Environment noise intelligent monitoring method and system

Abstract

The invention provides an environment noise intelligent monitoring method and system, and is applied to a noise monitoring system, the noise monitoring system comprises a gear assembly and a rotating rod, the rotating rod is used for supporting a sampling probe, the gear assembly is used for providing resistance for rotation of the rotating rod, and the method comprises the following steps: responding to monitored noise data; inputting the noise data into a de-noising model for de-noising processing to obtain target monitoring data; wherein the noise data is time series data in a continuous time period, and the denoising model is used for evaluating the abnormal degree of each data point in the noise data based on the time fluctuation condition of the noise data. According to the invention, the sampling probe on the rotating rod can be operated and maintained without climbing, meanwhile, the denoising model is utilized to carry out noise reduction processing on the monitored noise data, noise generated in the operation and maintenance process can be eliminated, monitoring does not need to be paused in the operation and maintenance process, the requirement of continuous environmental noise monitoring is met, and the operation and maintenance efficiency is improved. And the safety of operation and maintenance personnel is ensured.

Description

Technical Field

[0001] The present disclosure relates to the technical field of noise monitoring, and in particular to an intelligent environmental noise monitoring method and system. Background Art

[0002] At present, the relevant construction standards for fixed noise monitoring equipment require that the noise sampling probe in the equipment be no less than 4 meters, and the sampling probe must be operated and maintained at least once a month. In addition, if there is bad weather (for example, windy or rainy), the frequency of operation and maintenance should be appropriately increased.

[0003] Therefore, the sampling probe has a high maintenance frequency, and the operating height limit requires frequent operation of the ladder for maintenance. High-altitude operations are not only dangerous, but also easily damage equipment and tools if they are not operated properly. In addition, during the high-altitude operation and maintenance process, the operation and maintenance personnel need to first turn on the host of the noise monitoring station, and then press the pause monitoring and start operation and maintenance button to put the noise monitoring equipment in a shutdown state and start operation and maintenance. Otherwise, artificial noise interference may be generated during the operation and maintenance process. As mentioned above, the cumbersome operation and maintenance process in high-altitude operations will increase the danger, and such an operation and maintenance method cannot meet the needs of continuous environmental noise monitoring. Summary of the invention

[0004] The present invention provides an intelligent environmental noise monitoring method and system.

[0005] In a first aspect, an embodiment of the present disclosure provides an intelligent environmental noise monitoring method, which is applied to a noise monitoring system, wherein the noise monitoring system includes a gear assembly and a rotating rod, wherein the rotating rod is used to support a sampling probe, and the gear assembly is used to provide resistance to the rotation of the rotating rod. The method includes: in response to monitored noise data, inputting the noise data into a denoising model for noise reduction processing to obtain target monitoring data; wherein the noise data is time series data within a continuous time period, and the denoising model is used to evaluate the degree of abnormality of each data point in the noise data based on the time fluctuation of the noise data.

[0006] In a second aspect, an embodiment of the present disclosure provides a noise monitoring system, which includes: a main pole; a key block, a first end of the key block being clamped with the main pole; a rotating rod, a first end of the rotating rod being rotatably connected with the first end of the main pole, the second end of the key block being inserted into the interior of the rotating rod, and the second end of the rotating rod being used to support a sampling probe; a gear assembly, used to provide resistance to the rotation of the rotating rod; and a host, used to implement the intelligent environmental noise monitoring method described in the first aspect.

[0007] In a third aspect, an embodiment of the present disclosure provides a computer-readable medium, characterized in that a computer program is stored thereon, and when the computer program is executed by a processor, the method for intelligent monitoring of environmental noise described in the first aspect is implemented.

[0008] In the disclosed embodiment, when the sampling probe rotates with the rotating rod, the rotating rod rotates around the first end of the key block to trigger the gear assembly to provide resistance to the rotation of the rotating rod, thereby allowing the sampling probe on the rotating rod to be operated and maintained without climbing up. At the same time, a denoising model is used to perform denoising on the monitored noise data. The denoising model evaluates the degree of abnormality of the noise data in a continuous time period based on the time fluctuation of the noise data, eliminates the noise generated during the operation and maintenance process, and eliminates the need to pause the monitoring process every time during the operation and maintenance process, thereby satisfying the needs of continuous environmental noise monitoring and ensuring the safety of the operation and maintenance personnel. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In the accompanying drawings of the embodiments of the present disclosure: Figure 1 A flowchart of an intelligent environmental noise monitoring method provided in an embodiment of the present disclosure.

[0010] Figure 2 A flowchart of the training process of the denoising model provided in the embodiment of the present disclosure.

[0011] Figure 3 A schematic diagram of the noise reduction process provided in an embodiment of the present disclosure.

[0012] Figure 4 A schematic diagram of the structure of a noise monitoring system provided in an embodiment of the present disclosure.

[0013] Figure 5 A schematic diagram of the structure of a gear assembly provided in an embodiment of the present disclosure.

[0014] Figure 6 A schematic diagram of the mass relationship between the key block, the rotating rod and the sampling probe provided in an embodiment of the present disclosure.

[0015] Figure 7 A schematic diagram of the partial structure of the noise monitoring system provided by an embodiment of the present disclosure when the rotation angle of the rotating rod is 0° to 90° after the key block is removed.

[0016] Figure 8 A schematic diagram of the partial structure of the gear assembly when the rotation angle of the rotating rod is 0° to 90° after the key block is removed according to an embodiment of the present disclosure.

[0017] Fig. 9 A structural schematic diagram of a noise monitoring system provided by an embodiment of the present disclosure when the rotation angle of the rotating rod is 90° after the key block is removed.

[0018] Fig.10 A schematic diagram of the partial structure of the gear assembly when the rotation angle of the rotating rod is 90° after the key block is removed according to an embodiment of the present disclosure.

