Discharge noise control method, device, equipment and medium

Abstract

The invention discloses a discharge noise control method and device, equipment and a medium, and the method comprises the steps: carrying out the feature extraction and relation modeling of current related data through employing a multi-layer deep neural network, outputting a current noise prediction value, and rapidly and accurately providing data support for the subsequent electric field optimization; furthermore, a group of optimal electric field parameters are generated through iterative optimization by adopting a preset constraint condition, so that interference of high-frequency noise to a user is reduced, and a basis is provided for subsequent real-time electric field adjustment; besides, the positions of the electrodes are adjusted based on the optimal electric field parameters to ensure that the distance between the electrodes reaches the optimization target, and the power of the power supply is adjusted based on the optimal electric field parameters to ensure that the output power of the power supply meets the optimization requirement, so that the optimal discharge condition is maintained.

Description

Technical Field

[0001] The present application relates to the field of plasma technology, and in particular to a discharge noise control method, device, equipment and medium. Background Art

[0002] With the widespread adoption of intelligent home appliances and industrial equipment, plasma technology is increasingly being used in these devices. These devices generate plasma through high-frequency discharge technology, achieving functions such as heating, purification, cutting, and cleaning. However, due to the inherent characteristics of the high-frequency discharge process, these devices generate a certain degree of electromagnetic interference and discharge noise during operation. This noise not only affects the normal operation of the devices but also adversely affects the quality of life of users.

[0003] Existing noise control technologies mainly rely on traditional mechanical and physical means, such as adding sound insulation materials and adjusting the equipment's housing structure. However, existing discharge noise control methods have problems such as poor stability and low intelligence. Summary of the Invention

[0004] The present application aims to at least solve the technical problems existing in the prior art. To this end, the first aspect of the present application proposes a discharge noise control method, which includes:

[0005] Obtain current relevant data during high-frequency discharge;

[0006] Inputting the current relevant data into a preset noise prediction model for processing to generate a current noise prediction value; wherein the preset noise prediction model is trained based on a relevant data sample set during a high-frequency discharge process, the relevant data sample set including a discharge noise signal sample set, an electric field distribution data sample set, and an environmental data sample set. The preset noise prediction model is constructed using a multi-layer deep neural network architecture;

[0007] The current noise prediction value is minimized using preset constraints to generate optimal electric field parameters; the preset constraints include electrode spacing constraints, voltage preset conditions, current constraints, and environmental safety constraints;

[0008] Based on the optimal electric field parameters, the electrode position and power supply power are adjusted.

[0009] In one possible implementation, the current relevant data is input into a preset noise prediction model for processing to generate a current noise prediction value, including:

[0010] Obtaining an input eigenvector matrix based on current relevant data; wherein the current relevant data includes a discharge noise signal, electric field distribution data, and environmental data, the discharge noise signal includes frequency data and sound pressure level data, the electric field distribution data includes voltage data and current data, and the environmental data includes ambient temperature data, ambient humidity data, and air composition data;

[0011] The input feature vector matrix is received through the input layer of the multi-layer deep neural network and distributed to the hidden layer;

[0012] Extract and transform the input feature vector matrix through a preset number of hidden layers to obtain feature data;

[0013] The feature data is processed by the output layer using a linear activation function to generate the current noise prediction value.

[0014] In one possible implementation, the current noise prediction value includes current electric field parameters, and the current noise prediction value is minimized using preset constraints to generate optimal electric field parameters, including:

[0015] Under preset constraints, the current electric field parameters are iteratively solved based on a preset genetic algorithm to generate optimal electric field parameters.

[0016] In a possible implementation, iteratively solving the current electric field parameters based on a preset genetic algorithm under preset constraints to generate optimal electric field parameters includes:

[0017] generating an initial population; wherein the initial population includes a plurality of random samples, each random sample corresponds to a set of initial electric field distribution parameters, and the initial electric field distribution parameters satisfy preset constraints;

[0018] Repeat the steps of selecting individuals whose current fitness meets the preset fitness threshold for crossover and mutation operations to generate the next generation population until the preset termination condition is met; wherein the preset termination condition includes reaching the preset maximum number of iterations or the current noise prediction value reaching convergence, and the current fitness is opposite to the current noise prediction value.

[0019] In one possible implementation, individuals whose current fitness meets a preset fitness threshold are selected for crossover and mutation operations to generate the next generation population, including:

[0020] Selecting an initial electric field distribution parameter group to be operated corresponding to an individual whose current fitness meets a preset fitness threshold;

[0021] A preset simulated binary crossover algorithm is used to randomly select two individuals to perform crossover recombination on the initial electric field distribution parameters to be operated, thereby generating a new electric field distribution parameter group;

[0022] The new electric field distribution parameter group is randomly perturbed in the offspring population to obtain the next generation population.

[0023] In one possible implementation, the process of adjusting the electrode position includes:

[0024] Get the current electrode spacing;

[0025] Calculating a target adjustment amount of the electrode position based on the optimal electrode spacing in the optimal electric field parameters and the current electrode spacing;

[0026] Based on the target adjustment amount, the electrode position is adjusted.

[0027] In one possible implementation, the power adjustment process includes:

[0028] Get current voltage data and current current data;

[0029] Calculate the target voltage adjustment amount based on the optimal voltage data in the optimal electric field parameters and the current voltage data, and calculate the target current adjustment amount based on the optimal current data in the optimal electric field parameters and the current current data;

[0030] The power of the power supply is adjusted based on the target voltage adjustment amount and the target current adjustment amount.

