Noise reduction method for UPS chassis stamping

Abstract

The present invention relates to the field of stamping noise processing technology, and specifically to a noise reduction method suitable for UPS chassis stamping and forming, including obtaining historical data of metal plate stamping and forming, and parameter information of the metal plate before stamping; constructing a noise frequency prediction model, the noise frequency prediction model deeply learns the historical data to obtain the relationship between the parameter information and the stamping noise frequency; and predicts the noise frequency according to the learning result to obtain the predicted noise frequency; determines the initial parameters for installing a sound insulation device according to the predicted noise frequency; performs real-time noise monitoring on the stamping and forming, and adjusts the initial parameters of the sound insulation device according to the real-time noise monitoring results. The method of the present invention adjusts the initial structure and initial position of the sound insulation device through the monitoring results and the stamping quality feedback of the metal plate, effectively reducing the noise while meeting the heat dissipation and structural strength requirements of the chassis.

Description

Technical Field

[0001] The invention relates to the technical field of stamping noise processing, and in particular to a noise reduction method suitable for stamping forming of a UPS chassis. Background Art

[0002] With the rapid development of electronic equipment, the chassis design of the Uninterruptible Power Supply (UPS), as an important device to ensure the stable operation of key systems, must be comprehensively considered in many aspects; especially the balance between heat dissipation and mechanical structure strength, which has become a key technical challenge in the design process.

[0003] UPS chassis are usually made of metal plates (such as steel or aluminum), and their design needs to meet good heat dissipation performance to ensure that internal components will not fail due to overheating during operation. Therefore, stamping and forming heat dissipation holes on the top and side panels of the chassis has become a common solution. According to the principles of thermodynamics, the area and shape of the heat dissipation holes directly affect the heat dissipation effect. A larger opening ratio helps to improve the heat dissipation capacity of the chassis and prevent the equipment from malfunctioning due to excessive temperature.

[0004] However, the increase in the number of openings will significantly reduce the mechanical structural strength of the chassis, which may cause deformation or damage during use. Therefore, it is necessary to find a suitable balance between meeting the heat dissipation requirements and maintaining the strength of the chassis. For example, in order to protect the structural integrity of the chassis, the thickness of the chassis is increased, but this will bring challenges to the stamping process. Processing thicker metal plates requires higher stamping force and may cause other production process problems. In addition, the noise problem in the stamping process is also a serious technical bottleneck. When the stamping machine processes the metal plate, it will produce high-intensity impact sound and vibration, which will not only cause hearing damage to the workers, but also affect the overall noise level of the production environment, and even cause fatigue damage to the equipment. Therefore, how to effectively reduce the noise generated during the stamping process has become an urgent problem to be solved in the design and manufacturing of UPS chassis.

[0005] In response to the above technical challenges, existing technologies, when solving the stamping noise problem, mostly focus on the application of physical sound insulation materials or directly adjust the equipment working parameters, but they often lack specificity and systematicness, and cannot effectively reduce noise while meeting the heat dissipation and structural strength requirements of the chassis. Summary of the invention

[0006] To this end, the technical problem to be solved by the present invention is to overcome the difficulty in the prior art to effectively reduce the stamping noise of the UPS chassis in a targeted and systematic manner, and to provide a noise reduction method suitable for stamping and forming of a UPS chassis, which effectively reduces the noise while meeting the heat dissipation and structural strength requirements of the chassis.

[0007] In order to solve the above technical problems, the present invention provides a noise reduction method applicable to stamping of a UPS chassis, wherein the UPS chassis comprises a top plate and a side plate, wherein the top plate and the side plate both comprise a metal plate body, and heat dissipation holes are processed on the metal plate body based on a stamping process. The noise reduction method comprises the following steps:

[0008] Acquire historical data of stamping of a metal plate body and parameter information of the metal plate body before stamping; the parameter information includes attribute parameters of the metal plate body and attribute parameters of the heat dissipation holes to be processed;

[0009] Constructing a noise frequency prediction model, wherein the noise frequency prediction model deeply learns the historical data to obtain the relationship between the parameter information and the stamping noise frequency; and performing noise frequency prediction according to the learning result to obtain the predicted noise frequency;

[0010] Determine the initial parameters for installing the sound insulation device according to the predicted noise frequency; the initial parameters include an initial position and an initial structure;

[0011] Real-time noise monitoring is performed on the stamping process, and initial parameters of the sound insulation device are adjusted according to the real-time noise monitoring results.