[0019] Fig.11 A structural schematic diagram of a noise monitoring system provided by an embodiment of the present disclosure when the rotation angle of the rotating rod is 90° to 180° after the key block is removed.

[0020] Fig.12 A schematic diagram of the partial structure of the gear assembly when the rotation angle of the rotating rod is 90° to 180° after the key block is removed according to an embodiment of the present disclosure.

[0021] Fig.13 A structural schematic diagram of a noise monitoring system when restoring the monitoring state of a sampling probe 4 provided in an embodiment of the present disclosure.

[0022] Fig.14 A structural schematic diagram of another noise monitoring system when restoring the monitoring state of the sampling probe 4 provided in an embodiment of the present disclosure.

[0023] Fig.15 A schematic structural diagram of a noise monitoring system after key block installation provided in an embodiment of the present disclosure.

[0024] Reference numerals: 1-main pole; 2-key block; 3-rotating rod; 4-sampling probe; 5-gear assembly; 51-first gear; 511-damping bump; 52-second gear; 53-third gear; 531-damping landslide; 6-host. DETAILED DESCRIPTION

[0025] In order to enable those skilled in the art to better understand the technical solution of the present disclosure, the embodiments of the present disclosure are described in detail below with reference to the accompanying drawings.

[0026] The present disclosure will be described more fully below with reference to the accompanying drawings, but the embodiments shown may be embodied in different forms, and the present disclosure should not be construed as being limited to the embodiments set forth below. On the contrary, the purpose of providing these embodiments is to make the present disclosure thorough and complete, and will enable those skilled in the art to fully understand the scope of the present disclosure.

[0027] The accompanying drawings of the embodiments of the present disclosure are used to provide a further understanding of the embodiments of the present disclosure, and constitute a part of the specification, and are used together with the detailed embodiments to explain the present disclosure, and do not constitute a limitation of the present disclosure. By describing the detailed embodiments with reference to the accompanying drawings, the above and other features and advantages will become more apparent to those skilled in the art.

[0028] The present disclosure may be described with reference to plan views and / or cross-sectional views by means of ideal schematic views of the present disclosure. Therefore, the exemplary illustrations may be modified according to manufacturing techniques and / or tolerances.

[0029] In the absence of conflict, the various embodiments of the present disclosure and the various features therein may be combined with each other.

[0030] The terms used in the present disclosure are only used to describe specific embodiments and are not intended to limit the present disclosure. The term "and / or" as used in the present disclosure includes any and all combinations of one or more related enumerated items. The singular forms "one" and "the" as used in the present disclosure are also intended to include plural forms, unless the context clearly indicates otherwise. The terms "including", "made of..." as used in the present disclosure specify the presence of the features, wholes, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or groups thereof.

[0031] Unless otherwise defined, all terms (including technical and scientific terms) used in this disclosure have the same meaning as commonly understood by those of ordinary skill in the art. It will also be understood that terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant art and this disclosure, and will not be interpreted as having an idealized or overly formal meaning unless expressly defined in this disclosure.

[0032] The present disclosure is not limited to the embodiments shown in the drawings, but includes modifications of the configurations formed based on the manufacturing process. Therefore, the regions illustrated in the drawings have schematic properties, and the shapes of the regions shown in the drawings illustrate the specific shapes of the regions of the elements, but are not intended to be limiting.

[0033] In some related technologies, the construction standards for fixed noise monitoring systems have requirements for the location of the sampling probe in the noise monitoring system, which is usually not less than 4 meters, and there are requirements for the number of operations and maintenance of the sampling probe, which is usually required to be operated and maintained at least once a month. In addition, in case of bad weather (for example, windy or rainy, etc.), the frequency of operation and maintenance will be appropriately increased, that is, the sampling probe has the characteristics of being set higher than the ground and having a high frequency of operation and maintenance. The setting height and operation and maintenance frequency of the sampling probe increase the difficulty of equipment maintenance for the operation and maintenance personnel.

[0034] During the operation and maintenance process, the noise monitoring system is usually required to continuously monitor the noise. However, the noise monitored during the existing operation and maintenance process may contain both environmental noise and artificial noise generated by the operation and maintenance personnel, which may affect the monitoring results. If the operation and maintenance personnel manually suspend the noise monitoring system before starting the operation and maintenance, it may aggravate the dangerous state of high-altitude operations, and the suspension will also cause the collected data to be discontinuous.

[0035] Therefore, there is an urgent need to provide a noise monitoring solution to deal with the risks of high-altitude operations brought about by the existing noise monitoring system and the inability to meet the needs of continuous environmental noise monitoring.

[0036] In a first aspect, an embodiment of the present disclosure provides an intelligent environmental noise monitoring method, which is applied to a host of a noise monitoring system, wherein the noise monitoring system further comprises a gear assembly and a rotating rod, wherein the rotating rod is used to support a sampling probe, and the gear assembly is used to provide resistance to the rotation of the rotating rod, Figure 1 , the method includes step S1.

[0037] S1. In response to monitoring noise data, the noise data is input into a denoising model for noise reduction processing to obtain target monitoring data; wherein the noise data is time series data within a continuous time period, and the denoising model is used to evaluate the degree of abnormality of each data point in the noise data based on the time fluctuation of the noise data.

[0038] The noise data is data collected continuously and in real time by the host of the noise monitoring system according to a time sequence, and the denoising model is a pre-trained model that can at least be used to remove the artificial noise generated by operation and maintenance in the noise data.

[0039] In the embodiments of the present disclosure, a denoising model is used to perform denoising processing on the monitored noise data. The denoising model evaluates the degree of abnormality of the noise data in a continuous time period based on the time fluctuation of the noise data, so as to eliminate abnormal noise points generated during the operation and maintenance process, so that the operation and maintenance process does not need to suspend monitoring every time, thereby meeting the needs of continuous environmental noise monitoring, reducing the tedious operation and maintenance process in high-altitude operations, and ensuring the safety of operation and maintenance personnel.

