generators in the target workplace being correlated to a real-time noise sound field in the target workplace, the method comprising: Obtaining a first noise signal of a target staff member's location in a target workplace during a first time period; Predicting overall control parameters of the multiple signal generators in a second time period, which is a future time period of the first time period, based on the first noise signal, to obtain a current parameter prediction result; determining one or more target transmitters that need to be operated in a second time period from the plurality of signal transmitters based on the current parameter prediction result, and obtaining a parameter prediction value in the second time period for each target transmitter; and controlling each target emitter to emit a noise interference signal within the second time period according to the corresponding parameter prediction value to attenuate a second noise signal at the target staff's location in the target workplace within the second time period.

[0005] In a second aspect, the present disclosure provides a noise processing device applied to an electronic device, the electronic device communicates with a plurality of signal generators arranged in a target workplace, and an arrangement manner of the plurality of signal generators in the target workplace has a corresponding relationship with a real-time noise sound field in the target workplace, and the device includes: A first acquisition unit for acquiring a first noise signal of a target staff member's location in a target workplace within a first time period; A second obtaining unit for predicting overall control parameters of the plurality of signal generators in a second time period, which is a future time period of the first time period, based on the first noise signal, and obtaining a current parameter prediction result; a third obtaining unit for determining one or more target transmitters that need to be operated within a second period from the plurality of signal transmitters according to the current parameter prediction result, and obtaining a parameter prediction value of each target transmitter within the second period; and an emission control unit for causing each target emitter to emit a noise interference signal within the second time period according to the corresponding parameter prediction value to attenuate a second noise signal at the target staff's location in the target workplace within the second time period.

[0006] A third aspect provides an electronic device, the electronic device comprising: At least one processor; a memory in communication with the at least one processor; The memory stores instructions executable by the at least one processor, which, when executed by the at least one processor, cause the at least one processor to perform any of the methods of the embodiments of the present disclosure.

[0007] A fourth aspect provides a non-transitory computer-readable storage medium having stored thereon computer instructions for causing a computer to perform any of the methods according to the embodiments of the present disclosure.

[0008] A fifth aspect provides a computer program product including a computer program for, when executed by a processor, performing any of the methods according to the embodiments of the present disclosure. Effect of the Invention

[0009] According to the invention disclosed herein, after obtaining the first noise signal of the location of the target staff in the target workplace in the first period, the overall control parameters of the multiple signal generators in the second period (future period of the first period) can be predicted based on the first noise signal to obtain the current parameter prediction result. The current parameter prediction result is obtained by prediction based on the first noise signal, and is not obtained by directly analyzing the first noise signal, aiming to establish a strong correlation with the second period. Therefore, based on the current parameter prediction result, one or more target generators that need to operate in the second period from the multiple signal generators are determined, and after obtaining the parameter prediction value of each target generator in the second period, each target generator is controlled to emit a noise interference signal in the second period according to the corresponding parameter prediction value, which can generate a good offsetting effect on the second noise signal of the location of the target staff in the target workplace in the second period, and the second noise signal of the location of the target staff in the target workplace in the second period can be weakened, thereby reducing the noise pollution in the target workplace. [Brief description of the drawings]

[0010] [Figure 1] FIG. 1 is a schematic flow chart 1 of a noise processing method according to an embodiment of the present disclosure. [Diagram 2] FIG. 2 is a schematic flow chart 2 of a noise processing method according to an embodiment of the present disclosure. [Diagram 3] FIG. 3 is a schematic diagram of a configuration of a target time series model according to an embodiment of the present disclosure. [Figure 4] FIG. 4 is a diagram for explaining a process of removing an abnormal signal according to an embodiment of the present disclosure. [Diagram 5] FIG. 5 is a diagram for explaining a process of selecting a target point according to an embodiment of the present disclosure. [Figure 6] FIG. 6 is a diagram for explaining a process of adjusting the arrangement of signal transmitters (based on the overhead angle of a target workplace) according to an embodiment of the present disclosure. [Figure 7] FIG. 7 is a schematic configuration block diagram of a noise processing device according to an embodiment of the present disclosure. [Figure 8] FIG. 8 is a schematic configuration block diagram of an electronic device according to an embodiment of the present disclosure. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] It should be understood that the contents described in the Summary of the Invention section do not limit the key points or important features of the embodiments of the present disclosure, and do not limit the scope of the present disclosure. Other features of the present disclosure will be easily understood from the following description.

[0012] In the drawings, unless otherwise specified, the same reference numerals in multiple figures indicate the same or similar parts or elements. The drawings are not necessarily drawn to scale. It should be understood that the drawings merely illustrate some embodiments provided by the present disclosure and are not to be considered as limiting the scope of the present disclosure.

[0013] The present disclosure will now be described in more detail with reference to the drawings, in which like reference numbers indicate functionally identical or similar elements, and in which various aspects of the embodiments are illustrated, but which are not necessarily drawn to scale unless otherwise indicated.

[0014] In addition, in order to better explain the present disclosure, a number of specific details are described in the following specific embodiments. It should be understood by those skilled in the art that the present disclosure can be similarly implemented without specific details. In some instances, methods, means, elements, circuits, etc. that are well known to those skilled in the art are not described in detail, so as to emphasize the gist of the present disclosure.

[0015] As mentioned above, in the industrial environment of the spinning process, the complexity and length of the process flow means that it is necessary to set up multiple production workshops with different functions. Furthermore, a huge number of process equipments are installed in each production workshop, and significant noise pollution occurs when these process equipments operate. For example, in a spinning workshop that produces wound yarn package products, a huge number of spinning boxes, which are process equipments, are installed therein, and significant noise pollution occurs in the spinning workshop when these spinning boxes operate. In addition, for example, in a winding workshop that winds wound yarn package products, a huge number of winders, which are process equipments, are installed therein, and significant noise pollution occurs in the winding workshop when these winders operate.

[0016] In order to reduce noise pollution in a production workshop, an embodiment of the present disclosure provides a noise processing method, which is applied to an electronic device, which communicates with a plurality of signal generators arranged in a target workshop, and the arrangement manner of the plurality of signal generators in the target workshop correlates with a real-time noise sound field in the target workshop, where the target workshop can be any one of a plurality of production workshops with different functions related to a spinning process, and the signal generator can be a speaker.

[0017] In addition, in the embodiment of the present disclosure, the main types of the wound yarn package product may include at least one of partially oriented yarns (POY), fully drawn yarns (FDY), drawn textured yarns (DTY) (also called low elasticity yarns), etc. For example, the types of the wound yarn package product may specifically include polyester partially oriented yarns, polyester fully drawn yarns, polyester drawn yarns, polyester low elasticity yarns, etc.

[0018] Fig. 1 is a schematic flow chart 1 of a noise processing method according to an embodiment of the present disclosure. Hereinafter, the noise processing method according to the embodiment of the present disclosure will be described with reference to Fig. 1. It should be noted that although a logical order is shown in the schematic flow chart, in some cases the steps shown or described may be performed in other orders.

[0019] Step S101: Obtain a first noise signal of a location of a target staff member in a target workplace within a first period.

[0020] Here, the first period may be a current period having a time length of a first predetermined time length. In one example, the first predetermined time length is determined according to the moving speed of the target staff. For example, the first predetermined time length may show a negative correlation with the moving speed of the target staff, i.e., the faster the moving speed of the target staff, the shorter the first predetermined time length, and the slower the moving speed of the target staff, the longer the first predetermined time length.

[0021] In addition, in an embodiment of the present disclosure, the target staff may be a worker entering the target work area, and the first noise signal may be a real-time noise signal felt by the target staff within a first time period. In one example, the first noise signal is collected by a first pickup attached to the target staff and transmitted to an electronic device. The first pickup may be an electromagnetic pickup, a piezoelectric pickup, a fiber optic pickup, a digital pickup, etc.

[0022] Step S102: Predict overall control parameters of the multiple signal generators in a second period according to the first noise signal, and obtain a current parameter prediction result.