[0031] In a second aspect, the present application provides a discharge noise control device, comprising:

[0032] An acquisition module is used to obtain current relevant data during the high-frequency discharge process;

[0033] A first generation module is configured to input current relevant data into a preset noise prediction model for processing to generate a current noise prediction value; wherein the preset noise prediction model is trained based on a relevant data sample set during a high-frequency discharge process, the relevant data sample set including a discharge noise signal sample set, an electric field distribution data sample set, and an environmental data sample set, and the preset noise prediction model is constructed using a multi-layer deep neural network architecture;

[0034] The second generation module is used to minimize the current noise prediction value using preset constraints to generate optimal electric field parameters; wherein the preset constraints include electrode spacing constraints, voltage preset conditions, current constraints, and environmental safety constraints;

[0035] The adjustment module is used to adjust the electrode position and power supply power based on the optimal electric field parameters.

[0036] In a possible implementation, the first generating module is specifically configured to:

[0037] Obtaining an input eigenvector matrix based on current relevant data; wherein the current relevant data includes a discharge noise signal, electric field distribution data, and environmental data, the discharge noise signal includes frequency data and sound pressure level data, the electric field distribution data includes voltage data and current data, and the environmental data includes ambient temperature data, ambient humidity data, and air composition data;

[0038] The input feature vector matrix is received through the input layer of the multi-layer deep neural network and distributed to the hidden layer;

[0039] Extract and transform the input feature vector matrix through a preset number of hidden layers to obtain feature data;

[0040] The feature data is processed by the output layer using a linear activation function to generate the current noise prediction value.

[0041] In a possible implementation, the current noise prediction value includes a current electric field parameter, and the second generation module is specifically configured to:

[0042] Under preset constraints, the current electric field parameters are iteratively solved based on a preset genetic algorithm to generate optimal electric field parameters.

[0043] In a possible implementation, the second generation module is further configured to:

[0044] generating an initial population; wherein the initial population includes a plurality of random samples, each random sample corresponds to a set of initial electric field distribution parameters, and the initial electric field distribution parameters satisfy preset constraints;

[0045] Repeat the steps of selecting individuals whose current fitness meets the preset fitness threshold for crossover and mutation operations to generate the next generation population until the preset termination condition is met; wherein the preset termination condition includes reaching the preset maximum number of iterations or the current noise prediction value reaching convergence, and the current fitness is opposite to the current noise prediction value.

[0046] In a possible implementation, the second generation module is further configured to:

[0047] Selecting an initial electric field distribution parameter group to be operated corresponding to an individual whose current fitness meets a preset fitness threshold;

[0048] A preset simulated binary crossover algorithm is used to randomly select two individuals to perform crossover recombination on the initial electric field distribution parameters to be operated, thereby generating a new electric field distribution parameter group;

[0049] The new electric field distribution parameter group is randomly perturbed in the offspring population to obtain the next generation population.

[0050] In a possible implementation, the adjustment module is specifically configured to:

[0051] Get the current electrode spacing;

[0052] Calculating a target adjustment amount of the electrode position based on the optimal electrode spacing in the optimal electric field parameters and the current electrode spacing;

[0053] Based on the target adjustment amount, the electrode position is adjusted.

[0054] In a possible implementation, the adjustment module is further configured to:

[0055] Get current voltage data and current current data;

[0056] Calculate the target voltage adjustment amount based on the optimal voltage data in the optimal electric field parameters and the current voltage data, and calculate the target current adjustment amount based on the optimal current data in the optimal electric field parameters and the current current data;

[0057] The power of the power supply is adjusted based on the target voltage adjustment amount and the target current adjustment amount.

[0058] In a third aspect of the present application, an electronic device is proposed, comprising a processor and a memory, wherein the memory stores at least one instruction, at least one program, a code set, or an instruction set, and the at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by the processor to implement the discharge noise control method as described in the first aspect.

[0059] In a fourth aspect, the present application proposes a computer-readable storage medium, wherein the storage medium stores at least one instruction, at least one program, a code set, or an instruction set, and the at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by a processor to implement the discharge noise control method as described in the first aspect.

[0060] The embodiments of the present application have the following beneficial effects:

[0061] The discharge noise control method, device, equipment and medium provided by the embodiment of the present application include: obtaining current relevant data during high-frequency discharge, inputting the current relevant data into a preset noise prediction model for processing, generating a current noise prediction value, minimizing the current noise prediction value using preset constraints, generating optimal electric field parameters, and adjusting the electrode position and power supply power based on the optimal electric field parameters. This solution uses a multi-layer deep neural network to perform feature extraction and relationship modeling on the current relevant data, outputs the current noise prediction value, and quickly and accurately provides data support for subsequent electric field optimization; in addition, by using preset constraints and iterative optimization, a set of optimal electric field parameters is generated to reduce the interference of high-frequency noise on the user, providing a basis for subsequent real-time electric field adjustment; in addition, by adjusting the electrode position based on the optimal electric field parameters, it is ensured that the electrode spacing reaches the optimization target, and by adjusting the power supply power based on the optimal electric field parameters, it is ensured that the output power of the power supply reaches the optimization requirements, thereby maintaining the optimal discharge conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 A block diagram of a computer device provided in an embodiment of the present application;

[0063] Figure 2 A flow chart of the steps of a discharge noise control method provided in an embodiment of the present application;

[0064] Figure 3 A flowchart of the steps for generating a current noise prediction value provided in an embodiment of the present application;

[0065] Figure 4 A flowchart of the steps for generating optimal electric field parameters provided in an embodiment of the present application;

[0066] Figure 5 A flowchart of the steps for generating the next generation population provided in an embodiment of the present application;