[0040] In some embodiments, reference Figure 2 ,The denoising model training process includes steps S1 ' to S5'.

[0041] S1', obtaining N training data; each training data includes time series noise data in a continuous time period.

[0042] S2', taking m training data from the N training data according to a preset number of times and inputting them into the first preset model for processing to obtain a first target preset model; 1<m<N.

[0043] S3', inputting the nth training data among the N training data into the first target preset model for first noise reduction processing, and determining the intermediate abnormal data; n is taken as an integer from 1 to N in sequence.

[0044] S4′, inputting the intermediate abnormal data corresponding to the nth training data into the second preset model corresponding to the n-1th training data for second noise reduction processing to obtain the second preset model corresponding to the nth training data.

[0045] S5', determining a denoising model according to the first target preset model and the second preset model corresponding to the Nth training data.

[0046] In an embodiment of the present disclosure, a part of data randomly extracted from the entire set of training data (i.e., N training data), i.e., m training data, is randomly selected multiple times to train the first preset model, which can reduce the computational complexity of the training process. At the same time, since the m training data selected from the N training data each time are usually different, the diversity of the first target preset model after training can be increased through multiple training. In some embodiments, the first preset model is an isolation forest model iForest. The first target preset model after training can effectively identify globally abnormal data points in the training data. The embodiment of the present disclosure does not impose any special restrictions on the value of m, and it can be set according to training requirements.

[0047] Based on the trained first target preset model, take the nth training data from the N training data and input it into the first noise reduction process to obtain intermediate abnormal data, which is the data point of global abnormality in the nth training data identified by the first target preset model. Input the intermediate abnormal data into the second preset model corresponding to the n-1th training data for the second noise reduction process to obtain the second preset model corresponding to the nth training data, input all the N training data into the second preset model for training, and determine the model obtained (i.e., the second preset model corresponding to the Nth training data) as the second target preset model, thereby generating a denoising model according to the first target preset model and the second preset model corresponding to the Nth training data. Among them, the trained second preset model can effectively identify the data points of local abnormality in the training data. In some embodiments, the second preset model is a local abnormality factor model LOF.

[0048] The abnormal data points obtained after the preliminary noise reduction processing of the first target preset model are further monitored in detail using the second target preset model, which can more accurately determine the degree of abnormality, thereby improving the robustness of the monitoring of abnormal data points.

[0049] The disclosed embodiment does not impose any special restrictions on the model type of the second preset model, which may be LOF or a combination of a clustering algorithm and another iForest model.

[0050] In some embodiments, before step S1', the method further includes: Perform data preprocessing on N training data.

[0051] In the embodiment of the present disclosure, the original data matrix corresponding to N training data is calculated by formula (1): Standardization is performed, where n is the number of samples, that is, the total number of training data, and d is the feature dimension.

[0052] (1).

[0053] In formula (1), is the mean vector of each feature, satisfying , is the standard deviation vector of each feature, satisfying , represents element-wise division, is the original data matrix The value of the jth feature in the i-th sample.

[0054] In some embodiments, step S2' comprises: For each data point in the m training data extracted each time, the intermediate features of each data point and its corresponding intermediate split point are selected.

[0055] According to the intermediate features and their corresponding intermediate split points, each data point is divided into two data subsets; For each data point in each data subset, the steps of selecting the intermediate features of each data point and the corresponding intermediate cutting points are repeatedly performed, and each data point is divided into two data subsets according to the intermediate features and the corresponding intermediate cutting points, until the preset termination condition is met, and the intermediate decision tree corresponding to the m training data extracted this time is obtained; the preset termination condition is that the data points in the data subset are completely isolated or the intermediate decision tree reaches a preset maximum depth.

[0056] A first target preset model is generated according to the intermediate decision trees corresponding to the m training data extracted a preset number of times, wherein the intermediate features at least include the time fluctuations corresponding to the training data.

[0057] In an embodiment of the present disclosure, when the first preset model is an iForest model, the first preset model is composed of a decision tree iTree, and an intermediate decision tree is constructed for each selected m training data. The construction process of the intermediate decision tree includes randomly selecting an intermediate feature from all features, and randomly selecting an intermediate split point for the intermediate feature, dividing the data set composed of each data point of the training data into two data subsets, and repeatedly performing the operations of selecting the intermediate feature, the intermediate split point, and the data subset division until the training data is completely isolated or the depth of the intermediate decision tree reaches the preset maximum depth. The m training data extracted the preset number of times can construct the same number of intermediate decision trees as the preset number of times, thereby generating the first target preset model.

[0058] In some embodiments, the selection of intermediate features and intermediate split points is random, and in the preset number of selections of m training data, at least one of the intermediate features corresponding to the m training data selected is the time fluctuation corresponding to the training data.

[0059] In some embodiments, before constructing the intermediate decision tree for each of the m training data extracted, the further step includes: extracting intermediate features from the m training data, where the intermediate features at least include the time fluctuations corresponding to the training data. The disclosed embodiment does not impose any special restrictions on the method of extracting the intermediate features, which may be to decompose the time series of noise data in a continuous time period into trends, residuals, etc., or to determine through an LSTM (Long Short-Term Memory) network.

[0060] In some embodiments, reference Figure 3 , step S1 includes steps S11 to S13.

[0061] S11. Perform a first noise reduction process on the noise data to determine first abnormal data and a first evaluation value of each data point in the first abnormal data; the first abnormal data is determined based at least on a time fluctuation of the noise data.

[0062] S12: Perform a second noise reduction process on the first abnormal data to obtain a second evaluation value of each data point in the first abnormal data.

