A power plant noise simulation prediction analysis method

Abstract

The application provides a power plant noise simulation prediction analysis method, and belongs to the technical field of analysis and prediction, comprising: obtaining noise data, environment data and operation data of all equipment of the power plant, obtaining equipment distribution and spatial layout of the power plant; analyzing the noise data to determine a sound source characteristic matrix, constructing a noise propagation model based on the sound source characteristic matrix, the equipment distribution and the spatial layout of the power plant; analyzing the operation data to determine a main feature vector related to a noise level, constructing a noise prediction model based on the main feature vector, the environment data and the noise propagation model; simulating to determine a real-time noise prediction result based on the noise prediction model, and formulating a noise control strategy based on the real-time noise prediction result. Potential noise can be identified, the noise prediction and control accuracy can be improved, the equipment operation and the deployment of sound insulation measures can be dynamically adjusted, the noise pollution can be effectively reduced, the accurate noise prediction and control under the complex environment of the power plant are realized, and the environmental protection and safety level of the power plant is improved.

Description

Technical Field

[0001] This invention relates to the field of analysis and prediction technology, and in particular to a method for simulating and predicting power plant noise. Background Technology

[0002] Existing power plant noise methods often focus on simulating and predicting single noise sources or simplified models, typically neglecting the complex influence of environmental conditions and equipment operating status on noise levels, resulting in low prediction accuracy. Furthermore, some technologies lack in-depth analysis of the correlation between noise and equipment operating characteristics, making it difficult to optimize noise control strategies specifically. In addition, noise prediction models cannot adapt to the dynamic changes in the operating status of power plant equipment. These limitations in noise propagation simulation and accurate prediction make it difficult to meet the needs of noise control.

[0003] Therefore, the present invention provides a method for simulating and predicting power plant noise. Summary of the Invention

[0004] This invention provides a method for simulating and predicting noise in power plants. By analyzing noise data, equipment distribution, and spatial layout of the power plant, a noise propagation model is constructed. By analyzing operational data and environmental data and combining them with the noise propagation model, a noise prediction model is constructed. Based on the noise prediction model, simulations are performed to determine the real-time noise prediction results, and noise control strategies are formulated. This method can identify potential noise, improve the accuracy of noise prediction and control, dynamically adjust equipment operation and deploy sound insulation measures, effectively reduce noise pollution, and achieve accurate noise prediction and control in complex power plant environments, thereby improving the environmental protection and safety level of power plants.

[0005] This invention provides a method for simulating and predicting power plant noise, comprising:

[0006] 101: Obtain noise data, environmental data, and operational data of all equipment in the power plant, and obtain the equipment distribution and spatial layout of the power plant;

[0007] 102: Analyze noise data to determine the sound source feature matrix, and construct a noise propagation model based on the sound source feature matrix, the power plant's equipment distribution, and spatial layout;

[0008] 103: Analyze operational data to determine the main feature vectors related to noise levels, and construct a noise prediction model based on the main feature vectors, environmental data, and noise propagation models;

[0009] 104: Based on the noise prediction model, simulate and determine the real-time noise prediction results, and formulate noise control strategies based on the real-time noise prediction results.

[0010] According to the present invention, a method for simulating and predicting power plant noise is provided, which acquires noise data and environmental data of all equipment in a power plant, including:

[0011] Based on the first sensor group, acquire sub-noise data and corresponding sub-environment data of each device in the power plant within a specified time period;

[0012] All sub-noise data and corresponding sub-environment data are preprocessed. Based on the preprocessed sub-noise data, the power plant's noise data and environmental data are determined.

[0013] The second sensor group acquires noise-related sub-operational data for each device within a specified time period, and the operation data is determined based on the preprocessed sub-operational data of all devices.

[0014] According to the present invention, a method for simulating and predicting power plant noise is provided, which analyzes noise data to determine the sound source feature matrix, and constructs a noise propagation model based on the sound source feature matrix, the equipment distribution and spatial layout of the power plant, including:

[0015] Spectral analysis is performed on the preprocessed sub-noise data of each device in the noise data to determine multiple sub-frequency domain features of each device. At the same time, time domain features are extracted from the preprocessed sub-noise data of each device in the noise data to determine multiple sub-time domain features of each device.