[0067] Figure 6 A flowchart of the steps for adjusting the electrode position provided in an embodiment of the present application;

[0068] Figure 7 A flowchart of the steps for adjusting voltage and current provided in an embodiment of the present application;

[0069] Figure 8 This is a structural block diagram of a discharge noise control device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0070] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0071] With the widespread adoption of smart home appliances and industrial equipment, plasma technology is increasingly being used in these devices. These devices generate plasma through high-frequency discharge technology, achieving functions such as heating, purification, cutting, and cleaning. However, due to the inherent characteristics of the high-frequency discharge process, these devices generate a certain degree of electromagnetic interference and discharge noise during operation. This noise not only affects the normal operation of the equipment but also adversely affects the user's quality of life. In particular, the high-frequency discharge noise generated by home appliances such as plasma stoves can easily cause indoor noise pollution, thereby affecting the comfort of the home environment. Furthermore, this noise can pose certain health risks to the human body. Therefore, how to reduce high-frequency discharge noise while maintaining normal operation has become a critical issue for smart home and industrial plasma applications.

[0072] Existing noise control technologies primarily rely on traditional mechanical and physical methods, such as adding soundproofing materials and adjusting the device's housing structure. While these methods can reduce noise to a certain extent, they lack dynamic adjustment to varying operating environments and loads, resulting in unstable noise control and, in some cases, suboptimal noise suppression. Furthermore, traditional noise suppression methods are typically static: once the noise control measures are determined, the device is no longer dynamically adjusted. The noise generated during high-frequency discharge is closely related to the device's operating state. In particular, the characteristics of discharge noise vary significantly under varying power supply voltages, currents, and electrode positions. Existing technologies lack a real-time adjustment mechanism based on intelligent optimization algorithms, making it impossible to optimize the electric field distribution and reduce noise in real time based on the device's operating state. Furthermore, existing technologies often lack a clear mathematical model linking electric field distribution and noise, making it impossible to accurately quantify the relationship between electric field parameters and noise. Consequently, it is impossible to infer noise variations based on the device's current electric field distribution, nor to perform targeted optimization and adjustment based on noise characteristics during operation.

[0073] Based on this, the present application proposes a discharge noise control method, which includes: obtaining current relevant data during high-frequency discharge, inputting the current relevant data into a preset noise prediction model for processing, generating a current noise prediction value, minimizing the current noise prediction value using preset constraints, generating optimal electric field parameters, and adjusting the electrode position and power supply power based on the optimal electric field parameters. This solution uses a multi-layer deep neural network to extract features and model relationships of current relevant data, outputs current noise prediction values, and quickly and accurately provides data support for subsequent electric field optimization; in addition, by using preset constraints and iterative optimization, a set of optimal electric field parameters is generated to reduce the interference of high-frequency noise on users and provide a basis for subsequent real-time electric field adjustment; in addition, by adjusting the electrode position based on the optimal electric field parameters, it is ensured that the electrode spacing reaches the optimization target, and by adjusting the power supply power based on the optimal electric field parameters, it is ensured that the output power of the power supply reaches the optimization requirements, thereby maintaining the optimal discharge conditions.

[0074] In the following, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of the present disclosure, unless otherwise specified, "multiple" means two or more. In addition, the use of "based on" or "according to" means openness and inclusiveness, because the process, steps, calculations or other actions "based on" or "according to" one or more of the conditions or values may be based on additional conditions or values beyond the stated in practice.

[0075] The discharge noise control method provided in the present application can be applied to computer equipment (electronic equipment), which can be a server or a terminal, wherein the server can be a single server or a server cluster composed of multiple servers. The embodiments of the present application do not specifically limit this. The terminal can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers and portable wearable devices.

[0076] Take the computer device as an example, Figure 1 A block diagram of a server is shown, such as Figure 1 As shown, a server may include a processor and memory connected via a system bus. The server's processor provides computing and control capabilities. The server's memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operating system and computer program stored in the non-volatile storage medium. When executed by the processor, the computer program implements a method for controlling discharge noise.

[0077] Those skilled in the art will understand that Figure 1 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the server to which the solution of the present application is applied. Optionally, the server may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0078] It should be noted that the execution subject of the embodiments of the present application can be a computer device or a discharge noise control device. The following method embodiments will be described using a computer device as the execution subject.

[0079] Figure 2 This is a flow chart of the steps of a discharge noise control method provided in an embodiment of the present application. Figure 2 As shown, the method includes the following steps:

[0080] Step 202: Acquire current relevant data during the high-frequency discharge process.

[0081] In plasma cookers and other small household appliances using plasma technology, sensor modules can be used to collect current data related to the high-frequency discharge process. This data may include, but is not limited to, discharge noise signals, electric field distribution data, and environmental data. The discharge noise signal may include, but is not limited to, frequency data and sound pressure level data; the electric field distribution data may include, but is not limited to, voltage data and current data; and the environmental data may include, but is not limited to, ambient temperature data, ambient humidity data, and air composition data.

[0082] Optionally, a high-sensitivity noise sensor can be used to collect real-time discharge noise signals and record the spectrum information in the high-frequency range (usually 20kHz-200kHz). For example, the high-sensitivity noise sensor can be a micro-electromechanical system (MEMS) sensor or a dynamic microphone. The collected discharge noise signal can be recorded as S(t), S(t) = [f1, f2, ..., f n ,L1,L2,…,L n ], where f i Represents the i-th frequency component, in Hz; L i The unit of sound pressure level corresponding to the frequency component is dB.