[0063] S13. Determine data to be eliminated from the first abnormal data according to the first evaluation value and the second evaluation value, and eliminate the data to be eliminated from the noise data to obtain target monitoring data.

[0064] In the embodiment of the present disclosure, the noise data is subjected to a first noise reduction process by using a trained first target preset model, which is equivalent to a preliminary screening of each data point in the noise data, and the data points suspected of being abnormal constitute the first abnormal data, and the first evaluation value of each data point is determined; the data points in the first abnormal data are monitored for a second time using the second target preset model to determine whether these data points can be determined as abnormal data points, so that the abnormal data points are removed from the noise data to obtain the target monitoring data. Through the first noise reduction process and the second noise reduction process, abnormal data points can be automatically identified, with the advantage of a low misjudgment rate, especially the first noise reduction process, which can isolate data points that are significantly different from other data points earlier, and then combined with the second noise reduction process, it can ensure that the artificial noise generated by the operation and maintenance personnel is accurately removed.

[0065] In some embodiments, data points suspected of being abnormal obtained after being processed by the first target preset model may also be marked.

[0066] In some embodiments, during the first denoising process of the noise data using the first target preset model, a contamination parameter may also be set, which refers to the expected abnormal value percentage of the first target preset model for the training data in abnormality monitoring.

[0067] In some embodiments, before step S11, the process further includes: performing dimensionality reduction processing on the noise data. In the disclosed embodiment, there is no special limitation on the dimensionality reduction method, which may be PCA (Principal Component Analysis) or t-SNE (t-distributed Stochastic Neighbor Embedding).

[0068] In some embodiments, step S11 includes: For each data point in the noise data, the formula , calculating the average path length of the data point in at least one decision tree in the first target preset model; wherein, for The expected value of is the path length corresponding to the i-th decision tree, and t is the total number of decision trees; By formula , calculate the first evaluation value; wherein, is the standardization factor, is the first evaluation value.

[0069] In an embodiment of the present disclosure, a data point is input into the first target preset model to determine whether the data point is abnormal, and the average path length of the data point isolated in each decision tree corresponding to the first target preset model is calculated. When the average path length corresponding to the data point is shorter, it means that the data point is easy to be isolated and has a greater probability of being an abnormal value. Otherwise, it is not easy to be isolated and has a greater probability of being a normal value.

[0070] Where H(i) is the harmonic number estimated by ln(i)+0.5772156649 (i.e., the Euler constant γ), h(x) is the average path length of the isolated data point x in the first target preset model, E(h(x)) is the expected value of h(x), c(n) is the normalization factor of the average path length of any isolated data point in the first target preset model, n is the number of samples (i.e., the number of data points), and Score(x,n) is the anomaly score of the data point x (i.e., the first evaluation value).

[0071] When Score(x,n) approaches 1, x is an outlier; when Score(x,n) approaches 0, x is a normal value. In this solution, 0.5 can be set as the judgment threshold of the anomaly score to find outliers. In some embodiments, formula (2) is the normalization factor .

[0072] (2).

[0073] In some embodiments, .

[0074] In some embodiments, the data point with the highest evaluation value among the data points of the noise data is selected by formula (3). samples.

[0075] (3).

[0076] In some embodiments, step S12 includes: For each data point in the first abnormal data, a reachable distance between the data point and k adjacent data points is calculated.

[0077] The local density of the data points is determined according to the reachable distance.

[0078] A second evaluation value of each data point in the first abnormal data is determined according to a preset threshold and a ratio between a local density of the data point and a local density of the adjacent data points.

[0079] In an embodiment of the present disclosure, the reachable distance between a data point and k adjacent data points is calculated using formula (4).

[0080] (4).

[0081] in, is the data point, are k adjacent data points, the , is the i-th data point x i The k-nearest neighbor set of is the i-th data point x i The j-th data point in the k-nearest neighbor set of .

[0082] The local density of the data points is calculated using formula (5).

[0083] (5).

[0084] in, is the i-th data point x i The set of k-nearest neighbors of For data point x i The local density of .

[0085] The evaluation value of the data point is calculated using formula (6).

[0086] (6).

[0087] in, For data point x i The evaluation value of is the i-th data point x i The k-nearest neighbor set of , is the i-th data point x i The jth data point in the k-nearest neighbor set of For data points The local density of For data point x i The local density of .

[0088] In some embodiments, the data point with the highest evaluation value among the data points of the first abnormal data is selected by formula (7). samples.

[0089] (7).

[0090] in, To select the set of m samples (i.e., samples) with the highest evaluation values ​​among the data points of the first abnormal data, is the i-th data point, C is the candidate sample set consisting of all the first abnormal data, is the evaluation value of the ith data point, To sort the data points by evaluation value, is the ranking threshold.

[0091] In some embodiments, step S13 includes: Normalizing the first evaluation value and the second evaluation value to obtain a first intermediate evaluation value and a second intermediate evaluation value; For each data point in the first abnormal data, a first evaluation value and a second evaluation value of the data point are weightedly calculated according to a preset weight value to determine a comprehensive evaluation value; A preset number of data points in the first abnormal data with the highest comprehensive evaluation values ​​are determined as data to be eliminated.

[0092] In the embodiment of the present disclosure, the first evaluation value is normalized by formula (8).

[0093] (8).

[0094] in, is the first intermediate evaluation value of the ith data point, is the first evaluation value of the i-th data point, is the minimum value of the first evaluation values ​​of each data point in the first abnormal data, is the maximum value of the first evaluation values ​​of each data point in the first abnormal data.

[0095] The second evaluation value is normalized using formula (9).

[0096] (9).

[0097] in, is the second intermediate evaluation value of the ith data point, For data point x i The evaluation value of is the minimum value of the second evaluation values ​​of each data point in the first abnormal data, is the maximum value of the second evaluation values ​​of each data point in the first abnormal data.