[0023] The second period is a future period of the first period. For example, the second period may be a future period of the first period, the time length being a second predetermined time length. In an embodiment of the present disclosure, the second predetermined time length may be the same as the first predetermined time length or may be different from the first predetermined time length. In one example, the second predetermined time length is determined according to the movement speed of the target staff member. For example, the second predetermined time length may show a negative correlation with the movement speed of the target staff member, i.e., the faster the movement speed of the target staff member, the shorter the second predetermined time length, and the slower the movement speed of the target staff member, the longer the second predetermined time length.

[0024] It should be noted that in the embodiment of the present disclosure, the current parameter prediction result is used to control the operating state of each signal generator within the second period.

[0025] Step S103: Based on the current parameter prediction result, determine one or more target transmitters that need to operate within a second period from the multiple signal transmitters, and obtain parameter prediction values ​​within the second period for each target transmitter.

[0026] The current parameter prediction result is used to control the operating state of each signal generator in the second time period, so that for each signal generator, it can determine whether the signal generator is a target generator that needs to operate in the second time period based on the current parameter prediction result, and obtain a parameter prediction value in the second time period for each target generator, where the parameter prediction value may include a frequency prediction value, a phase prediction value, an amplitude prediction value, and a direction prediction value.

[0027] Step S104: Control each target emitter to emit a noise interference signal within the second time period according to the corresponding parameter prediction value, so as to weaken the second noise signal at the target staff's location in the target workplace within the second time period.

[0028] Here, the second noise signal may be a real-time noise signal that is sensed by the target person within the second period.

[0029] When each target transmitter is controlled to emit a noise interference signal within a second period according to a corresponding parameter predicted value, the noise interference signal can have a cancelling effect on a second noise signal at the target staff's location in the target workplace within the second period, thereby weakening the second noise signal at the target staff's location in the target workplace within the second period.

[0030] According to the invention disclosed herein, after obtaining a first noise signal at the location of the target staff in the target workplace in the first period, the overall control parameters of the multiple signal generators in the second period (future period of the first period) can be predicted based on the first noise signal to obtain a current parameter prediction result. The current parameter prediction result is obtained by prediction based on the first noise signal, and is not obtained by directly analyzing the first noise signal, aiming to establish a strong correlation with the second period. Therefore, based on the current parameter prediction result, one or more target generators that need to be activated in the second period from the multiple signal generators are determined, and after obtaining the parameter prediction value of each target generator in the second period, the noise interference signal emitted in the second period according to the corresponding parameter prediction value by controlling each target generator can generate a good offsetting effect on the second noise signal at the location of the target staff in the target workplace in the second period, thereby weakening the second noise signal at the location of the target staff in the target workplace in the second period, thereby reducing the noise pollution in the target workplace.

[0031] Fig. 2 is a schematic flow chart 2 of a noise processing method according to an embodiment of the present disclosure. Hereinafter, the noise processing method according to an embodiment of the present disclosure will be described with reference to Fig. 2. It should be noted that although a logical order is shown in the schematic flow chart, in some cases the steps shown or described may be performed in other orders.

[0032] Step S201: Obtain a first noise signal of a target staff member's location in a target workplace within a first period.

[0033] The first period may be a current period having a time length of a first predetermined time length. In one example, the first predetermined time length is determined according to the moving speed of the target staff. For example, the first predetermined time length may be negatively correlated with the moving speed of the target staff, i.e., the faster the moving speed of the target staff, the shorter the first predetermined time length, and the slower the moving speed of the target staff, the longer the first predetermined time length.

[0034] In addition, in an embodiment of the present disclosure, the target staff may be a worker entering the target work area, and the first noise signal may be a real-time noise signal felt by the target staff within a first time period. In one example, the first noise signal is collected by a first pickup attached to the target staff and transmitted to an electronic device. Here, the first pickup may be an electromagnetic pickup, a piezoelectric pickup, a fiber optic pickup, a digital pickup, etc.

[0035] Step S202: Construct a first feature sequence based on the first noise signal and the historical parameter prediction result.

[0036] Here, the historical parameter prediction result is a prediction of the overall control parameters of the multiple signal emitters in the first period based on the zeroth noise signal of the target staff's location in the target workplace in the zeroth period, and the zeroth period is a historical period of the first period. For example, the zeroth period may be a historical period of the first period with a time length of a third predetermined time length. In the embodiment of the present disclosure, the third predetermined time length may be the same as the first predetermined time length or may be different from the first predetermined time length. In one example, the third predetermined time length is determined according to the moving speed of the target staff. For example, the third predetermined time length may show a negative correlation with the moving speed of the target staff, i.e., the faster the moving speed of the target staff, the shorter the third predetermined time length, and the slower the moving speed of the target staff, the longer the third predetermined time length.

[0037] In addition, in the embodiment of the present disclosure, the historical parameter prediction result is used to control the operation state of each signal generator within a first period. More specifically, for each signal generator, a historical transmission parameter of the signal generator within the first period is obtained based on the historical parameter prediction result.

[0038] The historical transmission parameters may include a frequency value, a phase value, an amplitude value, and a direction value.

[0039] In an alternative embodiment, step S202 may include the following steps.

[0040] Step S202-1: Obtain a time-ordered noise signal sequence based on a first noise signal.

[0041] In one example, the time-ordered noise signal sequence obtained based on the first noise signal is {X 11 ,X 12 ,···,X 1n}, where X 11 is the time T in the first period 11 The noise data of the target staff position in the target workplace is shown, specifically, the frequency value F 11 , phase value P 11 , amplitude value A 11 and directional value D 11 X may also include 12 is the time T in the first period 12 The noise data of the target staff position in the target workplace is shown, specifically, the frequency value F 12 , phase value P 12 , amplitude value A 12 and directional value D 12 Similarly, the following may include X 1n is the time T in the first period 1n The noise data of the target staff position in the target workplace is shown, specifically, the frequency value F 1n , phase value P 1n , amplitude value A 1n and directional value D 1n may include:

[0042] Step S202-2: Based on the historical parameter prediction result, a known parameter sequence arranged in time order is obtained.

[0043] In one example, the number of signal generators in the target workplace is K (K≧2 and is an integer), and the known parameter sequence may include K known parameter subsequences that correspond one-to-one to the K signal generators, and the known parameter subsequence corresponding to the i-th signal generator among the K signal generators is {Qi 11 ,Qi 12 ,···,Qi 1n}, where 1≦i≦K, and i is a positive integer.

[0044] Here, for i=1, Q1 11 is the time T in the first period 11 In the example, the frequency prediction value F1 is a historical parameter prediction value of a first signal generator among the K signal generators. 11 , phase prediction value P1 11 , amplitude prediction value A1 11 and directional prediction value D1 11 Q1 may include 12 is the time T in the first period 12 In the example, the frequency prediction value F1 is a historical parameter prediction value of a first signal generator among the K signal generators. 12 , phase prediction value P1 12 , amplitude prediction value A1 12 and directional prediction value D1 12 Similarly, the following may include Q1 1n is the time T in the first period 1n In the example, the frequency prediction value F1 is a historical parameter prediction value of a first signal generator among the K signal generators. 1n , phase prediction value P1 1n , amplitude prediction value A1 1n and directional prediction value D1 1n It may also include.

[0045] For other values ​​of i, Qi 11 ,Qi 12 ,···,Qi 1n The above-mentioned related content can be referred to for understanding, and detailed description thereof will be omitted here.

[0046] Step S202-3: Splicing the known parameter sequence to the noise signal sequence to obtain a first initial sequence.

[0047] Step S202-4: Obtain a first additional feature based on the real-time task information of the target workplace.