[0083] Optionally, an electric field probe or a capacitive sensor can be used to collect key parameters of the electric field distribution between electrodes, namely, electric field distribution data, which is recorded as E(t). E(t) = [V1, V2, ..., V m ,I1,I2,…,I m]. Among them, V i represents the voltage of the i-th electric field point (V); I i represents the current (A) at the i-th electric field point.

[0084] Optionally, environmental data can be collected by temperature and humidity sensors and gas sensors, which are denoted as C(t), where C(t) = [T, H, A1, A2, ..., A p ], where T represents the ambient temperature data in °C; H represents the ambient humidity data in percentage; A i Represents air composition data, that is, the concentration of the i-th gas in the air composition, in ppm.

[0085] Step 204: Input the current relevant data into a preset noise prediction model for processing to generate a current noise prediction value.

[0086] Among them, the preset noise prediction model is trained based on the relevant data sample set in the high-frequency discharge process. The relevant data sample set may include the discharge noise signal sample set, the electric field distribution data sample set and the environmental data sample set. The preset noise prediction model is constructed based on a multi-layer deep neural network architecture.

[0087] In some optional embodiments, such as Figure 3 As shown, Figure 3 A flowchart of the steps for generating a current noise prediction value provided in an embodiment of the present application includes:

[0088] Step 302: Obtain an input feature vector matrix based on current relevant data.

[0089] Step 304: Receive the input feature vector matrix through the input layer of the multi-layer deep neural network and distribute it to the hidden layer.

[0090] Step 306: Extract and transform the input feature vector matrix through a preset number of hidden layers to obtain feature data.

[0091] Step 308: Process the feature data using a linear activation function through the output layer to generate a current noise prediction value.

[0092] After obtaining the current relevant data, the current relevant data can be standardized and vectorized to construct a unified input feature vector matrix. Taking the current relevant data including the discharge noise signal, electric field distribution data, and environmental data as an example, these three types of data are vectorized to obtain the vectorized data X(t), X(t) = [S(t), E(t), C(t)], where S(t) represents the noise signal vector; E(t) represents the electric field distribution vector; and C(t) represents the environmental data vector.

[0093] Then, the vectorized data X(t) can be standardized to obtain the input eigenvector matrix, denoted as X, X = [X(1), X(2), ..., X(T)], where T represents the time step, and the total data volume of the input eigenvector matrix is equal to the sampling frequency multiplied by the data monitoring time.

[0094] To quantitatively analyze the relationship between the electric field distribution and high-frequency discharge noise in devices like plasma cookers, a pre-defined noise prediction model is constructed based on a sample set of data collected during the high-frequency discharge process. This model utilizes a multi-layer deep neural network to extract features and model relationships within the input data. This model then outputs a current noise prediction value, providing data support for subsequent electric field optimization.

[0095] In some optional embodiments, during the process of training to obtain the preset noise prediction model, a mean square error may be used as a loss function, which is defined as: Among them, L represents the loss function value, which is the prediction error of the model. The smaller the loss function value, the better the prediction result of the model. The closer it is to the true value Y(t), the better the model performance; T represents the number of samples, that is, the number of time steps or data points in the feature vector matrix X. Since the loss function averages the prediction errors of all samples, T is a normalization factor used to prevent the loss function value from increasing as the number of samples increases.

[0096] After the preset noise prediction model is constructed, the current relevant data can be input into the preset noise prediction model for processing. Optionally, the input feature vector matrix can be received by the input layer of a multi-layer deep neural network and distributed to the hidden layer. The input feature vector matrix is subjected to feature extraction and transformation through a preset number of hidden layers to obtain feature data. The feature data is then processed using a linear activation function through the output layer to generate the current noise prediction value.

[0097] Among them, the input layer can receive the input feature vector matrix X and distribute it to the hidden layer. The hidden layer can set k hidden layers, each layer contains h_i neurons, and uses the ReLU activation function, z (l) =ReLU(W (l) ·z (l-1) +b (l) ),l=1,2,…,k, where W (l) and b (l) are the weight matrix and bias vector respectively.

[0098] The output layer is a single neuron that outputs the predicted noise value And the linear activation function is used to process the feature data. Among them, b (k+1)is the bias vector for this layer.

[0099] In some optional embodiments, the Adam optimizer can also be used to optimize the weight matrix and bias vector in the model parameters. In addition, the learning rate can be set as a dynamic adjustment strategy, which is adjusted according to the convergence of the loss function value to select the model that minimizes the prediction error.

[0100] Step 206: Minimize the current noise prediction value using preset constraints to generate optimal electric field parameters.

[0101] After obtaining the current noise prediction value based on the preset noise prediction model, the electric field parameters can be adjusted through an optimization algorithm to minimize noise during the discharge process. Optionally, a set of optimal electric field parameters can be generated through iterative optimization using preset constraints to reduce the interference of high-frequency noise on users and provide a basis for subsequent real-time electric field adjustments.

[0102] The optimization problem is modeled as a constrained optimization problem, where the goal is to minimize the noise prediction value At the same time, the physical constraints of the electric field distribution are satisfied. The mathematical form of the optimization problem is: in, The predicted noise value depends on the current electric field parameter P, that is, the current noise prediction value includes the current electric field parameter P, P = [d, V, I], where the current electric field parameter P can include the current electrode spacing d, the current voltage distribution V, and the current current intensity I.

[0103] The above-mentioned preset constraints may include electrode spacing constraints, voltage preset conditions, current constraints and environmental safety constraints. min ≤d≤d max , d min is the preset minimum electrode spacing, d max The preset maximum electrode distance. The voltage preset condition can be V min ≤V≤V max , V min is the preset minimum voltage, V max is the preset maximum voltage. The current constraint condition is I min ≤I≤I max , I min is the preset minimum current, I max The environmental safety constraint condition may be a limit on the temperature rise value.