[0098] The first evaluation value and the second evaluation value are weightedly calculated using formula (10).

[0099] (10).

[0100] in, is the comprehensive evaluation value of the i-th data point, is the weight, is the first intermediate evaluation value of the ith data point, is the second intermediate evaluation value of the i-th data point.

[0101] Formula (11) is used to preset a number k of data points with the highest comprehensive evaluation values.

[0102] (11).

[0103] in, is the set of k data points with the highest comprehensive evaluation value selected, To sort each data point according to the comprehensive evaluation value.

[0104] In the above-mentioned embodiment of the present disclosure, a denoising model is used to perform denoising processing on the monitored noise data. The denoising model evaluates the abnormality of the noise data in a continuous time period based on the time fluctuation of the noise data, so as to eliminate abnormal noise points generated during the operation and maintenance process, so that the operation and maintenance process does not need to suspend monitoring every time. That is, while the noise monitoring system is in normal noise monitoring, the operation and maintenance personnel can operate and maintain the equipment, thereby meeting the needs of continuous environmental noise monitoring, reducing the cumbersome operation and maintenance process in high-altitude operations, and ensuring the safety of the operation and maintenance personnel.

[0105] In a second aspect, the present disclosure provides a noise monitoring system, referring to Figure 4 , which includes: Main pole 1; A key block 2, wherein a first end of the key block 2 is clamped with the main pole 1; A rotating rod 3, wherein a first end of the rotating rod 3 is rotatably connected to a first end of the main vertical rod 1, a second end of the key block 2 is inserted into the interior of the rotating rod 3, and the second end of the rotating rod 3 is used to support a sampling probe 4; The gear assembly 5 is used to provide resistance to the rotation of the rotating rod 3; The host 6 is used to implement the intelligent environmental noise monitoring method described in the first aspect.

[0106] In the embodiment of the present disclosure, when the sampling probe 4 rotates with the rotating rod 3, the rotating rod 3 rotates around the first end of the main vertical pole 1 and triggers the gear assembly 5 to provide resistance to the rotation of the rotating rod 3, so that the second end of the rotating rod 3 stops rotating when it rotates close to the ground, thereby allowing the operation and maintenance personnel to operate and maintain the sampling probe 4 that was originally at a high place without having to climb up.

[0107] In some embodiments, the main pole 1 can be fixed vertically to the ground by pre-embedded means, or fixed to the ground by expansion screws, and the present disclosure is not limited thereto. The host 6 is fixed to the main pole 1 in the form of a wall mount, or the host 6 is placed on the ground. The sampling probe 4 is used to collect noise data, and the host 6 is used to further process the collected noise data.

[0108] In some embodiments, the first end of the key block 2 can be locked to the main pole 1 .

[0109] It is worth noting that when the key block 2 is close to the ground, clamped on the main pole 1 and inserted into the first end of the rotating rod 3, the mass of the key block 2 should be at least greater than the sum of the masses of the rotating rod 3 and the sampling probe 4, the rotating rod 3 with the sampling probe 4 fixed is perpendicular to the ground, and the first end of the rotating rod 3 is close to the ground and the second end is far from the ground. The connection between the key block 2, the main pole 1 and the rotating rod 3 can ensure that the noise monitoring system will not generate self-noise due to environmental interference (for example, wind), avoiding the influence on the monitored noise data.

[0110] As an example, see Figure 5 , when the mass of the key block 2 is m1, the distance from the center of gravity of the key block 2 to the axial center of the third gear 53 is L1, the total mass of the sampling probe 4 and the rotating rod 3 is m2, and the distance from the center of gravity of the sampling probe 4 and the rotating rod 3 to the axial center of the third gear 53 is L2, When the rotating rod 3 is in the state of being perpendicular to the ground, the end (second end) where the sampling probe 4 is located is away from the ground and the first end is close to the ground. The first end of the key block 2 can be engaged with the main vertical pole 1 and the key block 2 and the main vertical pole 1 can be locked by using a structural lock.

[0111] In some embodiments, reference Figure 6 , the gear assembly 5 comprises: A first gear 51, wherein the first gear 51 comprises a damping protrusion 511; a second gear 52, the second gear 52 meshing with the first gear 51; A third gear 53, the third gear 53 is meshed with the second gear 52 and is fixedly connected to the rotating rod 3, and the third gear 53 includes a damping slide 531; When the first end of the key block 2 is clamped on the first end of the main pole 1 and the second end of the key block 2 is inserted into the interior of the rotating rod 3, the key block 2 limits the rotating rod 3; when the first end of the key block 2 is not clamped on the first end of the main pole 1 and the second end of the key block 2 is not inserted into the interior of the rotating rod 3, the rotating rod 3 rotates around the first end of the key block 2, the third gear 53 rotates with the rotating rod 3, the damping protrusion 511 of the first gear 51 contacts the damping slide 531 of the third gear 53 and generates resistance during the rotation process.

[0112] In an embodiment of the present disclosure, the gear assembly 5 includes a first gear 51, a second gear 52 and a third gear 53 that are meshed in sequence. The damping protrusion 511 in the first gear 51 and the damping slope 531 in the third gear 53 can cooperate with each other to produce a damping effect on the rotation of the gears, thereby providing resistance to the rotation of the rotating rod 3.