[0048] Here, the real-time task information may include the product type, product specifications, etc. of the wound yarn package product processed at the target workshop. In addition, in the embodiment of the present disclosure, the operating parameters used when the process equipment in the target workshop operates can be obtained based on the real-time task information. If the target workshop is a spinning workshop, the process equipment in the target workshop is a spinning box, and the operating parameters used when the spinning box operates may include the specification parameters of the spinneret in the spinning box, the spinning speed, etc. If the target workshop is a winding workshop, the process equipment in the target workshop is a winding machine, and the operating parameters used when the winding machine operates may include the brand, specification parameters, winding speed, number of winding heads, etc. of the winding machine.

[0049] Based on this, in one example, the real-time task information of the target workplace and the first workplace indicator can both be the first additional feature, or the operating parameters used when the process equipment in the target workplace operates and the first workplace indicator of the target workplace can both be the first additional feature, where the first workplace indicator of the target workplace can be used to determine the processing task type, workplace location, etc. of the target workplace, and the processing task type can include producing a wound yarn package product, winding a wound yarn package product, etc.

[0050] Step S202-5: Construct a first feature sequence based on the first initial sequence and the first additional feature.

[0051] Continuing with the previous example, the noise signal sequence is {X 11 ,X 12 ,···,X 1n}, where the number of signal generators in the target factory is K (K≧2 and is an integer), and the known parameter sequence may include K known parameter subsequences that correspond one-to-one to the K signal generators, and the known parameter subsequence corresponding to the i-th signal generator among the K signal generators is expressed as {Qi 11 ,Qi 12 ,···,Qi 1n}, where 1≦i≦K, and i is a positive integer. When the first additional feature includes the real-time task information of the target workplace and the first distance sign, the first feature sequence constructed based on the first initial sequence obtained by stitching the known parameter sequence to the noise signal sequence and the first additional feature obtained based on the real-time task information of the target workplace may include feature data as shown in Table 1.

[0052] [Table 1]

[0053] Here, Y 11 is the time T in the first period 11 This shows the real-time task information of the target workplace at time T 11 Product type Ty of the wound yarn package product processed in the target workshop 11 , Product Specifications Sp 11 etc., Y 12 indicates the real-time task information of the target workplace at a time in the first period, specifically, at time T 12 Product type Ty of the wound yarn package product processed in the target workshop 12 , Product Specifications Sp1 12 etc., and so on, Y 1n is the time T in the first period 1n This shows the real-time task information of the target workplace at time T 1n Product type Ty of the wound yarn package product processed in the target workshop 1n, Product Specifications Sp 1n etc., where Wid indicates the first workstation marker of the target workstation.

[0054] Thus, in an embodiment of the present disclosure, the first feature sequence, in addition to including a first initial sequence obtained by splicing a known parameter sequence to a noise signal sequence, can also include a first additional feature obtained based on real-time task information of the target workplace, thereby ensuring the integrity of the feature data included in the first feature sequence, thereby increasing the reliability of the current parameter prediction result.

[0055] Step S203: Construct a second feature sequence based on the historical parameter prediction result and the future parameter sequence.

[0056] The prediction result of the history parameters can be understood by referring to the related content above, and a detailed description thereof will be omitted here.

[0057] In addition, in the embodiment of the present disclosure, the future parameter sequence is an input sequence corresponding to the current parameter prediction result, and the sequence length of the future parameter sequence may be the same as or different from the sequence length of the known parameter sequence.

[0058] Continuing with the above example, the number of signal generators in the target workplace is K (K≧2 and is an integer), and the future parameter sequence may similarly include K future parameter subsequences that correspond one-to-one to the K signal generators, and the future parameter subsequence corresponding to the i-th signal generator among the K signal generators is {Zi 21 ,Zi 22 ,···,Zi 2n}, where 1≦i≦K, and i is a positive integer.

[0059] Here, when i=1, Z1 21 is the time T in the second period21 In the above, the data element corresponds to a parameter prediction value of a first signal generator among the K signal generators, specifically, a frequency prediction value F1 21 The data element corresponding to the phase prediction value P1 21 The data element corresponding to the amplitude prediction value A1 21 The data element and the direction prediction value D1 corresponding to 21 and the values ​​of these four elements may be any set value, for example, 0, and Z1 22 is the time T in the second period 22 In the above, the data element corresponds to a parameter prediction value of a first signal generator among the K signal generators, specifically, a frequency prediction value F1 22 The data element corresponding to the phase prediction value P1 22 The data element corresponding to the amplitude prediction value A1 22 The data element and the direction prediction value D1 corresponding to 22 and the values ​​of these four elements may be any set value, for example, 0. Similarly, 2n is the time T in the second period 2n In the above, the data element corresponds to a parameter prediction value of a first signal generator among the K signal generators, specifically, a frequency prediction value F1 2n The data element corresponding to the phase prediction value P1 2n The data element corresponding to the amplitude prediction value A1 2n The data element and the direction prediction value D1 corresponding to 2n The data element may include data elements corresponding to:

[0060] For other values ​​of i, Zi 21 ,Zi 22 , , Zi 2n The above-mentioned related content can be referred to for understanding, and detailed description thereof will be omitted here.

[0061] In an alternative embodiment, step S203 may include the following steps.

[0062] Step S203-1: Based on the historical parameter prediction result, a known parameter sequence arranged in time order is obtained.

[0063] Here, the known parameter sequence can be understood by referring to the above-mentioned related content, and a detailed description thereof will be omitted here.

[0064] Step S203-2: Splice the future parameter sequence into the known parameter sequence to obtain a second initial sequence.

[0065] Step S203-3: Obtain a second additional feature based on the real-time task information of the target workplace.

[0066] As described above, the real-time task information may include the product type and product specifications of the wound yarn package product processed in the target workshop. In addition, in the embodiment of the present disclosure, the operating parameters used when the process equipment in the target workshop operates can be obtained based on the real-time task information. If the target workshop is a spinning workshop, the process equipment in the target workshop is a spinning box, and the operating parameters used when the spinning box operates may include the specification parameters of the spinneret in the spinning box, the spinning speed, etc.; if the target workshop is a winding workshop, the process equipment in the target workshop is a winder, and the operating parameters used when the winder operates may include the brand, specification parameters, winding speed, number of winding heads, etc. of the winder.

[0067] Based on this, in one example, the real-time task information of the target workplace and the first workplace indicator may both be the second additional feature, or the operating parameters used when the process equipment in the target workplace operates and the first workplace indicator of the target workplace may both be the second additional feature, where the first workplace indicator of the target workplace can be used to determine the process task type, workplace location, etc. of the target workplace, and the process task type can include producing a wound yarn package product, winding a wound yarn package product, etc.

[0068] Step S203-4: Construct a second feature sequence based on the second initial sequence and the second additional feature.

[0069] Continuing with the above example, the number of signal generators in the target workplace is K (K≧2 and is an integer), and the known parameter sequence may include K known parameter subsequences that correspond one-to-one to the K signal generators, and the known parameter subsequence corresponding to the i-th signal generator among the K signal generators is {Qi 11 ,Qi 12 ,···,Qi 1n}, and the future parameter subsequence corresponding to the i-th signal generator among the K signal generators may be represented as {Zi 21 ,Zi 22 ,···,Zi 2n}, where 1≦i≦K, and i is a positive integer. When the second additional feature includes the real-time task information of the target workplace and the first workplace mark, the second feature sequence constructed based on the second initial sequence obtained by stitching the future parameter sequence to the known parameter sequence and the second additional feature obtained based on the real-time task information of the target workplace may include feature data as shown in Table 2.