[0016] Based on all sub-frequency domain features and all sub-time domain features of each device, determine the sound source feature vector of each device;

[0017] The sound source feature matrix of the power plant is determined based on the sound source feature vectors of all devices;

[0018] The sub-noise propagation path and sub-noise propagation attenuation of each device are determined based on the equipment distribution and spatial layout of the power plant.

[0019] A noise propagation model is constructed based on the sub-noise propagation paths of all devices, the adjusted sub-noise propagation attenuation of all devices, and the sound source feature matrix.

[0020] According to the present invention, a method for simulating and predicting power plant noise is provided, which analyzes operating data to determine the main feature vectors related to noise levels, including:

[0021] For each piece of operational data in the sub-operational data of each device, a status label and a noise level label are applied. Based on the status label, the sub-operational data of each device is classified into states to determine the state operational data. The state operational data includes multiple sub-state operational data.

[0022] Feature extraction is performed on the sub-state operation data of each device in each state to determine the operation feature vector of the sub-state operation data of each device in each state;

[0023] Based on all noise level labels in the sub-state operating data of each device for each state, determine the fitted noise level of the sub-state operating data of each device for each state;

[0024] Based on the operational feature vectors of all sub-state operational data under all states of all devices, calculate the correlation value between each feature and the noise level.

[0025] The relevant feature threshold is determined based on the number of features in the running feature vector and the density of all relevant values.

[0026] Sort all features in the running feature vector by the correlation values ​​between the features and the noise level from largest to smallest, and select the top 10 most relevant features after sorting as the main features related to the noise level.

[0027] The principal feature vector is determined based on all selected principal features.

[0028] According to the power plant noise simulation and prediction analysis method provided by the present invention, based on the operating feature vector of all equipment under all states and all sub-state operating data, the correlation value between each feature and the noise level is calculated, including:

[0029]

[0030]

[0031] in, N1 represents the first sub-correlation value of the fitted noise level between the p-th feature value and the sub-state operation data of the i-th device, where N1 represents the number of sub-state operation data. This represents the p-th feature value in the running feature vector of the j-th sub-state running data of the i-th device. Y represents the p-th average eigenvalue of the i-th device in the runtime feature vector of all sub-state runtime data. ij This represents the fitting noise level of the j-th sub-state operating data of the i-th device. This represents the average noise level of all sub-state operation data of the i-th device. σY represents the standard deviation of the p-th eigenvalue in the operating feature vector of all sub-state operating data of all devices. ij This represents the standard deviation of the fitted noise level for all sub-state operating data of all devices. This represents the second sub-correlation value of the p-th feature value in the running data of the i-th device and the k-th device in the j-th sub-state, where N2 represents the number of devices. This represents the p-th average eigenvalue in the runtime feature vector of all devices running in the j-th sub-state. This represents the p-th feature value in the running feature vector of the j-th sub-state running data of the k-th device. R represents the second correlation value of the i-th device and all other devices except the i-th device. p This represents the correlation between the p-th eigenvalue in the running feature vector and the noise level.

[0032] According to the present invention, a method for simulating and predicting power plant noise is provided, which constructs a noise prediction model based on key feature vectors, environmental data, and a noise propagation model, including:

[0033] Feature extraction is performed on all preprocessed sub-environmental data in the environmental data to determine the sub-environmental vector of each preprocessed sub-environmental data.

[0034] Cluster analysis is performed on the sub-environment vectors of all sub-environment data to determine each cluster in the cluster analysis results as a sub-category, and the environment category is determined based on all sub-categories;

[0035] The sub-category vector for each sub-category is determined based on the sub-environment vectors of all sub-environment data in the cluster corresponding to each sub-category;

[0036] Based on the preprocessed sub-environment data corresponding to the preprocessed sub-running data in the input running data, determine the sub-category corresponding to the preprocessed sub-environment data;

[0037] The sub-noise propagation attenuation of each device is adjusted based on the sub-category vector corresponding to the sub-environment data;

[0038] The noise propagation model is optimized based on the adjusted sub-noise propagation attenuation.

[0039] The main feature vectors and the optimized noise propagation model are input into the noise prediction model, and the noise prediction model is trained based on the running data and noise data.

[0040] According to the present invention, a method for simulating and predicting power plant noise is provided, which determines real-time noise prediction results based on a noise prediction model, and formulates noise control strategies based on the real-time noise prediction results, including:

[0041] Real-time acquisition of real-time environmental data, real-time operational data, and real-time noise data from the power plant;

[0042] Real-time environmental data and real-time operational data are input into the noise prediction model to simulate the actual operating conditions of the power plant.