[0113] Compared with the related art that the sampling probe is fixedly installed on the top of the host to obtain data. In the present disclosure, the sampling probe 4 is fixedly connected to one end of the rotating rod 3, and the other end of the rotating rod 3 is fixedly connected to the third gear 53 of the gear assembly 5. When the first end of the key block 2 is clamped on the first end of the main vertical pole 1 and the second end of the key block 2 is inserted into the interior of the rotating rod 3, under the action of the gravity of the key block 2, the rotation of the rotating rod 3 is restricted, that is, the rotating rod 3 is limited, so that the sampling probe 4 can be fixed on one end of the rotating rod 3 and away from the ground; when the first end of the key block 2 is not clamped on the first end of the main vertical pole 1 and the second end of the key block 2 is not inserted into the interior of the rotating rod 3, the rotating rod 3 rotates, thereby providing a transmission force for the third gear 53, and the third gear 53 rotates with the side of the rotating rod 3 where the sampling probe 4 is fixed. The rotation of the third gear 53 drives the first gear 51 and the second gear 52 to rotate in turn, so that the damping protrusion 511 contacts the damping slide 531 and generates resistance in the opposite direction of the gear rotation during the rotation process.

[0114] Among them, the damping bump 511 and the damping slope 531 are arranged in the gear assembly 5, which can stop the rotating rod 3 from rotating when the sampling probe 4 rotates close to the ground, thereby avoiding excessive rotation due to inertia, thereby not causing damage to the sampling probe 4, and the operation and maintenance personnel can access the sampling probe 4 and perform maintenance without high-altitude operations.

[0115] The working principle of the noise monitoring system is explained below: Reference Figure 4 When the first end of the key block 2 is clamped on the first end of the main pole 1 and the second end of the key block 2 is inserted into the interior of the rotating rod 3, the key block 2 restricts the rotation of the rotating rod 3, and the end of the rotating rod 3 where the sampling probe 4 is fixed is away from the ground, and the rotating rod 3 is perpendicular to the ground and is in a stationary state. Figure 6 It can be seen that at this time, the gears do not rotate and the damping protrusion 511 is not in contact with the third gear 53.

[0116] As a specific form of the embodiment of the present disclosure, the operation and maintenance personnel remove the key block 2 and provide an initial speed for the rotating rod 3. The rotating rod 3 will no longer be able to maintain the vertical ground state when the first end of the key block 2 is clamped on the first end of the main vertical pole 1 and the second end of the key block 2 is inserted into the interior of the rotating rod 3, driving the third gear 53 to perform circular rotation. At this time, the first end of the key block 2 is not clamped on the first end of the main vertical pole 1 and the second end of the key block 2 is not inserted into the interior of the rotating rod 3. The rotation of the rotating rod 3 can provide transmission force for the third gear 53, and the rotation of the third gear 53 drives the rotation of the first gear 51 and the second gear 52 in turn. The embodiment of the present disclosure does not impose any special restrictions on the rotation direction of the rotating rod 3, which can be left or right.

[0117] After the key block 2 is removed, the rotation process of the rotating rod 3 can be divided into two stages. For the mass of the rotating rod 3 being m, the length of the rotating rod 3 being L, and the given initial speed being V0, the processes of the two stages are further described: 1) Reference Figure 7 and 9 The sampling probe 4 at the second end of the rotating rod 3 rotates along an arc from 0 to 1 / 4 of the circumference with the first end of the rotating rod 3 as the center and the rod length as the radius, that is, the rotating rod 3 rotates from 0 to 90 degrees. Figure 8 and 10 During the process, the damping protrusion 511 does not contact the damping slope 531, so the rotation process is an undamped acceleration motion process.

[0118] During the rotation process, the minimum value of the rotation speed is the initial speed V0, and the maximum value of the rotation speed is the speed V1 when the rotating rod 3 is parallel to the ground (ie, the rotating rod 3 rotates 90°).

[0119] When the rotating rod 3 is perpendicular to the ground, the center of gravity of the rotating rod 3 is located at the bottom. , the linear velocity V0 of the top end (i.e., the second end) of the rotating rod 3, so the initial potential energy at this time is , the initial kinetic energy is .

[0120] When the rotating rod 3 is parallel to the ground, the bottom end is the zero potential energy position, the center of gravity height of the rotating rod 3 is 0, and the speed is V1, so the initial potential energy at this time is , the initial kinetic energy is .

[0121] Furthermore, the total mechanical energy is conserved under ideal conditions (i.e., without external forces), so the initial mechanical energy is equal to the final mechanical energy. The relationship between the initial mechanical energy and the final mechanical energy is shown by formula (12).

[0122] (12).

[0123] According to formula (12), the maximum value of the rotation speed of the rotating rod 3 depends on the initial speed V0 and the rod length L. When the initial speed is large, the maximum value of the rotation speed is also large. If the initial speed is less than the critical value, the rotating rod 3 may not be able to move parallel to the ground.

[0124] In some embodiments, a first threshold is set for the rotation speed of the rotating rod 3. If the rotation speed of the rotating rod 3 is greater than the first threshold, the sampling probe 4 on the rotating rod 3 is damaged. In some embodiments, the value range of the first threshold is 0-3 m / s.

[0125] 2) Reference Fig.11 The sampling probe 4 at the second end of the rotating rod 3 rotates along a 1 / 4 to 1 / 2 arc with the first end of the rotating rod 3 as the center and the rod length as the radius, that is, the rotating rod 3 rotates from 90 to 180 degrees. Fig.12 During the process, the damping protrusion 511 contacts the damping slope 531, so the rotation process is a damped deceleration motion process.

[0126] During the rotation process, since the damping slide 531 in contact with the damping protrusion 511 is a slide structure, the closer to 180°, the greater the damping coefficient is, and at 180°, the damping coefficient is close to infinity, so the rotating rod 3 will rotate to stop between 90 and 180°.