[0070] [Table 2]

[0071] Here, Y 11 is the time T in the first period 11 This shows the real-time task information of the target workplace at time T 11 Product type Ty of the wound yarn package product processed in the target workshop 11 , Product Specifications Sp 11 etc., and Y 12 is the time T in the first period 12This shows the real-time task information of the target workplace at time T 12 Product type Ty of the wound yarn package product processed in the target workshop 12 , Product Specifications Sp1 12 etc., and so on, Y 1n is the time T in the first period 1n This shows the real-time task information of the target workplace at time T 1n Product type Ty of the wound yarn package product processed in the target workshop 1n , Product Specifications Sp 1n etc., Y 21 is the time T in the second period 21 The real-time task information of the target workplace is shown in Fig. 1. Specifically, the real-time task information of the target workplace is shown in Fig. 1 at time T 21 Product type TY of wound yarn package products processed in the target workshop 21 , Product Specifications Sp 21 etc., Y 22 is the time T in the second period 22 The real-time task information of the target workplace is shown in Fig. 1. Specifically, the real-time task information of the target workplace is shown in Fig. 1 at time T 22 Product type Ty of wound yarn package products processed in the target workshop 22 , Product Specifications Sp1 22 etc., and so on, Y 2n is the time T in the second period 2n The real-time task information of the target workplace is shown in Fig. 1. Specifically, the real-time task information of the target workplace is shown in Fig. 1 at time T 2n Product type Ty of the wound yarn package product processed in the target workshop 2n , Product Specifications Sp 2n etc., where Wid indicates the first workstation marker of the target workstation.

[0072] Thus, in an embodiment of the present disclosure, the second feature sequence, in addition to including a second initial sequence obtained by splicing the unknown parameter sequence into the known parameter sequence, can also include a second additional feature obtained based on real-time task information of the target workplace, which can ensure the integrity of the feature data included in the second feature sequence, thereby improving the reliability of the current parameter prediction result.

[0073] Step S204: Using the target time series model, predict overall control parameters of the multiple signal generators in a second time period according to the first feature sequence and the second feature sequence, and obtain a current parameter prediction result.

[0074] Here, the target time series model may be a trained time series model, such as an Informer model, an Autoregressive Integrated Moving Average Model, etc.

[0075] After obtaining the first feature sequence and the second feature sequence, the first feature sequence and the second feature sequence can be input into the target time series model to obtain the feature processing result output by the target time series model, and the current parameter prediction result can be obtained according to the feature processing result. As mentioned above, the second feature sequence can be obtained based on the second initial sequence and the second additional feature, and the second initial sequence can be obtained by splicing the future parameter sequence to the known parameter sequence, so that after obtaining the feature processing result, the output sequence corresponding to the future parameter sequence in the feature processing result can be the current parameter prediction result.

[0076] Since the second feature sequence is constructed based on the historical parameter prediction result and the future parameter sequence, in the process of using the target time series model to predict the overall control parameters of the multiple signal generators in the second period based on the first feature sequence and the second feature sequence and obtaining the current parameter prediction result, the historical parameter prediction result contained in the second feature sequence plays an active guiding role in the feature processing of the target time series model, thereby further improving the reliability of the current parameter prediction result.

[0077] Furthermore, in the embodiment of the present disclosure, the target time series model may have a model structure as shown in Figure 3, that is, the target time series model may include an encoder and a decoder. Based on this, in an optional embodiment, step S204 may include the following steps:

[0078] Step S204-1: A first feature sequence is input to an encoder, and a feature processing layer including a first self-attention module and a distillation module in the encoder is used to process the first feature sequence to obtain a first feature mapping result.

[0079] Here, the encoder may include a plurality of feature encoding structures connected in series, and each feature encoding structure may include a first self-attention module and a distillation module connected in series.

[0080] In addition, in an embodiment of the present disclosure, a first self-attention module is used to realize self-attention calculation on input features by using a ProbSparse thinning self-attention mechanism to obtain intermediate features, and the intermediate features are input to a distillation module belonging to the same feature coding structure as the first self-attention module, and the distillation module performs a distillation process on the intermediate features to reduce the complexity of the output features, and the distilled intermediate features are used as the output features of the feature coding structure.

[0081] It is understood that in the embodiment of the present disclosure, the output feature of the last feature encoding structure among the multiple feature encoding structures connected in series is the first feature mapping result.

[0082] Step S204-2: Input the second feature sequence to a second self-attention module in the encoder, and use the second self-attention module to process the second feature sequence to obtain a second feature mapping result.

[0083] Here, the second self-attention module is used to obtain a second feature mapping result by using the ProbSparse thinning self-attention mechanism and the mask mechanism to realize self-attention calculation for the second feature sequence.

[0084] Step S204-3: Input the first feature mapping result and the second feature mapping result to a cross-attention module in the encoder, and use the cross-attention module to process the first feature mapping result and the second feature mapping result to obtain a feature processing result.

[0085] The feature processing results include a current parameter prediction result obtained by predicting the overall control parameters of the plurality of signal generators within the second time period.

[0086] Continuing with the above example, the second feature sequence constructed based on the second initial sequence obtained by splicing the future parameter sequence to the known parameter sequence and the second additional feature obtained according to the real-time task information of the target workplace may include feature data as shown in Table 2. In this case, an output sequence corresponding to the future parameter sequence among the feature processing results is predicted as the overall control parameters of the multiple signal generators in the second time period to obtain a current parameter prediction result. For example, the current parameter prediction result may include K standby parameter subsequences in one-to-one correspondence with the K signal generators, and the standby parameter subsequence corresponding to the i-th signal generator among the K signal generators is expressed as {Z'i 21 ,Z'i 22 ,···,Z'i 2n}, where 1≦i≦K, and i is a positive integer.

[0087] Here, when i=1, Z'1 21 is the time T in the second period 21 In the above, the parameter prediction value of the first signal generator among the K signal generators is represented as F'1 21 , phase prediction value P'1 21 , amplitude prediction value A'1 21 and directional prediction value D'1 21 Z'1 22 is the time T in the second period 22 In the above, the parameter prediction value of the first signal generator among the K signal generators is represented as F'1 22 , phase prediction value P'1 22 , amplitude prediction value A'1 22 and directional prediction value D'1 22 Similarly, Z'1 2n is the time T in the second period 2n In the above, the parameter prediction value of the first signal generator among the K signal generators is represented as F'1 2n , phase prediction value P'1 2n , amplitude prediction value A'1 2n and directional prediction value D'12n may include:

[0088] That is, the current parametric prediction result may include feature data as shown in Table 3.

[0089] [Table 3]

[0090] Based on the above model structure of the target time series model, the current parameter prediction result can be quickly obtained, and the reliability of the current parameter prediction result can be further improved.

[0091] Step S205: Based on the current parameter prediction result, determine one or more target transmitters that need to activate operations within a second period from the multiple signal transmitters, and obtain parameter prediction values ​​within the second period for each target transmitter.

[0092] The current parameter prediction result can be used to control the operating state of each signal generator in the second period, so that for each signal generator, it can determine whether the signal generator is a target generator that needs to operate in the second period based on the current parameter prediction result, and obtain a parameter prediction value in the second period of each target generator, where the parameter prediction value can include a frequency prediction value, a phase prediction value, an amplitude prediction value, and a direction prediction value.

[0093] In one example, for each signal generator, if the frequency prediction value of the parameter prediction value of the signal generator at any time in the second period is 0, the signal generator can be an idleable generator that does not need to activate at that time in the second period, while otherwise, the signal generator can be a target generator that needs to activate at that time in the second period. For example, for a first signal generator of the K signal generators, at time T 21 The frequency prediction value F'1 included in the parameter prediction value of the first signal generator in21 is 0, the first signal generator is turned on at time T 21 On the other hand, for example, at time T 22 The frequency prediction value F'1 included in the parameter prediction value of the first signal generator in 21 If is not 0, the first signal generator is turned on at time T 22 and a time T 22 The parameter prediction value of the first signal generator in is the frequency prediction value F'1 22 , phase prediction value P'1 22 , amplitude prediction value A'1 22 and directional prediction value D'1 22 Includes.

[0094] Step S206: Control each target emitter to emit a noise interference signal within the second time period according to the corresponding parameter prediction value, so as to weaken the second noise signal at the target staff's location in the target workplace within the second time period.

[0095] Here, the second noise signal may be a real-time noise signal that is sensed by the target person within the second period.