[0012] In one embodiment of the present invention, the determining of the initial structure for installing the sound insulation device comprises:

[0013] Determining the height of the sound insulation device;

[0014] Determine the thickness of the sound insulation device; wherein the sound insulation device includes an inner cover body and an outer cover body formed into an integral structure, and an air layer provided between the inner cover body and the outer cover body; the thickness of the sound insulation device includes the sum of the body thickness of the inner cover body, the body thickness of the outer cover body and the thickness of the air layer;

[0015] And, determining the materials for making the outer cover and the inner cover.

[0016] In one embodiment of the present invention, after acquiring the historical data of metal plate stamping, the method further includes:

[0017] extracting a data set related to ram noise from the historical data;

[0018] Spectral features are extracted from the data set, and the initial structure of the sound insulation device is determined according to the spectral features.

[0019] In one embodiment of the present invention, determining the initial structure of the sound insulation device according to the frequency spectrum characteristics includes:

[0020] Obtain the main frequency of the noise according to the spectrum characteristics ;

[0021] According to the main frequency Determine wavelength ;

[0022] According to the wavelength Determine the height of the sound insulation device as ; and determine the thickness of the sound insulation device to be ; Wherein, h and n are both non-integer constants greater than zero.

[0023] In one embodiment of the present invention, extracting a data set related to stamping noise from the historical data includes:

[0024] Identifying noise impact factors in the historical data to obtain a set of noise impact factors;

[0025] Extracting a first noise impact factor from the noise impact factor set, and extracting factor data and noise data from the historical data according to the first noise impact factor to obtain a first noise factor sample data set;

[0026] Performing a straight line fitting on the first noise factor sample data set to obtain a first fitting result; performing a discrete distribution probability calculation on the first fitting result to obtain a first discrete probability;

[0027] If the first discrete probability is greater than or equal to a predetermined discrete probability threshold, adding the first noise impact factor to a set of non-influencing factors;

[0028] The difference between the noise influencing factor set and the non-influencing factor set is taken as the data set related to stamping noise.

[0029] In one embodiment of the present invention, the initial parameters of the sound insulation device are adjusted according to the real-time noise monitoring results, including:

[0030] Obtaining the initial structure of the sound insulation device and the real-time noise monitoring result;

[0031] Extracting frequency characteristics according to the real-time noise monitoring results, and analyzing the frequency characteristics;

[0032] The initial position of the sound insulation device is obtained, and the initial structure of the sound insulation device is adjusted according to the initial position and the frequency characteristic analysis result.

[0033] In one embodiment of the present invention, the initial parameters of the sound insulation device are adjusted according to the real-time noise monitoring results, including:

[0034] Obtaining the initial position of the sound insulation device and the real-time noise monitoring result;

[0035] Extracting frequency characteristics according to the real-time noise monitoring results, and analyzing the frequency characteristics;

[0036] The initial structure of the sound insulation device is obtained, and the initial position of the sound insulation device is adjusted according to the initial structure and the frequency characteristic analysis result.

[0037] In one embodiment of the present invention, a training set and a validation set are constructed to train the noise frequency prediction model, which includes:

[0038] Dividing the historical data into m subsets, extracting a first subset as the validation set, and combining the other subsets other than the first subset as the training set, to obtain a first performance parameter of the noise frequency prediction model;

[0039] Extracting a second subset from the m subsets as the verification set, and combining the other subsets other than the second subset as the training set to obtain a second performance parameter of the noise frequency prediction model;

[0040] Extracting a third subset from the m subsets as the validation set, and combining the other subsets except the third subset as the training set to obtain a third performance parameter of the noise frequency prediction model;

[0041] Repeat the above steps until the mth performance parameter of the noise frequency prediction model is obtained;

[0042] An evaluation result of the noise frequency prediction model is obtained according to the first performance parameter, the second performance parameter, the third performance parameter, and finally the mth performance parameter.