[0127] As another specific form of the embodiment of the present disclosure, refer to Figures 13 to 15 After the operation and maintenance personnel complete the operation and maintenance of the sampling probe 4, they restore the monitoring state of the sampling probe 4, that is, restore the vertical ground state of the rotating rod 3 when the first end of the key block 2 is clamped on the first end of the main vertical pole 1. At this time, the rotating rod 3 is pushed so that the damping protrusion 511 is pushed out of the damping area corresponding to the damping slope 531, the key block 2 is inserted into the bottom of the rotating rod 3, and the rotating rod 3 performs a circular rotation until the end of the rotating rod 3 where the sampling probe 4 is fixed is restored to a static state perpendicular to the ground, and the first end of the key block 2 is locked on the main vertical pole 1.

[0128] For the mass of the rotating rod 3 being m, the length of the rotating rod 3 being L, and the given initial velocity being V0, the rotation process of the rotating rod 3 is further described as follows: Under the gravity of the key block 2 , the potential energy change in the system is converted into the rotational kinetic energy of the rotating rod 3 , providing sufficient initial potential energy to rotate the rotating rod 3 , which mainly comes from the gravitational potential energy of the rotating rod 3 and the key block 2 .

[0129] For the rotating rod 3: The mass of the rotating rod 3 is m, and the total length is L. Initially, the rotating rod 3 is stationary and vertical or inclined to the ground. The center of gravity of the rotating rod 3 is at the midpoint of the rod. Therefore, the initial potential energy is ;in, is the vertical distance from the center of gravity to the fulcrum; further, since the initial inclination angle of the rotating rod 3 is , so , the potential energy of the rotating rod 3 is .

[0130] For key block 2: The mass of key block 2 is M, and the distance from the fulcrum is l, so its initial potential energy is .

[0131] Therefore, the total initial potential energy is .

[0132] By providing initial kinetic energy to the rotating rod 3, the rotating rod 3 and the sampling probe 4 can make a circular motion upward. The rotational kinetic energy of the upward circular motion is determined by the rotational inertia and angular velocity Provide, that is .

[0133] Among them, the moment of inertia of the rotating rod 3 about the fulcrum is , the moment of inertia of key block 2 is , so the total moment of inertia is .

[0134] By satisfying the relationship between potential energy and kinetic energy, the damping such as friction and air resistance is overcome, so that the rotating rod 3 and the sampling probe 4 begin to move in a circular motion upward.

[0135] Initially, the angular velocity , so the total energy at the beginning is potential energy. Based on the law of conservation of energy, potential energy is converted into kinetic energy: ,Right now .

[0136] The angular velocity can be determined from the relationship between kinetic energy and total energy With initial inclination The relationship between is formula (13).

[0137] (13).

[0138] As another specific form of the embodiment of the present disclosure, when the monitoring state of the sampling probe 4 is restored, that is, when the first end of the key block 2 is clamped on the first end of the main pole 1 and the rotating rod 3 is vertical to the ground, the rotation speed of the rotating rod 3 is set to 0-3m / s. According to the angular velocity calculation formula , where r is the distance from the sampling probe 4 to the rotation center. The maximum angular velocity can be determined as , and the initial tilt angle is .

[0139] In the above-mentioned embodiment of the present disclosure, by arranging the damping protrusion 511 and the damping slide 531 in the gear assembly 5, the rotating rod 3 can stop rotating when the sampling probe 4 rotates close to the ground, thereby avoiding excessive rotation due to inertia, thereby preventing damage to the sampling probe 4, and the operation and maintenance personnel can access the sampling probe 4 and perform maintenance without high-altitude operations.

[0140] In a third aspect, an embodiment of the present disclosure provides a computer-readable medium having a computer program stored thereon, and when the computer program is executed by a processor, the method for intelligent monitoring of environmental noise described in the first aspect is implemented.

[0141] It should be noted that the computer-readable medium is a medium corresponding to the above-mentioned environmental noise intelligent monitoring method. All implementation methods in the above-mentioned method embodiment are applicable to the embodiment of the computer-readable medium and can achieve the same technical effect.

[0142] In order to enable those skilled in the art to more clearly understand the technical solutions provided by the embodiments of the present disclosure, the technical solutions provided by the embodiments of the present disclosure are described in detail below through specific embodiments: Example 1 Exemplarily, as a specific form of the embodiment of the present disclosure, when the operation and maintenance personnel arrive at the noise monitoring system, the operation and maintenance process for the noise monitoring system includes: Confirm whether there are any new fixed sources or other situations within a certain range (for example, 200 meters) that may affect the monitoring results of noise data. If there are any abnormalities, give priority to the surrounding monitoring environment conditions.

[0143] Open the limit of key block 2 and remove key block 2, provide an initial speed to the rotating rod 3, and the rotating rod 3 rotates. When the sampling probe 4 in the rotating rod 3 rotates close to the ground, remove the windproof ball kit and put it into the sound calibrator.

[0144] Operate the sound calibrator to make it sound, monitor the equipment and perform automatic calibration within 1 minute of the calibrator making the sound.

[0145] When the linkage alarm light indicates that the calibration is complete, check whether the windshield kit is contaminated and needs to be cleaned or replaced. If maintenance is required, clean it in time and install the windshield kit back to the top of the rotating rod after cleaning.

[0146] The rotating rod 3 is pushed out of the damping area, and the key block 2 is inserted into the first end of the rotating rod 3. The rotating rod 3 performs circular rotation. When the sampling probe 4 and the rotating rod 3 are perpendicular to the ground and the sampling probe 4 is far away from the ground, the key block 2 is limited and locked in the main vertical pole 1.

[0147] The acoustic calibration operation and maintenance is completed after checking that there are no remaining operation and maintenance items on site.

[0148] Among them, the processor is a device with data processing capabilities, including but not limited to the central processing unit (CPU); the memory is a device with data storage capabilities, including but not limited to random access memory (RAM, more specifically SDRAM, DDR, etc.), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory (FLASH); the I / O interface (read-write interface) is connected between the processor and the memory, and can realize information exchange between the memory and the processor, including but not limited to the data bus (Bus), etc.