[0096] When each target transmitter is controlled to emit a noise interference signal within a second time period according to a corresponding parameter prediction value, the noise interference signal can have a cancelling effect on a second noise signal at the target staff's location in the target workplace within the second time period, thereby weakening the second noise signal at the target staff's location in the target workplace within the second time period.

[0097] In an embodiment of the present disclosure, after obtaining a first noise signal of the location of the target staff member in the target workplace during a first period by executing step S201, if the signal strength of the first noise signal is lower than a predetermined strength threshold, steps S201 to S206 are executed, whereas if the signal strength of the first noise signal is equal to or greater than the predetermined strength threshold, the historical parameter prediction result can be used as the current parameter prediction result as is.

[0098] In addition, in the embodiment of the present disclosure, before executing the noise processing method, a target time series model can be obtained by training the time series model as follows.

[0099] A third noise signal of the location of the target staff in the designated workplace during a third time period is obtained, where the designated workplace is one of a plurality of production workplaces having different functions related to the spinning process, and the target staff is an operator entering the target workplace.

[0100] Using the time series model, predict overall control parameters of the multiple signal emitters arranged in the target workplace within a fourth time period based on the third noise signal, and obtain a prediction result of the current training parameters, where the fourth time period is a future time period of the third time period.

[0101] Based on the current training parameter prediction result, one or more preliminary signal generators that need to operate within a fourth period are determined from the multiple signal generators, and a training parameter prediction value within the fourth training period of each preliminary signal generator is obtained.

[0102] Each of the preliminary oscillators is controlled to emit a noise interference signal within a fourth time period according to the corresponding training parameter predicted value.

[0103] Optimize model parameters of the time series model based on the difference between the second noise signal and an ideal noise signal at the target staff's location in the designated work area in the fourth time period, where the ideal noise signal can be set according to application needs.

[0104] The above contents can be understood by referring to the related contents, and detailed explanations will be omitted here.

[0105] Note that in the embodiment of the present disclosure, if the difference between the second noise signal and the ideal noise signal satisfies a predetermined difference requirement, the latest time series model may be the target time series module.

[0106] In some alternative embodiments, the noise processing method may include, before performing step 102 or step 202, the following.

[0107] An abnormal signal in the first noise signal is removed to obtain a new first noise signal.

[0108] Here, the abnormal signal may be a noise signal other than the noise signal generated during operation of the process equipment in the target workplace, such as a voice signal generated by a worker shouting in the target workplace or a noise signal generated by a worker touching the process equipment.

[0109] Based on this, when performing step S102, the overall control parameters of the multiple signal generators in the second time period can be predicted based on the new first noise signal to obtain a current parameter prediction result.When performing step 202, a first feature sequence can be constructed based on the new first noise signal and the historical parameter prediction result, so that subsequent steps in the noise processing method can be performed based on the first feature sequence.

[0110] Since the new first noise signal is the first noise signal with the abnormal signal removed, when performing subsequent steps in the noise processing method based on the new first noise signal, the interference of the abnormal signal can be avoided, thereby improving the reliability of the current parameter prediction result.

[0111] In one example, the step of "removing the abnormal signal in the first noise signal to obtain a new first noise signal" may include the steps of:

[0112] performing a wavelet packet transform on the first noise signal to obtain a plurality of initial wavelet packet coefficients;

[0113] performing a threshold process on each initial wavelet packet in the plurality of initial wavelet packet coefficients to obtain a plurality of target wavelet packet coefficients that correspond one-to-one to the plurality of initial wavelet packet coefficients;

[0114] and performing an inverse wavelet packet transform on the plurality of target wavelet packet coefficients to obtain a new first noise signal.

[0115] Referring to Fig. 4, in a specific example, a wavelet packet kernel constrained convolution network can be used to perform wavelet packet transform on a first noise signal to obtain a plurality of initial wavelet packet coefficients. When performing wavelet packet transform on a first noise signal, a predetermined multi-channel wavelet regularization term can be used as the transform constraint condition of the filter in the wavelet packet kernel constrained convolution network to decompose the first noise signal to obtain a plurality of initial wavelet packet coefficients. Here, the multi-channel wavelet regularization term can be expressed as follows:

number

[0116] where c denotes the channel of filter h in the wavelet packet kernel constrained convolutional network, C denotes the number of channels of filter h in the wavelet packet kernel constrained convolutional network, and R wave denotes the single-channel wavelet regularization term, p denotes the number of filters h, and u denotes the order of filter h.

[0117] After obtaining a plurality of initial wavelet packet coefficients, a threshold process can be performed on each initial wavelet packet in the plurality of initial wavelet packet coefficients using an activation network to obtain a plurality of target wavelet packet coefficients that correspond one-to-one to the plurality of initial wavelet packet coefficients. For each initial wavelet packet coefficient, a threshold process can be performed on the initial wavelet packet coefficient according to a preset soft shrinkage function in the activation network to obtain a target wavelet packet coefficient corresponding to the initial wavelet packet coefficient. Here, the soft shrinkage function can be expressed as follows:

number

[0118] where λ denotes a learnable parameter, sigmoid(λ) denotes using the sigmoid function to process λ, σ denotes the standard deviation operation, and x denotes the initial wavelet packet coefficients that need to be thresholded.

[0119] After obtaining a plurality of target wavelet packet coefficients having one-to-one correspondence with a plurality of initial wavelet packet coefficients, an inverse wavelet packet transform is performed on the plurality of target wavelet packet coefficients using an inverse wavelet packet kernel constrained convolution network, and a new first noise signal can be obtained by performing reconstruction of the noise signal.

[0120] In this way, in the embodiment of the present disclosure, the improved wavelet packet denoising technology can be used to remove the abnormal signal in the first noise signal to obtain a new first noise signal. The improved wavelet packet denoising technology has a good removal effect on the abnormal signal, so that the interference of the abnormal signal can be avoided to the maximum extent, and the reliability of the current parameter prediction result can be further improved.

[0121] Furthermore, in the embodiment of the present disclosure, the series-connected wavelet packet core constrained convolution network, the activation network and the inverse wavelet packet core constrained convolution network are treated as an abnormal signal processing network, and multiple abnormal signal processing networks are connected in series to obtain an abnormal signal processing model, and further, the abnormal signal in the first noise signal is removed by the abnormal signal processing model to obtain a new first noise signal, thereby further improving the reliability of the current parameter prediction result.

[0122] As described above, in the embodiment of the present disclosure, the arrangement of the plurality of signal generators in the target workplace is correlated with the real-time noise sound field in the target workplace. Based on this, in an optional embodiment, the noise processing method includes, before performing step S101 or step S201,

[0123] Selecting a number of intense noise points in the target workspace that are maximum noise points based on a real-time noise sound field in the target workspace;

[0124] Selecting a plurality of object points on top of a target work area, the object points corresponding one-to-one to a plurality of strong noise points;

[0125] The method may further include adjusting an arrangement manner of the multiple signal transmitters in the target workplace such that the multiple signal transmitters are arranged at the multiple target points in one-to-one correspondence with the multiple target points.

[0126] Here, the real-time noise sound field in the target workplace can be obtained by means of room acoustic simulation.

[0127] In one example, a workspace model of the target workspace can be obtained by first modeling the target workspace according to workspace characteristics of the target workspace in acoustic simulation software.

[0128] Here, the acoustic simulation software may be COMSOL multiphysics simulation software, the workplace characteristics may include building characteristics and equipment characteristics of the process equipment, the building characteristics may include building structure, building size, building materials used, etc., and the equipment characteristics may include the number of process equipment installed, layout manner, and equipment structure, equipment size, manufacturing materials used, etc. of the process equipment. Based on this, in this example, the workplace model may include a building model and an equipment model of the process equipment. Further, in this example, according to the building characteristics of the target workplace, building modeling is performed to obtain a building model of the target workplace, equipment modeling is performed according to the equipment characteristics of the process equipment in the target workplace, and an equipment model of the process equipment in the target workplace can be obtained, and an acoustic simulator can be added in each equipment model.