[0043] In one embodiment of the present invention, the method further includes collecting the stamping quality of the metal plate body, and adjusting the initial parameters of the sound insulation device according to feedback of the stamping quality of the metal plate body.

[0044] The above technical solution of the present invention has the following beneficial effects compared with the prior art:

[0045] The noise reduction method applicable to stamping forming of a UPS chassis of the present invention comprises the following steps: providing a sound insulation device to reduce the noise of a stamping noise source; determining the initial structure and initial position of the sound insulation device by learning historical data, monitoring the noise and the noise reduction effect in real time, and adjusting the initial structure and initial position of the sound insulation device by the monitoring result and the stamping quality feedback of the metal plate body, thereby effectively reducing the noise and meeting the heat dissipation and structural strength requirements of the chassis. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to make the content of the present invention more clearly understood, the present invention is further described in detail below according to specific embodiments of the present invention in conjunction with the accompanying drawings, wherein:

[0047] Figure 1 This is a schematic diagram of the UPS chassis structure;

[0048] Figure 2 It is a schematic diagram of the structure of a metal plate after heat dissipation holes are processed on the metal plate based on a stamping process;

[0049] Figure 3 It is a flow chart of a noise reduction method applicable to UPS chassis stamping and forming in a preferred embodiment of the present invention;

[0050] Figure 4 The flowchart of extracting a data set related to stamping noise from historical data in a preferred embodiment of the present invention.

[0051] Explanation of the reference numerals in the specification: 2 - chassis; 4 - metal plate; 6 - heat dissipation hole. DETAILED DESCRIPTION

[0052] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments so that those skilled in the art can better understand the present invention and implement it, but the embodiments are not intended to limit the present invention.

[0053] Reference Figure 1 The UPS case is shown, which is used to assemble and place UPS modules. The case 2 is assembled from a base, a top plate and side plates; wherein the top plate and the side plates both include a metal plate body 4, and heat dissipation holes 6 are processed on the metal plate body 4 based on a stamping process; Figure 2 The figure shows a schematic diagram of the structure of a metal plate after heat dissipation holes are processed on the metal plate based on the stamping process. On the one hand, the number, diameter and spacing of the heat dissipation holes affect the overall heat dissipation effect of the chassis 2. At the same time, the number, diameter and spacing of the heat dissipation holes affect the overall mechanical structure strength of the chassis 2. Since the UPS module has requirements for both the heat dissipation effect and the mechanical structure strength of the chassis 2, to a certain extent (for example, when the overall volume and weight meet the conditions), the thickness of the metal plate can be increased to meet both requirements. However, the increase in the thickness of the metal plate brings certain problems to the stamping process, especially the increase in the stamping force, and the resulting stamping noise problem becomes particularly significant.

[0054] The embodiment of the present invention discloses a method for reducing noise during stamping and forming of a UPS chassis, the purpose of which is to reduce noise generated during stamping and forming in accordance with the requirements of heat dissipation performance and mechanical structural strength of the UPS chassis. Figure 3 As shown, the noise reduction method includes the following steps:

[0055] S100: Acquire historical data of stamping of a metal plate and parameter information of the metal plate before stamping; the parameter information includes property parameters of the metal plate and property parameters of the heat dissipation holes to be processed;

[0056] S200: constructing a noise frequency prediction model, wherein the noise frequency prediction model deeply learns the historical data to obtain the relationship between the parameter information and the stamping noise frequency; and predicting the noise frequency according to the learning result to obtain the predicted noise frequency;

[0057] S300: Determine initial parameters for installing a sound insulation device according to the predicted noise frequency; the initial parameters include an initial position and an initial structure;

[0058] S400: performing real-time noise monitoring on the stamping process, and adjusting initial parameters of the sound insulation device according to the real-time noise monitoring result.