[0043] The real-time noise prediction result is determined based on the output of the noise prediction model.

[0044] According to the power plant noise simulation and prediction analysis method provided by the present invention, the method determines the real-time noise prediction result based on a noise prediction model, and formulates a noise control strategy based on the real-time noise prediction result, and further includes:

[0045] The noise prediction model is evaluated and optimized by analyzing the consistency between the real-time noise prediction results and the real-time noise data collected in real time.

[0046] Noise control strategies are developed based on the optimized noise prediction model.

[0047] Compared with the prior art, the beneficial effects of this application are as follows:

[0048] By analyzing power plant noise data, equipment distribution, and spatial layout, a noise propagation model is constructed. By analyzing operational and environmental data and combining them with the noise propagation model, a noise prediction model is built. Based on the noise prediction model, simulations are performed to determine the real-time noise prediction results, and noise control strategies are formulated. This approach can identify potential noise, improve the accuracy of noise prediction and control, dynamically adjust equipment operation and deploy sound insulation measures, effectively reduce noise pollution, and achieve accurate noise prediction and control in the complex environment of power plants, thereby improving the environmental protection and safety level of power plants. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0050] Figure 1 This is a flowchart illustrating a power plant noise simulation and prediction analysis method provided in an embodiment of the present invention. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0052] Example 1:

[0053] This invention provides a method for simulating and predicting power plant noise, such as... Figure 1 As shown, it includes:

[0054] 101: Obtain noise data, environmental data, and operational data of all equipment in the power plant, and obtain the equipment distribution and spatial layout of the power plant;

[0055] 102: Analyze noise data to determine the sound source feature matrix, and construct a noise propagation model based on the sound source feature matrix, the power plant's equipment distribution, and spatial layout;

[0056] 103: Analyze operational data to determine the main feature vectors related to noise levels, and construct a noise prediction model based on the main feature vectors, environmental data, and noise propagation models;

[0057] 104: Based on the noise prediction model, simulate and determine the real-time noise prediction results, and formulate noise control strategies based on the real-time noise prediction results.

[0058] In this embodiment, the equipment distribution and spatial layout clearly define the location of each piece of equipment within the power plant and the overall spatial layout of the power plant, in order to simulate the noise propagation path and impact range.

[0059] In this embodiment, the noise propagation model is based on the location and spatial layout of equipment within the power plant, combined with environmental conditions, to simulate the propagation of noise from the sound source outward. This model is used to predict the propagation effect of noise at different locations, taking into account factors such as spatial attenuation.

[0060] In this embodiment, noise data includes sub-noise data of all equipment in the power plant; environmental data includes environmental data that affects noise propagation, such as temperature, humidity, and wind speed; and operational data includes operational data of all equipment in the power plant, such as operating status, power output, and rotational speed. These information are closely related to the noise level.

[0061] In this embodiment, a noise prediction model is used to perform real-time simulation and prediction to obtain the noise level of the power plant under the current operating state and environmental conditions.

[0062] The beneficial effects of the above technical solution are as follows: By analyzing the noise data, equipment distribution, and spatial layout of the power plant, a noise propagation model is constructed. By analyzing the operating data and environmental data and combining them with the noise propagation model, a noise prediction model is constructed. Based on the noise prediction model, simulations are performed to determine the real-time noise prediction results, and noise control strategies are formulated. This can identify potential noise, improve the accuracy of noise prediction and control, dynamically adjust equipment operation and deploy sound insulation measures, effectively reduce noise pollution, achieve accurate noise prediction and control in the complex environment of the power plant, and improve the environmental protection and safety level of the power plant.

[0063] Example 2:

[0064] This invention provides a method for simulating and predicting power plant noise, which acquires noise data and environmental data from all equipment in the power plant, including:

[0065] Based on the first sensor group, acquire sub-noise data and corresponding sub-environment data of each device in the power plant within a specified time period;

[0066] All sub-noise data and corresponding sub-environment data are preprocessed. Based on the preprocessed sub-noise data, the power plant's noise data and environmental data are determined.

[0067] The second sensor group acquires noise-related sub-operational data for each device within a specified time period, and the operation data is determined based on the preprocessed sub-operational data of all devices.