[0104] In some optional embodiments, the current electric field parameters may be iteratively solved based on a preset genetic algorithm under preset constraints to generate optimal electric field parameters.

[0105] Optionally, a preset iterative optimization algorithm may be used for iterative solution, which is not specifically limited in this embodiment.

[0106] In some optional embodiments, the preset iterative optimization algorithm may be a genetic algorithm, such as Figure 4 As shown, Figure 4 A flowchart of steps for generating optimal electric field parameters provided in an embodiment of the present application includes:

[0107] Step 402: Generate an initial population.

[0108] Step 404 , repeatedly executing the step of selecting individuals whose current fitness meets the preset fitness threshold to perform crossover and mutation operations to generate the next generation population, until the preset termination condition is met.

[0109] The initial population includes multiple random samples, each of which corresponds to a set of initial electric field distribution parameters, and the initial electric field distribution parameters satisfy preset constraints. Optionally, an initial population P_0 can be generated, including N random samples, each of which corresponds to a set of initial electric field distribution parameters P0 = [d0, V0, I0], and ensures that the samples satisfy the preset constraints.

[0110] Next, the fitness function Fitness(P) can be defined as the negative value of the current predicted noise value, that is, Individuals with higher fitness have lower current noise prediction values. Therefore, individuals whose current fitness meets the preset fitness threshold can be selected for crossover and mutation operations to generate the next generation population.

[0111] In some optional embodiments, such as Figure 5 As shown, Figure 5 A flowchart of the steps for generating the next generation population provided in an embodiment of the present application includes:

[0112] Step 502: Select an initial electric field distribution parameter group to be operated corresponding to an individual whose current fitness meets a preset fitness threshold.

[0113] Step 504: Use a preset simulated binary crossover algorithm to randomly select two individuals to perform crossover recombination on the initial electric field distribution parameters to be operated, and generate a new electric field distribution parameter group.

[0114] Step 506: Randomly perturb the new electric field distribution parameter group in the offspring population to obtain the next generation population.

[0115] Among them, the preset fitness threshold is usually a large value, that is, it is necessary to select two individuals with higher current fitness to cross-recombine the initial electric field distribution parameters to be operated, so as to generate a new electric field distribution parameter group. Then, the preset simulated binary crossover algorithm can be used to randomly select two individuals to cross-recombine the initial electric field distribution parameters to be operated, so as to obtain the next generation population, that is, to obtain a new electric field distribution parameter group, which is recorded as P', P ′ =αP1+(1-α)P2,α∈[0,1], where P_1 and P_2 are two randomly selected individuals; α represents the weight factor when two individuals undergo crossover recombination, which determines the proportion of the next generation individuals inheriting the characteristics of the two parents.

[0116] Then, the new electric field distribution parameter group can be randomly perturbed in the offspring population to obtain the next generation population, that is, a mutation operation can be performed. Optionally, a small range of random perturbations can be performed on the new electric field distribution parameter group in the offspring population to enhance the global search capability of the algorithm to obtain the next generation population, denoted as P i ′ , Among them, P_i represents the parent individual; P_i' represents the new individual generated by mutation; δ represents the mutation amount, which is a value randomly selected from the uniform distribution U(-σ,σ); U(-σ,σ) represents the uniform distribution, which represents the range of the disturbance amount; σ represents the parameter that controls the amplitude of the mutation and determines the intensity of the mutation.

[0117] Thus, the steps of selecting individuals whose current fitness meets the preset fitness threshold for crossover and mutation operations to generate the next generation population can be repeated until a preset termination condition is met. The preset termination condition includes reaching a preset maximum number of iterations, or the current noise prediction value reaches convergence, or the current fitness is opposite to the current noise prediction value.

[0118] Finally, the optimal electric field parameters can be output, denoted as P * , P * =[d * ,V * ,I * ], where d * represents the optimal electrode spacing, V * Indicates the optimal voltage data, I * Indicates optimal current data.

[0119] Step 208: Adjust the electrode position and power supply based on the optimal electric field parameters.

[0120] Among them, after the optimization algorithm generates the optimal electric field parameters, the optimal electric field parameters need to be converted into actual equipment adjustment parameters to achieve the optimal electric field distribution and discharge effect, thereby reducing noise.

[0121] Alternatively, as Figure 6 As shown, Figure 6 A flowchart of the steps for adjusting the electrode position provided in an embodiment of the present application includes:

[0122] Step 602: Obtain the current electrode spacing.

[0123] Step 604: Calculate the target adjustment amount of the electrode position based on the optimal electrode spacing in the optimal electric field parameters and the current electrode spacing.

[0124] Step 606: Adjust the electrode position based on the target adjustment amount.

[0125] The electrode spacing is a key factor affecting discharge characteristics and noise generation. Too large or too small an electrode spacing can lead to unstable discharge or excessive noise.

[0126] Optionally, the current electrode spacing d can be obtained through the real-time monitoring module current , so that the target adjustment amount of the electrode position can be calculated based on the optimal electrode spacing in the optimal electric field parameters and the current electrode spacing. The target adjustment amount is recorded as Δd, Δd = d * -d current , so that the electrode position can be adjusted based on the target adjustment amount to ensure that the electrode spacing reaches the optimization target.

[0127] Alternatively, as Figure 7 As shown, Figure 7 A flowchart of the steps for adjusting voltage and current provided in an embodiment of the present application includes:

[0128] Step 702: Obtain current voltage data and current current data.