[0059] In specific application scenarios, the property parameters of the metal plate and the property parameters of the heat dissipation holes to be processed will seriously affect the noise frequency generated by the stamping machine. The influencing channels include: (1) The elastic modulus, density and internal friction characteristics of different materials will affect the noise frequency generated during stamping; (2) The thickness and hardness of the stamping material affect the noise frequency: thicker materials may produce lower-frequency sounds, while thinner materials may produce higher-frequency sounds; materials with greater hardness may produce higher noise frequencies during stamping; (3) The running speed, stamping force and punch design of the stamping machine affect the noise frequency: the faster the running speed, the higher the noise frequency; the size of the stamping force will affect the vibration during the stamping process, thereby affecting the frequency characteristics of the noise; the shape, design and wear degree of the punch will directly affect the noise characteristics during the forming process. (4) The stamping mode and stamping gap affect the noise frequency: different stamping modes (such as single stamping, step-by-step stamping, etc.) will have different effects on the noise frequency; the size of the gap between the punch and the die will affect the deformation mode of the material, and then affect the noise frequency. The above influencing channels also affect the heat dissipation effect and mechanical structure strength of the chassis after the heat dissipation holes are processed.

[0060] The noise frequency prediction model is used to match the stamping noise frequency corresponding to the parameter information of the metal plate body that needs to be stamped. The noise frequency prediction model uses deep learning technology, such as a neural network model, to learn the historical data of stamping and obtain the relationship between the parameter information of the metal plate body and the stamping noise frequency. It should be noted here that the noise frequency prediction model should be understood as using a neural network architecture to build a neural network basic model, and the neural network basic model is trained through historical data. During the training, the sample parameter information is used as input and the sample stamping noise frequency is used as output. During the training process, the model parameters are continuously optimized until the convergence requirements are met. The neural network basic model that is qualified after the training is completed and verified is the noise frequency prediction model. After the noise frequency prediction model is obtained, the parameter information is input into the noise frequency prediction model to obtain the stamping noise frequency. The initial parameters of the sound insulation device are set by learning the noise frequency prediction model, and the actual working parameters of the sound insulation device are adjusted according to the real-time noise monitoring results and the initial parameters of the sound insulation device, so that the stamping noise is effectively reduced under the premise of meeting the heat dissipation and mechanical structure strength of the chassis. At the same time, the application of the prediction model can discover potential problems in advance and reduce downtime in production.

[0061] The noise reduction method applicable to stamping forming of a UPS chassis of the present invention comprises the following steps: providing a sound insulation device to reduce the noise of a stamping noise source; determining an initial position and an initial structure of the sound insulation device by learning historical data, monitoring the noise and the noise reduction effect in real time, and adjusting the initial position and the initial structure of the sound insulation device by the monitoring result and the stamping quality feedback of the metal plate body, thereby effectively reducing the noise and meeting the heat dissipation and structural strength requirements of the chassis.

[0062] It should be noted that the sound insulation device is arranged outside the area where the stamping machine is located, and has a certain straight-line distance from the stamping machine. The area where the distance is located is defined as the noise reduction area; the sound insulation device reduces various indicators of stamping noise in the noise reduction area. Specifically, the sound insulation device includes an inner cover body and an outer cover body formed as an integral structure, and an air layer arranged between the inner cover body and the outer cover body. The sound insulation device includes an inner cover body and an outer cover body formed as an integral structure, which should be understood as follows: the sound insulation device includes an inner cover body and an outer cover body, and the inner cover body and the outer cover body form an integral structure. The sound insulation device composed of the inner cover body, the air layer and the outer cover body reduces various indicators of stamping noise based on sound absorption, sound insulation, increased energy loss, and reduced sound wave reflection.

[0063] Furthermore, determining the initial structure for installing the sound insulation device includes determining the height of the sound insulation device; determining the thickness of the sound insulation device; and determining the materials for making the outer cover and the inner cover. The thickness of the sound insulation device includes the sum of the thickness of the inner cover, the thickness of the outer cover, and the thickness of the air layer. The effect of the height, thickness, and material of the sound insulation device on noise reduction is related to the noise frequency, and the influencing ways include:

[0064] Sound-absorbing materials (such as spray polyurethane, sound-absorbing cotton, mineral wool, etc.) reduce the reflection and propagation of sound waves; use multi-layer composite materials and combine different material properties (such as density, elasticity, etc.) to enhance the sound insulation effect; damping materials can reduce the noise caused by structural vibration, increase energy loss, and thus reduce sound propagation; enhance the sound insulation effect through the air layer; reasonable sound insulation device size can reduce the reflection and focusing of sound waves, etc.