[0068] In this embodiment, the first sensor group represents a set of sensors used to collect device noise and environmental information, including noise sensors and environmental sensors, etc.

[0069] In this embodiment, sub-noise data represents the noise data generated by each device within a specified time period.

[0070] In this embodiment, the sub-environmental data represents the environmental data of each device within a specified time period, such as environmental factors that affect noise propagation, such as temperature and humidity.

[0071] In this embodiment, preprocessing refers to operations such as cleaning, smoothing, and denoising the data to improve data quality and ensure the accuracy of subsequent analysis.

[0072] In this embodiment, the noise data represents the overall noise information of the power plant obtained after preprocessing and integrating all sub-noise data, representing the overall noise level within the power plant.

[0073] In this embodiment, the environmental data represents the overall environmental information of the power plant obtained after preprocessing and integrating all sub-environmental data, which is used to describe the environmental conditions inside the power plant.

[0074] In this embodiment, the second sensor group refers to a sensor group used to collect the operating status of the device, such as a speed sensor, which is closely related to noise generation.

[0075] In this embodiment, sub-operation data represents the operating status data of each device within a specified time period, including the device's load, speed, temperature, etc., which affect the noise generation of the device.

[0076] In this embodiment, the operating data represents the overall operating status of the power plant obtained after preprocessing and integrating all sub-operating data, and is used for noise prediction and control analysis.

[0077] The beneficial effects of the above technical solution are: acquiring noise data and environmental data of all equipment in the power plant enables detailed monitoring and response to noise sources of each device, providing a more efficient data foundation for noise management.

[0078] Example 3:

[0079] This invention provides a method for simulating and predicting power plant noise, which analyzes noise data to determine a sound source feature matrix, and constructs a noise propagation model based on the sound source feature matrix, the equipment distribution and spatial layout of the power plant, including:

[0080] Spectral analysis is performed on the preprocessed sub-noise data of each device in the noise data to determine multiple sub-frequency domain features of each device. At the same time, time domain features are extracted from the preprocessed sub-noise data of each device in the noise data to determine multiple sub-time domain features of each device.

[0081] Based on all sub-frequency domain features and all sub-time domain features of each device, determine the sound source feature vector of each device;

[0082] The sound source feature matrix of the power plant is determined based on the sound source feature vectors of all devices;

[0083] The sub-noise propagation path and sub-noise propagation attenuation of each device are determined based on the equipment distribution and spatial layout of the power plant.

[0084] A noise propagation model is constructed based on the sub-noise propagation paths of all devices, the adjusted sub-noise propagation attenuation of all devices, and the sound source feature matrix.

[0085] In this embodiment, the frequency domain features represent the noise signal characteristics obtained by analyzing in the frequency domain, including main frequency components, frequency peaks, frequency band energy, etc., which describe the frequency distribution of the noise signal.

[0086] In this embodiment, time-domain features represent the noise signal features obtained by analyzing in the time domain, such as mean, variance, peak value, etc., which describe the changes of the signal over time.

[0087] In this embodiment, the sound source feature vector represents a vector formed by integrating frequency domain features and time domain features, which is used to describe the sound source characteristics of each device.

[0088] In this embodiment, the sound source feature matrix represents a matrix containing the feature vectors of all device sound sources, which is the feature dataset of the noise sources of the entire power plant.

[0089] In this embodiment, the sub-noise propagation path represents the propagation route of noise from each device within the power plant space, which is affected by the device location and the power plant layout.

[0090] In this embodiment, sub-noise propagation attenuation represents the amount of energy loss of sound waves during propagation due to distance and obstruction factors, which is usually related to propagation distance, medium and obstacles.

[0091] In this embodiment, the noise propagation model represents a model constructed based on the sound source feature matrix, propagation path, and attenuation characteristics, used to simulate the propagation mode and attenuation of noise in a power plant.

[0092] The beneficial effects of the above technical solution are as follows: constructing a noise propagation model based on the sound source feature matrix, the equipment distribution and spatial layout of the power plant can enhance the understanding and prediction accuracy of noise propagation, provide a more scientific basis for noise control in power plants, and effectively improve the noise control effect.

[0093] Example 4:

[0094] This invention provides a method for simulating and predicting power plant noise, which analyzes operating data to determine the main feature vectors related to noise levels, including:

[0095] For each piece of operational data in the sub-operational data of each device, a status label and a noise level label are applied. Based on the status label, the sub-operational data of each device is classified into states to determine the state operational data. The state operational data includes multiple sub-state operational data.