[0129] Step 704 : Calculate the target voltage adjustment amount based on the optimal voltage data in the optimal electric field parameters and the current voltage data, and calculate the target current adjustment amount based on the optimal current data in the optimal electric field parameters and the current current data.

[0130] Step 706 : Adjust the power supply power based on the target voltage adjustment amount and the target current adjustment amount.

[0131] Among them, the operating parameters of the power supply are adjusted according to the optimized optimal voltage data and optimal current data, so as to ensure the stability of the discharge process and minimize noise.

[0132] Optionally, the current voltage data V of the power supply can be obtained through the real-time monitoring module current And current data I current, the target voltage adjustment is calculated based on the optimal voltage data in the optimal electric field parameters and the current voltage data. The target voltage adjustment is recorded as ΔV, ΔV = V * -V current The target current adjustment amount is calculated based on the optimal current data in the optimal electric field parameters and the current current data. The target current adjustment amount is recorded as ΔI, ΔI=I * -I current Finally, the power supply power can be adjusted based on the target voltage adjustment amount and the target current adjustment amount, thereby ensuring that the output power of the power supply meets the optimization requirements and thus maintains the optimal discharge conditions.

[0133] The present application provides a discharge noise control method, which includes: obtaining current relevant data during high-frequency discharge, inputting the current relevant data into a preset noise prediction model for processing, generating a current noise prediction value, minimizing the current noise prediction value using preset constraints, generating optimal electric field parameters, and adjusting the electrode position and power supply power based on the optimal electric field parameters. This solution uses a multi-layer deep neural network to extract features and model relationships of current relevant data, outputs current noise prediction values, and quickly and accurately provides data support for subsequent electric field optimization; in addition, by using preset constraints and iterative optimization, a set of optimal electric field parameters is generated to reduce the interference of high-frequency noise on users and provide a basis for subsequent real-time electric field adjustment; in addition, by adjusting the electrode position based on the optimal electric field parameters, it is ensured that the electrode spacing reaches the optimization target, and by adjusting the power supply power based on the optimal electric field parameters, it is ensured that the output power of the power supply reaches the optimization requirements, thereby maintaining the optimal discharge conditions.

[0134] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0135] Figure 8 This is a structural block diagram of a discharge noise control device provided in an embodiment of the present application.

[0136] like Figure 8 As shown, the discharge noise control device 800 includes:

[0137] The acquisition module 802 is used to acquire current relevant data during the high-frequency discharge process.

[0138] The first generation module 804 is used to input the current relevant data into the preset noise prediction model for processing to generate the current noise prediction value; wherein, the preset noise prediction model is trained based on the relevant data sample set during the high-frequency discharge process, and the relevant data sample set includes a discharge noise signal sample set, an electric field distribution data sample set and an environmental data sample set. The preset noise prediction model is constructed based on a multi-layer deep neural network architecture.

[0139] The second generation module 806 is used to minimize the current noise prediction value using preset constraints to generate optimal electric field parameters; wherein the preset constraints include electrode spacing constraints, voltage preset conditions, current constraints and environmental safety constraints.

[0140] The adjustment module 808 is used to adjust the electrode position and power supply power based on the optimal electric field parameters.

[0141] Regarding the apparatus in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method and will not be elaborated on here. The various modules in the above graph-based retrieval enhancement generation apparatus can be implemented in whole or in part by software, hardware, or a combination thereof. The above modules can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory of the computer device in software form, so that the processor can call and execute the operations of the above modules.

[0142] In one embodiment of the present application, a computer device is provided. The computer device includes a memory and a processor. The memory stores a computer program. When the processor executes the computer program, the following steps are implemented:

[0143] Obtain current relevant data during high-frequency discharge;

[0144] Inputting the current relevant data into a preset noise prediction model for processing to generate a current noise prediction value; wherein the preset noise prediction model is trained based on a relevant data sample set during a high-frequency discharge process, the relevant data sample set including a discharge noise signal sample set, an electric field distribution data sample set, and an environmental data sample set. The preset noise prediction model is constructed using a multi-layer deep neural network architecture;

[0145] The current noise prediction value is minimized using preset constraints to generate optimal electric field parameters; the preset constraints include electrode spacing constraints, voltage preset conditions, current constraints, and environmental safety constraints;

[0146] Based on the optimal electric field parameters, the electrode position and power supply power are adjusted.

[0147] In one embodiment of the present application, when the processor executes the computer program, the processor further implements the following steps:

[0148] Obtaining an input eigenvector matrix based on current relevant data; wherein the current relevant data includes a discharge noise signal, electric field distribution data, and environmental data, the discharge noise signal includes frequency data and sound pressure level data, the electric field distribution data includes voltage data and current data, and the environmental data includes ambient temperature data, ambient humidity data, and air composition data;

[0149] The input feature vector matrix is received through the input layer of the multi-layer deep neural network and distributed to the hidden layer;

[0150] Extract and transform the input feature vector matrix through a preset number of hidden layers to obtain feature data;

[0151] The feature data is processed by the output layer using a linear activation function to generate the current noise prediction value.

[0152] In one embodiment of the present application, when the processor executes the computer program, the processor further implements the following steps:

[0153] Under preset constraints, the current electric field parameters are iteratively solved based on a preset genetic algorithm to generate optimal electric field parameters.