[0065] Specifically, after obtaining historical data of stamping forming of metal plates, a data set related to stamping noise is extracted from the historical data; spectrum features are extracted from the data set, and the initial structure of the sound insulation device is determined according to the spectrum features; wherein determining the initial structure of the sound insulation device according to the spectrum features includes obtaining the main frequency of the noise according to the spectrum features According to the main frequency Determine wavelength According to the wavelength Determine the height of the sound insulation device as ; and determine the thickness of the sound insulation device to be ; Wherein, h and n are non-integer constants greater than zero; for example, h is 1.3; n is 0.8. The noise frequency prediction model extracts a data set related to stamping noise from historical data through deep learning, extracts spectral features from the data set, and obtains the main frequency of the noise based on the spectral features. , thereby matching the predicted noise frequency corresponding to the parameter information of the metal plate that needs to be stamped, and determining the thickness and height of the sound insulation device according to the corresponding main noise frequency. Among them, the thickness and height of the sound insulation device are in direct proportion to the wavelength of the main noise frequency, which can improve the sound insulation effect; especially the proportional relationship is a non-integer multiple relationship, which can improve the sound insulation effect while avoiding the influence of resonance on the noise reduction effect.

[0066] In the method of the present invention, the noise frequency prediction model can be obtained by training based on any model training method of the prior art. However, the existing model training method often automatically learns and extracts features related to noise frequency based on multi-source heterogeneous data such as material properties of historical sound insulation devices, press parameters, stamping process parameters, chassis structure design and environmental factors, and then uses machine learning algorithms to train the prediction model; however, multi-source heterogeneous data such as material properties of historical sound insulation devices, press parameters, stamping process parameters, chassis structure design and environmental factors have different effects on the accuracy of the prediction model. If all the data are directly used for model training, it is easy to lead to insufficient prediction accuracy, thereby affecting the adjustment accuracy and accuracy of the structure and installation position of the sound insulation device.

[0067] In order to solve the above problem, the method of the present invention extracts a data set related to stamping noise from the historical data, referring to Figure 4 As shown, it includes the following steps,

[0068] Step 1: Identify the noise impact factors in the historical data to obtain a set of noise impact factors;

[0069] In specific application scenarios, historical data refers to stamping record data at historical times, including attribute parameters of metal plates (materials, structures, dimensions, etc.), attribute parameters of heat dissipation holes (aperture, hole spacing, hole shape, number of holes, etc.), stamping machine attributes (stamping speed, stamping force, punch design, stamping mode, stamping gap, quality, stiffness, damping characteristics of punch and die base, etc.); and the running time, running environment, maintenance records, etc. of the stamping machine, and further noise influencing factor identification is performed on the historical data to obtain a set of noise influencing factors. Specifically, existing analysis methods, such as factor analysis, can be used to analyze historical data to obtain multiple indicators that are associated with noise, such as stamping force, metal plate thickness, number of heat dissipation holes, etc. as a set of noise influencing factors.

[0070] Step 2: extracting a first noise impact factor from the noise impact factor set, extracting factor data and noise data from the historical data according to the first noise impact factor, and obtaining a first noise factor sample data set;

[0071] In a specific application scenario, the first noise impact factor in the noise impact factor set is extracted, and the factor data and noise data are extracted from the historical data according to the first noise impact factor, that is, the factor and noise data corresponding to the first noise impact factor are extracted. The corresponding factor data and noise data are extracted for all noise impact factors to form a first noise factor sample data set. For example, if the first noise impact factor is the thickness of the metal plate, it is only necessary to extract the thickness of the metal plate and the corresponding noise data from the historical data set and put them into the first noise factor sample data set.