[0129] Then, first simulation parameters are obtained based on the real-time task information of the target workplace, and the acoustic simulation software controls the acoustic simulators added in each equipment model in the workplace model of the target workplace to emit noise simulation signals according to the first simulation parameters, and further obtains the real-time noise sound field in the target workplace.

[0130] Here, the real-time task information may include the product type, product specifications, etc. of the wound yarn package product processed in the target workshop, and the operating parameters used when the process equipment in the target workshop operates can be obtained based on the real-time task information. Based on this, in this example, the operating parameters used when the process equipment in the target workshop operates are obtained based on the real-time task information of the target workshop, any process equipment in the target workshop is selected as a test equipment, and the startup operation of the test equipment is controlled based on the operating parameters, and a second pickup installed near the test equipment is used to collect real-time noise signals generated when the test equipment operates, the real-time noise signals are analyzed, and first simulation parameters are obtained, and an acoustic simulator is controlled to emit a noise simulation signal that is the same as the real-time noise signal (e.g., the same in frequency, phase, amplitude, and direction). When the process equipment is a spinning box, the operating parameters used when the process equipment operates may include the specification parameters of the spinneret in the spinning box, the spinning speed, etc. When the process equipment is a winding machine, the operating parameters used when the process equipment operates may include the brand, specification parameters, winding speed, number of winding heads, etc. of the winding machine, and the second pickup may be an electromagnetic pickup, a piezoelectric pickup, an optical fiber pickup, a digital pickup, etc.

[0131] After obtaining the real-time noise sound field in the target workplace, a number of strong noise points in the target workplace where the noise is the largest can be selected based on the real-time noise sound field in the target workplace, and a number of target points that correspond one-to-one with the number of strong noise points at the top of the target workplace can be further selected, and the arrangement manner of the number of signal transmitters in the target workplace can be adjusted so that the number of signal transmitters are arranged at the number of target points in one-to-one correspondence with the number of target points.

[0132] In one example, a mounting frame for a signal generator is installed on the top of the target workplace, the mounting frame includes a number of long rails fixed to the top of the target workplace, and a number of movable rails for mounting a signal generator are installed between two adjacent long rails. Here, the movable rails can move on the two long rails corresponding to the movable rails in response to a first movement control command, and the signal generator can move on the movable rail corresponding to the signal generator in response to a second movement control command. Based on this, a number of target points that correspond one-to-one to a number of strong noise points on the top of the target workplace are selected, and then the signal generators can be arranged on the target points in one-to-one correspondence with the number of target points by moving the movable rails and / or the signal generators attached to the movable rails.

[0133] 5, for example, based on the real-time noise sound field in the target work area 501, eight strong noise points with the maximum noise are selected from the target work area 501, the eight strong noise points are respectively strong noise point A1, strong noise point A2, strong noise point A3, strong noise point A4, strong noise point A5, strong noise point A6, strong noise point A7 and strong noise point A8. Then, eight object points having one-to-one correspondence with the plurality of strong noise points can be selected from the top 501-1 of the target work area 501, the eight object points are respectively object point A1', object point A2', object point A3', object point A4', object point A5', object point A6', object point A7' and object point A8'.

[0134] Further referring to FIG. 6, after selecting eight target points on the top of the target workplace which correspond one-to-one with the eight strong noise points, the movable rail 601 can be moved and the signal transmitter 602 attached to the movable rail 601 can be moved so that the signal transmitter 602 is located at the target point A1', and in the same manner, the signal transmitter 604 can be located at the target point A2', the signal transmitter 606 can be located at the target point A3', the signal transmitter 608 can be located at the target point A4', the signal transmitter 610 can be located at the target point A5', the signal transmitter 612 can be located at the target point A6', the signal transmitter 614 can be located at the target point A7', and the signal transmitter 616 can be located at the target point A8'.

[0135] In the embodiment of the present disclosure, the arrangement manner of the multiple signal transmitters in the target workplace is automatically adjusted, so that the automation degree of the noise processing method can be improved. Meanwhile, since the multiple target points correspond one-to-one with the multiple strong noise points in the target workplace where the noise is the largest, after the multiple signal transmitters are arranged at the multiple target points in one-to-one correspondence with the multiple target points, the noise interference signal emitted by each target transmitter in the second period according to the corresponding parameter prediction value can generate a better offsetting effect on the second noise signal at the target staff's location in the target workplace in the second period.

[0136] In order to better implement the above noise processing method, the embodiment of the present disclosure further provides a noise processing device, which can be applied to an electronic device, which communicates with a plurality of signal generators arranged in a target workplace, and the arrangement manner of the plurality of signal generators in the target workplace has a corresponding relationship with the real-time noise sound field in the target workplace.

[0137] Hereinafter, a noise processing device according to an embodiment of the present disclosure will be described with reference to the configuration diagram shown in FIG.

[0138] A noise processing device,

[0139] A first acquisition unit 701 for acquiring a first noise signal of a target staff member's location in a target workplace within a first time period;

[0140] a second obtaining unit 702 for predicting overall control parameters of the multiple signal generators in a second period, which is a future period of the first period, based on the first noise signal, and obtaining a current parameter prediction result;

[0141] a third obtaining unit 703 for determining one or more target transmitters that need to activate operations within a second time period from the multiple signal transmitters according to the current parameter prediction result, and obtaining a parameter prediction value of each target transmitter within the second time period; and

[0142] a noise processing device including an emission control unit 704 for controlling each target emitter to emit a noise interference signal within a second period, which is a future period of the first period, according to the corresponding parameter prediction value, so as to attenuate a second noise signal at the location of the target staff in the target workplace within the second period.

[0143] In one alternative embodiment, the second acquisition unit 702 is used to do the following:

[0144] A first feature sequence is constructed based on the first noise signal and a historical parameter prediction result obtained by predicting the overall control parameters of the multiple signal emitters in the first period according to the zeroth noise signal of the location of the target staff member in the target workplace in the zeroth period, which is a historical period of the first period.

[0145] A second feature sequence is constructed based on the historical parameter prediction results and a future parameter sequence, which is an input sequence that corresponds to the current parameter prediction results.

[0146] Using the target time series model, predict overall control parameters of the multiple signal generators in a second time period according to the first feature sequence and the second feature sequence to obtain a current parameter prediction result.

[0147] In one alternative embodiment, the second acquisition unit 702 is used to do the following:

[0148] A time-sequentially arranged noise signal sequence is obtained based on the first noise signal.

[0149] A time-ordered sequence of known parameters is obtained based on the historical parameter prediction results.

[0150] A first initial sequence is obtained by splicing the known parameter sequence to the noise signal sequence.

[0151] A first additional feature is obtained based on real-time task information of the target workplace.

[0152] A first feature sequence is constructed based on the first initial sequence and the first additional feature.

[0153] In one alternative embodiment, the second acquisition unit 702 is used to do the following:

[0154] A time-ordered sequence of known parameters is obtained based on the historical parameter prediction results.

[0155] A second initial sequence is obtained by splicing the future parameter sequence to the known parameter sequence.

[0156] A second additional feature is obtained based on the real-time task information of the target workplace.

[0157] A second feature sequence is constructed based on the second initial sequence and the second additional feature.

[0158] In an alternative embodiment, the target time series model includes an encoder and a decoder, and the second acquisition unit 702 is used to:

[0159] The first feature sequence is input to the encoder, and a feature processing layer including a first self-attention module and a distillation module in the encoder is used to process the first feature sequence to obtain a first feature mapping result.

[0160] The second feature sequence is input to a second self-attention module in the encoder, and the second self-attention module is used to process the second feature sequence to obtain a second feature mapping result.

[0161] The first feature mapping result and the second feature mapping result are input to a cross-attention module in the encoder, and the cross-attention module is used to process the first feature mapping result and the second feature mapping result to obtain a feature processing result including a current parameter prediction result obtained by predicting the overall control parameters of the multiple signal generators within a second period.

[0162] In an alternative embodiment, the noise processing device further includes a noise removal unit.