[0154] In one embodiment of the present application, when the processor executes the computer program, the processor further implements the following steps:

[0155] generating an initial population; wherein the initial population includes a plurality of random samples, each random sample corresponds to a set of initial electric field distribution parameters, and the initial electric field distribution parameters satisfy preset constraints;

[0156] Repeat the steps of selecting individuals whose current fitness meets the preset fitness threshold for crossover and mutation operations to generate the next generation population until the preset termination condition is met; wherein the preset termination condition includes reaching the preset maximum number of iterations or the current noise prediction value reaching convergence, and the current fitness is opposite to the current noise prediction value.

[0157] In one embodiment of the present application, when the processor executes the computer program, the processor further implements the following steps:

[0158] Selecting an initial electric field distribution parameter group to be operated corresponding to an individual whose current fitness meets a preset fitness threshold;

[0159] A preset simulated binary crossover algorithm is used to randomly select two individuals to perform crossover recombination on the initial electric field distribution parameters to be operated, thereby generating a new electric field distribution parameter group;

[0160] The new electric field distribution parameter group is randomly perturbed in the offspring population to obtain the next generation population.

[0161] In one embodiment of the present application, when the processor executes the computer program, the processor further implements the following steps:

[0162] Get the current electrode spacing;

[0163] Calculating a target adjustment amount of the electrode position based on the optimal electrode spacing in the optimal electric field parameters and the current electrode spacing;

[0164] Based on the target adjustment amount, the electrode position is adjusted.

[0165] In one embodiment of the present application, when the processor executes the computer program, the processor further implements the following steps:

[0166] Get current voltage data and current current data;

[0167] Calculate the target voltage adjustment amount based on the optimal voltage data in the optimal electric field parameters and the current voltage data, and calculate the target current adjustment amount based on the optimal current data in the optimal electric field parameters and the current current data;

[0168] The power of the power supply is adjusted based on the target voltage adjustment amount and the target current adjustment amount.

[0169] The computer device provided in the embodiment of the present application has similar implementation principles and technical effects to those of the above-mentioned method embodiment, and will not be described in detail here.

[0170] In one embodiment of the present application, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0171] Obtain current relevant data during high-frequency discharge;

[0172] Inputting the current relevant data into a preset noise prediction model for processing to generate a current noise prediction value; wherein the preset noise prediction model is trained based on a relevant data sample set during a high-frequency discharge process, the relevant data sample set including a discharge noise signal sample set, an electric field distribution data sample set, and an environmental data sample set. The preset noise prediction model is constructed using a multi-layer deep neural network architecture;

[0173] The current noise prediction value is minimized using preset constraints to generate optimal electric field parameters; the preset constraints include electrode spacing constraints, voltage preset conditions, current constraints, and environmental safety constraints;

[0174] Based on the optimal electric field parameters, the electrode position and power supply power are adjusted.

[0175] In one embodiment of the present application, when the computer program is executed by a processor, the following steps are further implemented:

[0176] Obtaining an input eigenvector matrix based on current relevant data; wherein the current relevant data includes a discharge noise signal, electric field distribution data, and environmental data, the discharge noise signal includes frequency data and sound pressure level data, the electric field distribution data includes voltage data and current data, and the environmental data includes ambient temperature data, ambient humidity data, and air composition data;

[0177] The input feature vector matrix is received through the input layer of the multi-layer deep neural network and distributed to the hidden layer;

[0178] Extract and transform the input feature vector matrix through a preset number of hidden layers to obtain feature data;

[0179] The feature data is processed by the output layer using a linear activation function to generate the current noise prediction value.

[0180] In one embodiment of the present application, when the computer program is executed by a processor, the following steps are further implemented:

[0181] Under preset constraints, the current electric field parameters are iteratively solved based on a preset genetic algorithm to generate optimal electric field parameters.

[0182] In one embodiment of the present application, when the computer program is executed by a processor, the following steps are further implemented:

[0183] generating an initial population; wherein the initial population includes a plurality of random samples, each random sample corresponds to a set of initial electric field distribution parameters, and the initial electric field distribution parameters satisfy preset constraints;

[0184] Repeat the steps of selecting individuals whose current fitness meets the preset fitness threshold for crossover and mutation operations to generate the next generation population until the preset termination condition is met; wherein the preset termination condition includes reaching the preset maximum number of iterations or the current noise prediction value reaching convergence, and the current fitness is opposite to the current noise prediction value.

[0185] In one embodiment of the present application, when the computer program is executed by a processor, the following steps are further implemented:

[0186] Selecting an initial electric field distribution parameter group to be operated corresponding to an individual whose current fitness meets a preset fitness threshold;

[0187] A preset simulated binary crossover algorithm is used to randomly select two individuals to perform crossover recombination on the initial electric field distribution parameters to be operated, thereby generating a new electric field distribution parameter group;

[0188] The new electric field distribution parameter group is randomly perturbed in the offspring population to obtain the next generation population.

[0189] In one embodiment of the present application, when the computer program is executed by a processor, the following steps are further implemented:

[0190] Get the current electrode spacing;

[0191] Calculating a target adjustment amount of the electrode position based on the optimal electrode spacing in the optimal electric field parameters and the current electrode spacing;

[0192] Based on the target adjustment amount, the electrode position is adjusted.

[0193] In one embodiment of the present application, when the computer program is executed by a processor, the following steps are further implemented:

[0194] Get current voltage data and current current data;

[0195] Calculate the target voltage adjustment amount based on the optimal voltage data in the optimal electric field parameters and the current voltage data, and calculate the target current adjustment amount based on the optimal current data in the optimal electric field parameters and the current current data;

[0196] The power of the power supply is adjusted based on the target voltage adjustment amount and the target current adjustment amount.

[0197] The computer-readable storage medium provided in this embodiment has similar implementation principles and technical effects to those of the above-mentioned method embodiment, and will not be described in detail here.