[0072] Step 3: performing a straight line fitting on the first noise factor sample data set to obtain a first fitting result; performing a discrete distribution probability calculation on the first fitting result to obtain a first discrete probability;

[0073] In a specific application scenario, performing straight-line fitting on the first noise factor sample data set is to fit a correlation straight line between the noise and the first noise influencing factor as the first fitting result, and performing discrete distribution probability calculation on the first fitting result is to analyze the sample distribution probability of the first noise factor sample data set other than the first fitting result to obtain the first discrete probability. It should be noted that the larger the first discrete probability, the worse the fitting effect between the noise and the first noise influencing factor. If the first noise influencing factor is used as a feature of the training model, it is easy to make the model difficult to train to converge, the convergence speed is slow, and the prediction accuracy is low.

[0074] Step 4: If the first discrete probability is greater than or equal to a predetermined discrete probability threshold, add the first noise impact factor to a set of non-influencing factors;

[0075] In specific application scenarios, the predetermined discrete probability threshold is set by professional and technical personnel based on experimental results or industry regulations. When the first discrete probability is greater than or equal to the predetermined discrete probability threshold, it means that in the result of straight-line fitting of the first noise factor sample data set, there are more data distributed in a farther range of the first fitting result, that is, the first fitting result is not representative and it is difficult to truly reflect the relationship between noise and the first noise influencing factor. Therefore, the first noise influencing factor is added to the set of non-influencing factors.

[0076] Step 5: taking the difference between the noise influencing factor set and the non-influencing factor set as the data set related to stamping noise.

[0077] In a specific application scenario, the difference between the set of noise influencing factors and the set of non-influencing factors is the data set related to stamping noise, that is, sample data with unrepresentative fitting results are eliminated to improve the model training speed and prediction accuracy.

[0078] Furthermore, it also includes constructing a training set and a validation set to train the noise frequency prediction model, which includes,

[0079] Dividing the historical data into m subsets, extracting a first subset as the validation set, and combining the other subsets other than the first subset as the training set, to obtain a first performance parameter of the noise frequency prediction model;

[0080] Extracting a second subset from the m subsets as the verification set, and combining the other subsets other than the second subset as the training set to obtain a second performance parameter of the noise frequency prediction model;

[0081] Extracting a third subset from the m subsets as the validation set, and combining the other subsets except the third subset as the training set to obtain a third performance parameter of the noise frequency prediction model;

[0082] Repeat the above steps until the mth performance parameter of the noise frequency prediction model is obtained;

[0083] An evaluation result of the noise frequency prediction model is obtained according to the first performance parameter, the second performance parameter, the third performance parameter, and finally the mth performance parameter.

[0084] In specific application scenarios, historical data are divided into m subsets, and each subset is used as a validation set. Historical data can be fully utilized for model training and evaluation. By using different subsets as validation sets, the generalization ability of the model can be verified, that is, whether the model has good predictive ability for data in different time periods or situations, and the risk of model overfitting can be reduced. Because each subset is used as a validation set, the model will not rely too much on a specific data set during training. Finally, the average or weighted average of the performance parameters of each subset as a validation set is calculated to obtain a comprehensive evaluation result of the model on the entire historical data, thereby comprehensively evaluating the performance of the model.

[0085] Specifically, adjusting the initial parameters of the sound insulation device includes adjusting the initial structure and initial position of the sound insulation device to obtain a target structure and installation position that meet the requirements of the sound insulation device in actual work. The adjustment strategy for the initial structure is:

[0086] The initial structure, initial position and real-time noise monitoring result of the sound insulation device are obtained; frequency characteristics are extracted according to the real-time noise monitoring result, and the frequency characteristics are analyzed; the initial structure of the sound insulation device is adjusted according to the initial position of the sound insulation device and the frequency characteristic analysis result.

[0087] The adjustment strategy for the initial position is:

[0088] The initial position, initial structure and real-time noise monitoring result of the sound insulation device are obtained; frequency characteristics are extracted according to the real-time noise monitoring result, and the frequency characteristics are analyzed; the initial position of the sound insulation device is adjusted according to the initial structure and frequency characteristic analysis result of the sound insulation device.