[0096] Feature extraction is performed on the sub-state operation data of each device in each state to determine the operation feature vector of the sub-state operation data of each device in each state;

[0097] Based on all noise level labels in the sub-state operating data of each device for each state, determine the fitted noise level of the sub-state operating data of each device for each state;

[0098] Based on the operational feature vectors of all sub-state operational data under all states of all devices, calculate the correlation value between each feature and the noise level.

[0099] The relevant feature threshold is determined based on the number of features in the running feature vector and the density of all relevant values.

[0100] Sort all features in the running feature vector by the correlation values ​​between the features and the noise level from largest to smallest, and select the top 10 most relevant features after sorting as the main features related to the noise level.

[0101] The principal feature vector is determined based on all selected principal features.

[0102] In this embodiment, the status tag represents the device status label tagged on each piece of running data, such as "idle", "running", "fault", etc., which describes the current operating status of the device.

[0103] In this embodiment, the noise level marker indicates the noise level corresponding to each piece of running data, which is used to analyze the changes in noise.

[0104] In this embodiment, state classification means classifying the sub-operational data of the device according to state labels to form a set of operation data for each state.

[0105] In this embodiment, the state operation data represents the overall operation data set of a device in all states, including multiple sub-state operation data.

[0106] In this embodiment, the sub-state operation data represents the operation data of a device in a certain state, which facilitates the extraction of state features.

[0107] In this embodiment, the running feature vector represents the feature set extracted from the sub-state running data of a device in a certain state, and each state of each device corresponds to a running feature vector.

[0108] In this embodiment, the fitted noise level represents a representative noise level value determined based on all noise level markers of the corresponding device in the corresponding state, which is used to describe the noise situation of the device in that state.

[0109] In this embodiment, the relevant feature threshold is a value set according to the number and density of relevant values, used to filter the features most relevant to the noise level.

[0110] In this embodiment, the key features refer to the features most closely related to the noise level among all features, which are used for subsequent analysis.

[0111] In this embodiment, the principal feature vector represents a vector composed of all principal features, representing the set of features that have a key impact on the noise level.

[0112] The beneficial effects of the above technical solution are: analyzing operational data to determine the main feature vectors related to noise levels, identifying key operational factors affecting noise, and providing a scientific basis for precise noise reduction and optimization of equipment operation.

[0113] Example 5:

[0114] This invention provides a method for simulating and predicting power plant noise, which calculates the correlation value between each feature and the noise level based on the operating feature vector of all equipment under all states and all sub-state operating data, including:

[0115]

[0116] in, N1 represents the first sub-correlation value of the fitted noise level between the p-th feature value and the sub-state operation data of the i-th device, where N1 represents the number of sub-state operation data. This represents the p-th feature value in the running feature vector of the j-th sub-state running data of the i-th device. Y represents the p-th average eigenvalue of the i-th device in the runtime feature vector of all sub-state runtime data. ij This represents the fitting noise level of the j-th sub-state operating data of the i-th device. This represents the average noise level of all sub-state operation data of the i-th device. σY represents the standard deviation of the p-th eigenvalue in the operating feature vector of all sub-state operating data of all devices. ij This represents the standard deviation of the fitted noise level for all sub-state operating data of all devices. This represents the second sub-correlation value of the p-th feature value in the running data of the i-th device and the k-th device in the j-th sub-state, where N2 represents the number of devices. This represents the p-th average eigenvalue in the runtime feature vector of all devices running in the j-th sub-state. This represents the p-th feature value in the running feature vector of the j-th sub-state running data of the k-th device. R represents the second correlation value of the i-th device and all other devices except the i-th device. p This represents the correlation between the p-th eigenvalue in the running feature vector and the noise level.

[0117] In this embodiment, the correlation value represents the correlation between the corresponding feature in the running feature vector and all noise levels of all devices.

[0118] In this embodiment, This represents the correlation value between the p-th feature value and the noise level in all sub-state operation data of the i-th device and all other devices excluding the i-th device.

[0119] In this embodiment, This represents the combined second correlation value among all devices.

[0120] The beneficial effects of the above technical solution are as follows: Based on the operating feature vectors of all sub-state operating data under all states of all devices, the correlation value between each feature and the noise level can be calculated, which can improve the data basis for screening the main features and determining the main feature vectors.