[0198] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, which can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0199] It is easy to understand that those skilled in the art can combine, split, reorganize, etc. the embodiments of the present application based on the several embodiments provided in the present application to obtain other embodiments, and these embodiments do not exceed the scope of protection of the present application.

[0200] The above specific implementation methods further explain in detail the purpose, technical solutions and beneficial effects of the embodiments of the present application. It should be understood that the above are only specific implementation methods of the embodiments of the present application and are not intended to limit the scope of protection of the embodiments of the present application. Any modifications, equivalent replacements, improvements, etc. made on the basis of the technical solutions of the embodiments of the present application should be included in the scope of protection of the embodiments of the present application.

Claims

1. A discharge noise control method, characterized in that: The method comprises: Obtain current relevant data during high-frequency discharge; Inputting the current relevant data into a preset noise prediction model for processing to generate a current noise prediction value; wherein the preset noise prediction model is trained based on a relevant data sample set during a high-frequency discharge process, the relevant data sample set including a discharge noise signal sample set, an electric field distribution data sample set, and an environmental data sample set, and the preset noise prediction model is constructed using a multi-layer deep neural network architecture; Minimizing the current noise prediction value using preset constraints to generate optimal electric field parameters; wherein the preset constraints include electrode spacing constraints, voltage preset conditions, current constraints, and environmental safety constraints; Based on the optimal electric field parameters, the electrode positions and power supply power are adjusted.

2. The method according to claim 1, characterized in that The step of inputting the current relevant data into a preset noise prediction model for processing to generate a current noise prediction value includes: Obtaining an input eigenvector matrix obtained based on the current relevant data; wherein the current relevant data includes a discharge noise signal, electric field distribution data, and environmental data, the discharge noise signal includes frequency data and sound pressure level data, the electric field distribution data includes voltage data and current data, and the environmental data includes ambient temperature data, ambient humidity data, and air composition data; The input feature vector matrix is received through the input layer of the multi-layer deep neural network and distributed to the hidden layer; Performing feature extraction and transformation on the input feature vector matrix through a preset number of hidden layers to obtain feature data; The characteristic data is processed by an output layer using a linear activation function to generate the current noise prediction value.

3. The method according to claim 1 or 2, characterized in that The current noise prediction value includes a current electric field parameter, and the minimizing process of the current noise prediction value using a preset constraint condition to generate an optimal electric field parameter includes: The current electric field parameters are iteratively solved based on a preset genetic algorithm under preset constraint conditions to generate optimal electric field parameters.

4. The method according to claim 3, characterized in that The iteratively solving the current electric field parameters based on a preset genetic algorithm under preset constraints to generate optimal electric field parameters includes: Generate an initial population; wherein the initial population includes a plurality of random samples, each of the random samples corresponds to a set of initial electric field distribution parameters, and the initial electric field distribution parameters satisfy the preset constraint conditions; Repeat the steps of selecting individuals whose current fitness meets a preset fitness threshold to perform crossover and mutation operations to generate the next generation population until a preset termination condition is met; wherein the preset termination condition includes reaching a preset maximum number of iterations or the current noise prediction value reaching convergence, and the current fitness is opposite to the current noise prediction value.

5. The method according to claim 4, characterized in that The process of selecting individuals whose current fitness meets a preset fitness threshold to perform crossover and mutation operations to generate the next generation population includes: Selecting an initial electric field distribution parameter group to be operated corresponding to an individual whose current fitness meets a preset fitness threshold; Using a preset simulated binary crossover algorithm, two individuals are randomly selected to perform crossover recombination on the initial electric field distribution parameters to be operated, thereby generating a new electric field distribution parameter group; The new electric field distribution parameter group is randomly perturbed in the offspring population to obtain the next generation population.

6. The method according to claim 1 or 2, characterized in that The process of adjusting the electrode position includes: Get the current electrode spacing; Calculating a target adjustment amount of the electrode position based on the optimal electrode spacing in the optimal electric field parameters and the current electrode spacing; The electrode position is adjusted based on the target adjustment amount.

7. The method according to claim 1 or 2, characterized in that The power supply adjustment process includes: Get current voltage data and current current data; Calculating a target voltage adjustment amount based on the optimal voltage data in the optimal electric field parameters and the current voltage data, and calculating a target current adjustment amount based on the optimal current data in the optimal electric field parameters and the current current data; The power supply power is adjusted based on the target voltage adjustment amount and the target current adjustment amount.

8. A discharge noise control device, characterized in that: The device comprises: An acquisition module is used to obtain current relevant data during the high-frequency discharge process; A first generation module is configured to input the current relevant data into a preset noise prediction model for processing to generate a current noise prediction value; wherein the preset noise prediction model is trained based on a relevant data sample set during a high-frequency discharge process, the relevant data sample set including a discharge noise signal sample set, an electric field distribution data sample set, and an environmental data sample set, and the preset noise prediction model is constructed using a multi-layer deep neural network architecture; a second generation module, configured to minimize the current noise prediction value using preset constraints to generate optimal electric field parameters; wherein the preset constraints include electrode spacing constraints, voltage preset conditions, current constraints, and environmental safety constraints; The adjustment module is used to adjust the electrode position and power supply power based on the optimal electric field parameters.

9. An electronic device, characterized in that: The electronic device includes a processor and a memory, wherein the memory stores at least one instruction, at least one program, a code set, or an instruction set, and the at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by the processor to implement the discharge noise control method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The storage medium stores at least one instruction, at least one program, a code set, or an instruction set, and the at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by the processor to implement the discharge noise control method according to any one of claims 1 to 7.

Images