[0089] In a specific application scenario, the height, thickness, manufacturing material and initial position of the sound insulation device are determined according to the predicted noise frequency; it can be understood that after obtaining the predicted noise frequency, the installation position, height, thickness and manufacturing material of the sound insulation device required to reduce the stamping noise with the predicted noise frequency to meet the requirements are determined, thereby obtaining the initial structure and initial position of the sound insulation device. Combined with the real-time noise monitoring results, and analyzing the frequency characteristics of the noise monitoring results, the real-time noise frequency is obtained. According to the real-time noise frequency, the height, thickness, manufacturing material and installation position of the sound insulation device actually required can be determined. At this time, on the one hand, the target height, target thickness and target manufacturing material of the sound insulation device are determined while the initial installation position remains unchanged; at the same time, the actual installation position of the sound insulation device is determined while the initial structure remains unchanged. In this way, the initial position and initial structure of the sound insulation device are cross-adjusted so that the noise reduction capacity of the sound insulation device has a margin, which significantly improves the noise reduction effect. Among them, the noise reduction capacity of the sound insulation device has a margin, which can be understood as:

[0090] The installation position and structure (including height, thickness and manufacturing material) of the sound insulation device jointly affect the sound insulation effect of the sound insulation device; for example, the farther the distance between the sound insulation device and the noise source within a certain distance, the better the sound insulation and noise reduction effect; the sound insulation and noise reduction effect can be changed by changing the material of the sound insulation device, etc. The initial position and initial structure of the sound insulation device are obtained according to historical data prediction. In the historical data, the sound insulation device achieves the required noise reduction capacity through the initial position and initial structure. When adjusting, the actual height, actual thickness and actual manufacturing material of the sound insulation device are determined under the condition that the initial installation position remains unchanged, and the coordination of the adjusted structure determined thereby with the initial installation position makes the noise reduction capacity of the sound insulation device have a margin compared with the noise reduction capacity of the sound insulation device under the historical data; similarly, the actual installation position of the sound insulation device is determined under the condition that the initial structure remains unchanged, and the coordination of the adjusted installation position determined thereby with the initial structure makes the noise reduction capacity of the sound insulation device have a margin compared with the noise reduction capacity of the sound insulation device under the historical data.

[0091] In other embodiments of the present invention, the stamping quality of the metal plate is collected, and the initial parameters of the sound insulation device are adjusted according to the feedback of the stamping quality of the metal plate. The stamping quality of the metal plate includes dimensional accuracy, surface quality, and mechanical properties. The parameters of the stamping process are adjusted according to the feedback of the stamping quality of the metal plate. After the stamping process parameters are changed, the stamping noise changes accordingly. By adjusting the initial parameters of the sound insulation device to respond to the change of the stamping noise, the heat dissipation and structural strength requirements of the chassis are met while effectively reducing the noise.

[0092] In summary, the noise reduction method applicable to stamping forming of a UPS chassis described in the present invention provides a sound insulation device to reduce the noise of the stamping noise source; and determines the initial structure and initial position of the sound insulation device by learning historical data, and monitors the noise and noise reduction effect in real time, and adjusts the initial structure and initial position of the sound insulation device according to the monitoring results and the stamping quality feedback of the metal plate body, so as to effectively reduce the noise while meeting the heat dissipation and structural strength requirements of the chassis.

[0093] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0094] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0095] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0096] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0097] Obviously, the above embodiments are merely examples for clear explanation and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived from these are still within the protection scope of the invention.