[0121] Example 6:

[0122] This invention provides a method for simulating and predicting power plant noise, which constructs a noise prediction model based on key feature vectors, environmental data, and a noise propagation model, including:

[0123] Feature extraction is performed on all preprocessed sub-environmental data in the environmental data to determine the sub-environmental vector of each preprocessed sub-environmental data.

[0124] Cluster analysis is performed on the sub-environment vectors of all sub-environment data to determine each cluster in the cluster analysis results as a sub-category, and the environment category is determined based on all sub-categories;

[0125] The sub-category vector for each sub-category is determined based on the sub-environment vectors of all sub-environment data in the cluster corresponding to each sub-category;

[0126] Based on the preprocessed sub-environment data corresponding to the preprocessed sub-running data in the input running data, determine the sub-category corresponding to the preprocessed sub-environment data;

[0127] The sub-noise propagation attenuation of each device is adjusted based on the sub-category vector corresponding to the sub-environment data;

[0128] The noise propagation model is optimized based on the adjusted sub-noise propagation attenuation.

[0129] The main feature vectors and the optimized noise propagation model are input into the noise prediction model, and the noise prediction model is trained based on the running data and noise data.

[0130] In this embodiment, the sub-environment vector represents the feature vector extracted from the corresponding sub-environment data, describing the multidimensional environmental features of the sub-environment data.

[0131] In this embodiment, a cluster family represents a group in the clustering analysis results, and each cluster family represents a subcategory, reflecting a data set with similar environmental characteristics.

[0132] In this embodiment, the sub-category represents a sub-category with similar environmental characteristics obtained by cluster analysis, which contains a data set with specific environmental characteristics.

[0133] In this embodiment, the environment category is a comprehensive classification of multiple subcategories, used to represent the overall characteristics of the environment.

[0134] In this embodiment, the sub-category vector represents the vector of the feature center of each sub-category, indicating the typical environmental characteristics of that category.

[0135] In this embodiment, the noise propagation model is a mathematical model that simulates the propagation of noise in different environments and spaces, taking into account the noise source, path, and attenuation.

[0136] In this embodiment, the noise prediction model is a model trained based on operating data, environmental data, and an optimized noise propagation model, used to predict the noise level of the equipment.

[0137] The beneficial effects of the above technical solution are as follows: By constructing a noise prediction model based on the main feature vectors, environmental data, and noise propagation model, the noise prediction model can more accurately reflect the impact of environmental changes on noise, providing reliable support for noise prediction and control in complex environments.

[0138] Example 7:

[0139] This invention provides a method for simulating and predicting power plant noise, which involves simulating and determining real-time noise prediction results based on a noise prediction model, and formulating noise control strategies based on the real-time noise prediction results. The method includes:

[0140] Real-time acquisition of real-time environmental data, real-time operational data, and real-time noise data from the power plant;

[0141] Real-time environmental data and real-time operational data are input into the noise prediction model to simulate the actual operating conditions of the power plant.

[0142] The real-time noise prediction result is determined based on the output of the noise prediction model.

[0143] In this embodiment, real-time environmental data includes current environmental parameters such as temperature, humidity, air pressure, and wind speed, which affect the propagation characteristics of noise.

[0144] In this embodiment, the real-time operating data includes the operating parameters of various equipment in the power plant, such as operating status, power, speed, and load, reflecting the working status of the equipment.

[0145] In this embodiment, the real-time noise data includes actual measurements of the current noise level, typically provided by a noise sensor, and is used for subsequent comparison of noise prediction results and model optimization.

[0146] In this embodiment, the collected real-time environmental data and real-time operational data are fed into the noise prediction model. The noise prediction model simulates the noise level of equipment in the power plant under the current environmental conditions based on the trained equipment noise characteristics and environmental noise propagation laws.

[0147] In this embodiment, the real-time noise prediction result is the noise level prediction value output by the noise prediction model under the current conditions, representing the predicted noise situation of the device in the real-time environment and operating state.

[0148] The beneficial effects of the above technical solution are: by simulating and determining the real-time noise prediction results based on the noise prediction model, and by formulating noise control strategies based on the real-time noise prediction results, the noise propagation model can be optimized and the accuracy of the noise prediction model can be improved.