[0163] The noise removal unit is used for removing the abnormal signal in the first noise signal to obtain a new first noise signal.

[0164] Also, obtaining a prediction result of the current parameters of the plurality of signal generators within the second time period based on the first noise signal includes:

[0165] and predicting an overall control parameter of the plurality of signal generators within a second time period based on the new first noise signal to obtain a current parameter prediction result.

[0166] In one alternative embodiment, the noise reduction unit is used to do the following:

[0167] A wavelet packet transform is performed on the first noise signal to obtain a number of initial wavelet packet coefficients.

[0168] A threshold process is performed on each initial wavelet packet in the plurality of initial wavelet packet coefficients to obtain a plurality of target wavelet packet coefficients that correspond one-to-one to the plurality of initial wavelet packet coefficients.

[0169] An inverse wavelet packet transform is performed on the multiple target wavelet packet coefficients to obtain a new first noise signal.

[0170] In an alternative embodiment, the noise processing device further includes a transmitting head placement unit for:

[0171] Based on the real-time noise sound field in the target workspace, a number of strong noise points are selected where the noise in the target workspace is maximum.

[0172] A number of object points on top of the object work area that correspond one-to-one with a number of strong noise points are selected.

[0173] The arrangement of the plurality of signal transmitters in the target workplace is adjusted so that the plurality of signal transmitters are arranged at the plurality of target points in one-to-one correspondence with the plurality of target points.

[0174] For specific functions and exemplary descriptions of each module and sub-module of the apparatus according to the embodiments of the present disclosure, reference may be made to the relevant descriptions of the corresponding steps in the embodiments of the above method, and detailed descriptions will be omitted here.

[0175] The acquisition, storage, and application of users' personal information in the technical solution disclosed herein complies with the provisions of relevant laws and regulations and does not violate public order and morals.

[0176] FIG. 8 is a block diagram of an electronic device according to an embodiment of the present disclosure. As shown in FIG. 8, the electronic device includes a memory 801 and a processor 802. The memory 801 stores a computer program executable by the processor 802. The number of the memory 801 and the processor 802 may be one or more. The memory 801 may store one or more computer programs. When the one or more computer programs are executed by the electronic device, the electronic device executes the method according to the above method embodiment. The electronic device may further include a communication interface 803 for communicating with an external device and interactively transmitting data.

[0177] If the memory 801, the processor 802, and the communication interface 803 are separate, the memory 801, the processor 802, and the communication interface 803 can be connected to each other and communicate with each other via a bus. The bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. For convenience, only one thick line is shown in FIG. 8, but this does not mean that there is only one bus or one type of bus.

[0178] Optionally, in a specific implementation, when the memory 801, the processor 802, and the communication interface 803 are integrated into one chip, the memory 801, the processor 802, and the communication interface 803 can communicate with each other via an internal interface.

[0179] It should be understood that the processor may be a Central Processing Unit (CPU) or other general purpose processor, a Digital Signal Processing (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, a discrete gate, transistor logic device, a discrete hardware component, etc. The general purpose processor may be a microprocessor or any conventional processor, etc. It should be noted that the processor may be a processor capable of supporting an Advanced Reduced Instruction Set Machine (ARM) architecture.

[0180] Additionally, optionally, the memory may include read-only memory and random access memory, or may include non-volatile random access memory. The memory may be volatile or non-volatile memory, or may include both volatile and non-volatile memory. Non-volatile memory may include Read-Only Memory (ROM), Programmable Read-Only Memory (PROM), Erasable PROM (EPROM), Electrically EPROM (EEPROM), or flash memory. Volatile memory may include Random Access Memory (RAM) used as an external cache. The above description is illustrative only and not restrictive. Many forms of RAM are available. For example, static random access memory (Static RAM, SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (Synchronous DRAM, SDRAM), double data rate synchronous dynamic random access memory (Double Data Rate SDRAM, DDR SDRAM), enhanced synchronous dynamic random access memory (Enhanced SDRAM, ESDRAM), synchlink dynamic random access memory (Synchlink DRAM, SLDRAM), and direct memory bus random access memory (Direct RAM BUS RAM, DR RAM) may be used.

[0181] In the above embodiments, all or part of the above may be realized by software, hardware, firmware, or any combination thereof. When realized by software, all or part of the above may be realized in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer instructions are loaded and executed by a computer, the flow or function described in the embodiments of the present disclosure is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or may be transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wire (e.g., coaxial cable, optical fiber, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, Bluetooth, microwave, etc.). The computer-readable storage medium may be any available medium accessible by a computer, or may be a data storage device such as a server, data center, etc. that includes one or more available media. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, or a magnetic tape), an optical medium (e.g., a Digital Versatile Disc (DVD)), a semiconductor medium (e.g., a Solid State Disk (SSD)), etc. Note that the computer-readable storage medium according to the present disclosure may be a non-volatile storage medium, in other words, a non-transitory storage medium.

[0182] Those skilled in the art can understand that all or part of the steps for realizing the above embodiments may be completed by hardware, or may be completed by instructing related hardware by a program, and the program may be stored in a computer-readable storage medium, and the storage medium may be a read-only memory, a magnetic disk, an optical disk, etc.

[0183] In describing the embodiments of the present disclosure, the use of the terms "one embodiment", "some embodiments", "examples", "particular examples", or "some examples" means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present disclosure. In addition, the specific features, structures, materials, or characteristics described can be combined in any suitable manner in any one or more embodiments or examples. In addition, unless mutually inconsistent, a person skilled in the art can combine features in different embodiments or examples and different embodiments or examples described herein.

[0184] In the description of the embodiments of the present disclosure, unless otherwise specified, " / " means "or", for example, "A / B" can represent "A" or "B". In this specification, "and / or" is merely a relation describing related objects, and means that three relations may exist, for example, "A and / or B" can indicate three situations: "A" exists alone, "A" and "B" exist simultaneously, and "B" exists alone.

[0185] In describing the embodiments of the present disclosure, the terms "first" and "second" are descriptive only and should not be understood to indicate or imply a relative importance or number of the indicated components. Thus, a feature qualified with "first" or "second" can explicitly or implicitly include one or more of the feature. In describing the embodiments of the present disclosure, unless otherwise specified, "plurality" means two or more.

[0186] The above are merely illustrative examples of the present disclosure, and are not intended to limit the present disclosure. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present disclosure should be included within the scope of the claims of the present disclosure.

Claims

1. A noise processing method applied to an electronic device, comprising: The electronic device communicates with a plurality of signal emitters disposed in the target workplace, and a layout of the plurality of signal emitters in the target workplace correlates with a real-time noise sound field in the target workplace; The noise processing method includes: Obtaining a first noise signal of a target staff member's location in the target workplace during a first time period; Predicting overall control parameters of the plurality of signal generators in a second time period, which is a future time period of the first time period, based on the first noise signal, to obtain a current parameter prediction result; determining one or more target transmitters that need to operate within the second time period from the plurality of signal transmitters based on the current parameter prediction result, and obtaining a parameter prediction value within the second time period for each of the target transmitters; and controlling each of the target emitters to emit a noise interference signal during the second time period according to the corresponding parameter prediction value, so as to attenuate a second noise signal at a target staff position in the target workplace during the second time period; Predicting overall control parameters of the plurality of signal generators within a second period based on the first noise signal to obtain a current parameter prediction result; constructing a first feature sequence based on the first noise signal and a historical parameter prediction result obtained by predicting an overall control parameter of the plurality of signal generators during the first period according to a zeroth noise signal of a location of a target staff member in the target workplace during a zeroth period, which is a historical period of the first period; constructing a second feature sequence based on the historical parameter prediction results and a future parameter sequence, the future parameter sequence being an input sequence corresponding to the current parameter prediction results; and predicting an overall control parameter of the plurality of signal generators in the second time period according to the first feature sequence and the second feature sequence by using a target time series model to obtain the current parameter prediction result. Noise processing method.