Claims

1. A noise reduction method applicable to stamping of a UPS chassis, wherein the UPS chassis comprises a top plate and a side plate, wherein the top plate and the side plate both comprise a metal plate body, and heat dissipation holes are processed on the metal plate body based on a stamping process, wherein: The noise reduction method The following steps are included: Acquire historical data of metal plate stamping, wherein the historical data includes process data of metal plate stamping and parameter information of the metal plate before stamping; the parameter information includes property parameters of the metal plate and property parameters of the heat dissipation holes to be processed; Constructing a noise frequency prediction model, wherein the noise frequency prediction model deeply learns the historical data to obtain the relationship between the parameter information and the stamping noise frequency; And predict the noise frequency according to the learning result to obtain the predicted noise frequency; Determine the initial parameters for installing the sound insulation device according to the predicted noise frequency; the initial parameters include an initial position and an initial structure; Performing real-time noise monitoring on the stamping process, and adjusting initial parameters of the sound insulation device according to the real-time noise monitoring result; Wherein, after acquiring the historical data of stamping forming of metal plates, it also includes extracting a data set related to stamping noise from the historical data, extracting the data set includes identifying the noise influencing factors in the historical data to obtain a set of noise influencing factors; extracting a first noise influencing factor from the set of noise influencing factors, extracting factor data and noise data from the historical data according to the first noise influencing factor to obtain a first noise factor sample data set; performing a straight line fitting on the first noise factor sample data set to obtain a first fitting result; performing a discrete distribution probability calculation on the first fitting result to obtain a first discrete probability; if the first discrete probability is greater than or equal to a predetermined discrete probability threshold, adding the first noise influencing factor to a set of factors that do not affect the stamping noise; taking the difference between the set of noise influencing factors and the set of factors that do not affect the noise as the data set related to stamping noise; The data set is used to train the noise frequency prediction model; the predicted noise frequency is obtained according to the noise frequency prediction model, and the main frequency of the noise is determined according to the predicted noise frequency According to the main frequency Determine wavelength According to the wavelength Determine the height of the sound insulation device as ; and determine the thickness of the sound insulation device to be ; Wherein, h and n are both non-integer constants greater than zero.

2. The noise reduction method for UPS chassis stamping according to claim 1 is characterized in that: The initial structure for installing the sound insulation device includes: Determining the height of the sound insulation device; Determine the thickness of the sound insulation device; wherein the sound insulation device includes an inner cover body and an outer cover body formed into an integral structure, and an air layer provided between the inner cover body and the outer cover body; the thickness of the sound insulation device includes the sum of the body thickness of the inner cover body, the body thickness of the outer cover body and the thickness of the air layer; And, determining the materials for making the outer cover body and the inner cover body.

3. The noise reduction method applicable to UPS chassis stamping and forming according to claim 1, characterized in that: The initial parameters of the sound insulation device are adjusted according to the real-time noise monitoring results, including: Obtaining the initial structure of the sound insulation device and the real-time noise monitoring result; Extracting frequency characteristics according to the real-time noise monitoring results, and analyzing the frequency characteristics; The initial position of the sound insulation device is obtained, and the initial structure of the sound insulation device is adjusted according to the initial position and the frequency characteristic analysis result.

4. The noise reduction method applicable to UPS chassis stamping and forming according to claim 1 or 3, characterized in that: The initial parameters of the sound insulation device are adjusted according to the real-time noise monitoring results, including: Obtaining the initial position of the sound insulation device and the real-time noise monitoring result; Extracting frequency characteristics according to the real-time noise monitoring results, and analyzing the frequency characteristics; The initial structure of the sound insulation device is obtained, and the initial position of the sound insulation device is adjusted according to the initial structure and the frequency characteristic analysis result.

5. The noise reduction method applicable to UPS chassis stamping and forming according to claim 1, characterized in that: Constructing a training set and a validation set to train the noise frequency prediction model, which includes: Dividing the historical data into m subsets, extracting a first subset as the validation set, and combining the other subsets other than the first subset as the training set, to obtain a first performance parameter of the noise frequency prediction model; Extracting a second subset from the m subsets as the verification set, and combining the other subsets other than the second subset as the training set to obtain a second performance parameter of the noise frequency prediction model; Extracting a third subset from the m subsets as the validation set, and combining the other subsets except the third subset as the training set to obtain a third performance parameter of the noise frequency prediction model; Repeat the above steps until the mth performance parameter of the noise frequency prediction model is obtained; An evaluation result of the noise frequency prediction model is obtained according to the first performance parameter, the second performance parameter, the third performance parameter, and finally the mth performance parameter.

6. The noise reduction method applicable to UPS chassis stamping and forming according to claim 1, characterized in that: The method also includes collecting the stamping quality of the metal plate body and adjusting the initial parameters of the sound insulation device according to feedback of the stamping quality of the metal plate body.

Images