[0149] Example 8:

[0150] This invention provides a method for simulating and predicting power plant noise, which involves determining real-time noise prediction results based on a noise prediction model, formulating noise control strategies based on the real-time noise prediction results, and further includes:

[0151] The noise prediction model is evaluated and optimized by analyzing the consistency between the real-time noise prediction results and the real-time noise data collected in real time.

[0152] Noise control strategies are developed based on the optimized noise prediction model.

[0153] In this embodiment, analyzing the consistency between the real-time noise prediction results of the noise prediction model and the real-time noise data collected in real time includes checking the deviation between the noise prediction results and the actual noise data, analyzing the degree of agreement between the prediction results and the actual noise data under different environments and operating conditions, and identifying the causes of prediction errors, such as the influence of environmental conditions and equipment parameters on noise prediction.

[0154] After determining the prediction error of the model under certain conditions through consistency analysis, the noise prediction model is evaluated and optimized. The optimization process includes parameter tuning: adjusting the parameters in the model according to the characteristics of the noise data and the actual situation to improve the model's adaptability to different environments and operating conditions; data updating: adding new real-time noise data and environmental data for training to enhance the model's generalization ability and adapt it to new operating scenarios; and model structure optimization: optimizing the model structure (such as adding new input features or improving the model algorithm) based on the results of error analysis to improve prediction accuracy.

[0155] Based on the optimized noise prediction model, noise control strategies are formulated. Utilizing the model's predictive capabilities, power plants can take preventative measures against potentially excessive or risky noise conditions. These measures include: equipment operation adjustments (reducing the operating power of certain equipment or changing its operating mode in high-noise-risk situations to reduce noise generation); noise shielding (using sound barriers, sound-absorbing materials, and other measures to reduce noise propagation during peak noise periods or high-risk areas); and dynamic control (dynamically adjusting equipment operating parameters or using noise control devices based on real-time predicted noise levels to keep noise levels within safe limits).

[0156] The beneficial effects of the above technical solution are as follows: By analyzing the noise data, equipment distribution, and spatial layout of the power plant, a noise propagation model is constructed. By analyzing the operating data and environmental data and combining them with the noise propagation model, a noise prediction model is constructed. Based on the noise prediction model, simulations are performed to determine the real-time noise prediction results, and noise control strategies are formulated. This can identify potential noise, improve the accuracy of noise prediction and control, dynamically adjust equipment operation and deploy sound insulation measures, effectively reduce noise pollution, achieve accurate noise prediction and control in the complex environment of the power plant, and improve the environmental protection and safety level of the power plant.

[0157] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0158] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0159] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method of power plant noise simulation prediction analysis, characterized by, The method comprises the following steps: 101: obtaining noise data, environment data and operation data of all equipment in the power plant, obtaining equipment distribution and spatial layout of the power plant; 102: analyzing the noise data to determine the sound source feature matrix, and constructing a noise propagation model based on the sound source feature matrix, the equipment distribution and the spatial layout of the power plant; 103: analyzing the operation data to determine the main feature vector related to the noise level, and constructing a noise prediction model based on the main feature vector, the environment data and the noise propagation model; 104: determining real-time noise prediction results based on the noise prediction model, and formulating a noise control strategy based on the real-time noise prediction results; Obtaining noise data and environment data of all equipment in the power plant comprises: Based on the first sensor group, obtaining sub-noise data and corresponding sub-environment data of each device in the power plant within a specified time period; Preprocessing all sub-noise data and corresponding sub-environment data, determining noise data of the power plant based on all preprocessed sub-noise data, and determining environment data of the power plant based on all preprocessed sub-environment data; Based on the second sensor group, obtaining sub-operation data related to noise of each device within a specified time period, and determining operation data based on preprocessed sub-operation data of all devices; Analyzing the operation data to determine the main feature vector related to the noise level comprises: Marking each piece of operation data in the sub-operation data of each device with a state and a noise level, classifying the sub-operation data of each device based on the state to determine state operation data, wherein the state operation data comprises a plurality of sub-state operation data; Extracting features from the sub-state operation data of each state of each device to determine the operation feature vector of the sub-state operation data of each state of each device; Based on all noise level markers in the sub-state operation data of each state of each device, determining the fitted noise level of the sub-state operation data of each state of each device; Based on the operation feature vectors of all sub-state operation data of all states of all devices, calculating the correlation value of each feature with the noise level; Determining a correlation feature threshold based on the number of features in the operation feature vector and the density of all correlation values; Sorting all correlation values of features in the operation feature vector from large to small, and selecting the top correlation feature threshold feature values as the main features related to the noise level; Determining the main feature vector based on all selected main features.