2. Constructing a first feature sequence based on the first noise signal and historical parameter prediction results includes: obtaining a time-ordered noise signal sequence based on the first noise signal; obtaining a time-ordered sequence of known parameters based on the historical parameter prediction results; splicing the known parameter sequence onto the noise signal sequence to obtain a first initial sequence; Obtaining a first additional feature based on real-time task information of the target workplace; and constructing the first feature sequence based on the first initial sequence and the first additional feature. The noise processing method according to claim 1 .

3. Constructing a second feature sequence based on the historical parameter prediction results and a future parameter sequence includes: obtaining a time-ordered sequence of known parameters based on the historical parameter prediction results; splicing an unknown parameter sequence into the known parameter sequence to obtain a second initial sequence; Obtaining a second additional feature based on real-time task information of the target workplace; and and constructing the second feature sequence based on the second initial sequence and the second additional feature. The noise processing method according to claim 1 .

4. the target time series model includes an encoder and a decoder; predicting an overall control parameter of the plurality of signal generators in the second time period according to the first feature sequence and the second feature sequence by using the target time series model, and obtaining the current parameter prediction result; inputting the first feature sequence into the encoder, and processing the first feature sequence using a feature processing layer in the encoder, the feature processing layer including a first self-attention module and a distillation module, to obtain a first feature mapping result; inputting the second feature sequence into a second self-attention module in an encoder, and using the second self-attention module to process the second feature sequence to obtain a second feature mapping result; inputting the first feature mapping result and the second feature mapping result into a cross-attention module in the encoder, and using the cross-attention module to process the first feature mapping result and the second feature mapping result to obtain a feature processing result, the feature processing result including the current parameter prediction result obtained by predicting an overall control parameter of the plurality of signal generators within the second time period; The noise processing method according to any one of claims 1 to 3.

5. The noise processing method includes: The method further includes removing an abnormal signal in the first noise signal to obtain a new first noise signal; Obtaining current parameter prediction results of the plurality of signal generators within a second time period based on the first noise signal includes: predicting an overall control parameter of the plurality of signal generators within the second time period based on the new first noise signal, and obtaining the current parameter prediction result; The noise processing method according to claim 1 .

6. Removing the abnormal signal in the first noise signal and obtaining a new first noise signal performing a wavelet packet transform on the first noise signal to obtain a plurality of initial wavelet packet coefficients; performing a threshold process on each initial wavelet packet in the plurality of initial wavelet packet coefficients to obtain a plurality of target wavelet packet coefficients that correspond one-to-one to the plurality of initial wavelet packet coefficients; performing an inverse wavelet packet transform on the plurality of target wavelet packet coefficients to obtain the new first noise signal; Including, The noise processing method according to claim 5.

7. The noise processing method includes: Selecting a number of intense noise points in the target workspace that are maximum noise points based on a real-time noise sound field in the target workspace; selecting a plurality of object points on top of the object work area that correspond one-to-one to the plurality of strong noise points; Adjusting the arrangement of the plurality of signal transmitters in the target workplace so that the plurality of signal transmitters are arranged at the plurality of target points in one-to-one correspondence with the plurality of target points; Including, The noise processing method according to claim 1 .

8. A noise processing device applied to an electronic device, The electronic device communicates with a plurality of signal emitters disposed in the target workplace, and a layout of the plurality of signal emitters in the target workplace correlates with a real-time noise sound field in the target workplace; The noise processing device includes: A first acquisition unit for acquiring a first noise signal of a target staff member's location in the target workplace within a first time period; a second acquisition unit for predicting an overall control parameter of the plurality of signal generators in a second period, which is a future period of the first period, based on the first noise signal, and acquiring a current parameter prediction result; a third obtaining unit for determining one or more target transmitters that need to operate within the second time period from the plurality of signal transmitters according to the current parameter prediction result, and obtaining a parameter prediction value of each of the target transmitters within the second time period; an emission control unit for controlling each of the target emitters to emit a noise interference signal within the second time period according to a corresponding parameter prediction value, so as to attenuate a second noise signal at a target staff position in the target workplace within the second time period; The second acquisition unit is constructing a first feature sequence based on the first noise signal and a historical parameter prediction result obtained by predicting an overall control parameter of the plurality of signal generators during the first period according to a zeroth noise signal of the location of the target staff member in the target workplace during a zeroth period, which is a historical period of the first period; constructing a second feature sequence based on the historical parameter prediction results and a future parameter sequence, the future parameter sequence being an input sequence corresponding to the current parameter prediction results; Utilizing a target time series model, predicting an overall control parameter of the plurality of signal generators in the second time period according to the first feature sequence and the second feature sequence, and obtaining the current parameter prediction result. Noise processing device.

9. The second acquisition unit is obtaining a time-ordered noise signal sequence based on the first noise signal; obtaining a time-ordered sequence of known parameters based on the historical parameter prediction results; splicing the known parameter sequence onto the noise signal sequence to obtain a first initial sequence; Obtain a first additional feature based on real-time task information of the target workplace; used to construct the first feature sequence based on the first initial sequence and the first additional feature. The noise processing device according to claim 8 .

10. The second acquisition unit is obtaining a time-ordered sequence of known parameters based on the historical parameter prediction results; splicing the future parameter sequence into the known parameter sequence to obtain a second initial sequence; Obtain a second additional feature based on the real-time task information of the target workplace; used to construct the second feature sequence based on the second initial sequence and the second additional feature. The noise processing device according to claim 8 .

11. The target time series model includes an encoder and a decoder. The second acquisition unit is inputting the first feature sequence into the encoder, and processing the first feature sequence using a feature processing layer in the encoder, the feature processing layer including a first self-attention module and a distillation module, to obtain a first feature mapping result; inputting the second feature sequence into a second self-attention module in an encoder, and using the second self-attention module to process the second feature sequence to obtain a second feature mapping result; inputting the first feature mapping result and the second feature mapping result into a cross-attention module in the encoder, and using the cross-attention module to process the first feature mapping result and the second feature mapping result to obtain a feature processing result, the feature processing result including the current parameter prediction result obtained by predicting an overall control parameter of the plurality of signal generators within the second time period; The noise processing device according to any one of claims 8 to 10.

12. The noise processing device includes: The method further includes a noise removal unit for removing an abnormal signal in the first noise signal to obtain a new first noise signal; Obtaining current parameter prediction results of the plurality of signal generators within a second time period based on the first noise signal includes: predicting an overall control parameter of the plurality of signal generators within the second time period based on the new first noise signal, and obtaining the current parameter prediction result; The noise processing device according to claim 8 .

13. The noise removal unit includes: Perform a wavelet packet transform on the first noise signal to obtain a number of initial wavelet packet coefficients; Perform a threshold process on each initial wavelet packet in the plurality of initial wavelet packet coefficients to obtain a plurality of target wavelet packet coefficients that correspond one-to-one to the plurality of initial wavelet packet coefficients; performing an inverse wavelet packet transform on the plurality of target wavelet packet coefficients to obtain the new first noise signal; The noise processing device according to claim 12.

14. The noise processing device includes: Further comprising a transmitting head placement unit; The transmitter head arrangement unit includes: Selecting a plurality of strong noise points in the target workspace, which are the maximum noise points, based on the real-time noise sound field in the target workspace; Select a plurality of target points on top of the target work area that correspond one-to-one to the plurality of strong noise points; is used to adjust the arrangement of the plurality of signal transmitters in the target workplace so that the plurality of signal transmitters are arranged at the plurality of target points in one-to-one correspondence with the plurality of target points; The noise processing device according to claim 8 .

15. At least one processor; a memory in communication with the at least one processor; The memory stores instructions executable by the at least one processor, the instructions, when executed by the at least one processor, causing the at least one processor to perform a method according to any one of claims 1 to 3. electronic equipment.

16. A non-transitory computer readable storage medium having stored thereon instructions for causing a computer to perform the method of any one of claims 1 to 3.

Images