2. A method of power plant noise simulation prediction analysis according to claim 1, characterized in that, Analyzing the noise data to determine the sound source feature matrix, and constructing a noise propagation model based on the sound source feature matrix, the equipment distribution and the spatial layout of the power plant comprises: Performing frequency spectrum analysis on the preprocessed sub-noise data of each device in the noise data to determine a plurality of sub-frequency domain features of each device, and simultaneously performing time domain feature extraction on the preprocessed sub-noise data of each device in the noise data to determine a plurality of sub-time domain features of each device; Determining the sound source feature vector of each device based on all sub-frequency domain features and all sub-time domain features of each device; Determining the sound source feature matrix of the power plant based on the sound source feature vectors of all devices; Determine sub-noise propagation paths and sub-noise propagation attenuations of each device based on the distribution and spatial layout of the devices in the power plant; Construct a noise propagation model based on the sub-noise propagation paths of all devices, the adjusted sub-noise propagation attenuations of all devices, and the sound source feature matrix.

3. The method of claim 1, wherein, Calculate the correlation value of each feature and the noise level based on the operating feature vectors of all sub-states under all states of all devices, including: ； ； ； wherein, represents a first sub-correlation value of the pth feature value in all sub-state running data of the i th device and the fitting noise level of all sub-state running data of the i th device, N1 represents the number of sub-state running data, represents the pth feature value in the running feature vector of the jth sub-state running data of the i th device, represents the pth average feature value in the running feature vector of all sub-state running data of the i th device, represents the fitting noise level of the jth sub-state running data of the i th device, represents the average noise level of all sub-state running data of the i th device, represents the standard deviation of the pth feature value in the running feature vector of all sub-state running data of all devices, represents the standard deviation of the fitting noise level of all sub-state running data of all devices, represents a second sub-correlation value of the pth feature value in the jth sub-state running data of the i th device and the k th device, N2 represents the number of devices, represents the pth average feature value in the running feature vector of the jth sub-state running data of all devices, represents the pth feature value in the running feature vector of the jth sub-state running data of the k th device, represents a second correlation value of the i th device and all other devices except the i th device, represents a correlation value of the pth feature value in the running feature vector and the noise level.

4. The method of claim 1, wherein, Construct a noise prediction model based on the main feature vector, environmental data, and the noise propagation model, including: Extract features from all pre-processed sub-environmental data in the environmental data to determine the sub-environmental vectors of each pre-processed sub-environmental data; Determine each cluster family in the clustering analysis result as a sub-category based on the clustering analysis of the sub-environmental vectors of all sub-environmental data, and determine the environmental category based on all sub-categories; Determine the sub-category vector of each sub-category based on the sub-environmental vectors of all sub-environmental data in the cluster family corresponding to each sub-category; Determine the sub-category corresponding to the pre-processed sub-environmental data corresponding to the pre-processed sub-operation data in the input operating data; Adjust the sub-noise propagation attenuation of each device based on the sub-category vector of the sub-category corresponding to the sub-environmental data; Optimize the noise propagation model based on the adjusted sub-noise propagation attenuation; Input the main feature vector and the optimized noise propagation model into the noise prediction model, and train the noise prediction model based on the operating data and the noise data.

5. The method of claim 1, wherein, Determine the real-time noise prediction result based on the simulation of the noise prediction model, and develop a noise control strategy based on the real-time noise prediction result, including: Real-time acquisition of real-time environmental data, real-time operating data, and real-time noise data of the power plant; Input the real-time environmental data and real-time operating data into the noise prediction model to simulate the real operating conditions of the power plant; Determine the real-time noise prediction result based on the output result of the noise prediction model.

6. A method of power plant noise simulation prediction analysis according to claim 5, wherein, Determine the real-time noise prediction result based on the simulation of the noise prediction model, and develop a noise control strategy based on the real-time noise prediction result, also including: Analyze the consistency of the real-time noise prediction result of the noise prediction model and the real-time noise data collected in real time, evaluate and optimize the noise prediction model; Develop a noise control strategy based on the optimized noise prediction model.

Images