[0149] Those skilled in the art will appreciate that all or some of the steps, systems, and functional modules / units in the apparatus disclosed above may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0150] In hardware implementations, the division between functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, one physical component may have multiple functions, or one function or step may be performed by several physical components in cooperation.

[0151] Some or all physical components may be implemented as software executed by a processor, such as a central processing unit (CPU), a digital signal processor or a microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or non-transitory medium) and a communication medium (or transient medium). As is well known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules or other data). Computer storage media include, but are not limited to, random access memory (RAM, more specifically SDRAM, DDR, etc.), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory (FLASH) or other disk storage; read-only compact disk (CD-ROM), digital versatile disk (DVD) or other optical disk storage; magnetic cassettes, tapes, disk storage or other magnetic storage; any other medium that can be used to store desired information and can be accessed by a computer. Furthermore, it is well known to those skilled in the art that communication media typically embodies computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0152] The present disclosure has disclosed example embodiments, and although specific terms are employed, they are used and should be interpreted only in a general illustrative sense and not for limiting purposes. In some instances, it will be apparent to those skilled in the art that, unless otherwise expressly stated, features, characteristics, and / or elements described in conjunction with a particular embodiment may be used alone or in combination with features, characteristics, and / or elements described in conjunction with other embodiments. Therefore, those skilled in the art will appreciate that various changes in form and detail may be made without departing from the scope of the present disclosure as set forth in the appended claims.

Claims

1. An intelligent environmental noise monitoring method, characterized in that: A host applied to a noise monitoring system, the noise monitoring system further comprising a gear assembly and a rotating rod, the rotating rod being used to support a sampling probe, the gear assembly being used to provide resistance to the rotation of the rotating rod, the method comprising: In response to monitoring noise data, the noise data is input into a denoising model for noise reduction processing to obtain target monitoring data; wherein the noise data is time series data within a continuous time period, and the denoising model is used to evaluate the degree of abnormality of each data point in the noise data based on the time fluctuation of the noise data.

2. The method according to claim 1, characterized in that: The training process of the denoising model includes: Obtain N training data; each training data includes time series noise data in a continuous time period; Take m training data from the N training data according to a preset number of times and input them into the first preset model for processing to obtain a first target preset model; 1<m<N; Inputting the nth training data among the N training data into the first target preset model for first noise reduction processing to determine the intermediate abnormal data; n is sequentially selected from integers from 1 to N; Inputting the intermediate abnormal data corresponding to the nth training data into the second preset model corresponding to the n-1th training data for second noise reduction processing to obtain the second preset model corresponding to the nth training data; A denoising model is determined according to the first target preset model and the second preset model corresponding to the Nth training data.

3. The method according to claim 2, characterized in that: The step of taking m training data from the N training data according to a preset number of times and inputting them into a first preset model for processing to obtain a first target preset model includes: For each data point in the m training data extracted each time, select the intermediate features of each data point and its corresponding intermediate split point; According to the intermediate features and their corresponding intermediate split points, each data point is divided into two data subsets; For each data point in each data subset, repeatedly perform the steps of selecting the intermediate features of each data point and the corresponding intermediate split points, and dividing each data point into two data subsets according to the intermediate features and the corresponding intermediate split points, until a preset termination condition is met, and an intermediate decision tree corresponding to the m training data extracted this time is obtained; the preset termination condition is that each data point in the data subset is completely isolated or the intermediate decision tree reaches a preset maximum depth; wherein the intermediate features at least include the time fluctuation of each data point; A first target preset model is generated according to the intermediate decision tree corresponding to the m training data extracted a preset number of times.

4. The method according to claim 1, characterized in that: The step of inputting the noise data into a denoising model for denoising to obtain target monitoring data includes: Performing a first noise reduction process on the noise data to determine first abnormal data and a first evaluation value of each data point in the first abnormal data; the first abnormal data is determined based at least on a time fluctuation of the noise data; Performing a second noise reduction process on the first abnormal data to obtain a second evaluation value of each data point in the first abnormal data; According to the first evaluation value and the second evaluation value, data to be eliminated is determined from the first abnormal data, and the data to be eliminated is eliminated from the noise data to obtain target monitoring data.

5. The method according to claim 4, characterized in that: The performing a first noise reduction process on the noise data to determine first abnormal data and a first evaluation value of each data point in the first abnormal data includes: For each data point in the noise data, the formula , calculating the average path length of the data point in at least one decision tree in the first target preset model; wherein, for The expected value of is the path length corresponding to the i-th decision tree, and t is the total number of decision trees; By formula , calculate the first evaluation value; wherein, is the standardization factor.

6. The method according to claim 4, characterized in that: The performing a second noise reduction process on the first abnormal data to obtain a second evaluation value of each data point in the first abnormal data includes: For each data point in the first abnormal data, calculating the reachable distance between the data point and k adjacent data points; Determining a local density of the data points according to the reachable distance; A second evaluation value of each data point in the first abnormal data is determined according to a preset threshold and a ratio between a local density of the data point and a local density of the adjacent data points.

7. The method according to claim 4, characterized in that: The step of determining the data to be removed from the first abnormal data according to the first evaluation value and the second evaluation value includes: Normalizing the first evaluation value and the second evaluation value to obtain a first intermediate evaluation value and a second intermediate evaluation value; For each data point in the first abnormal data, a first evaluation value and a second evaluation value of the data point are weightedly calculated according to a preset weight value to determine a comprehensive evaluation value; A preset number of data points in the first abnormal data with the highest comprehensive evaluation values ​​are determined as data to be eliminated.

8. A noise monitoring system, characterized in that: It includes: Main pole; A key block, a first end of which is clamped with the main pole; A rotating rod, wherein the first end of the rotating rod is rotatably connected to the first end of the main vertical rod, the second end of the key block is inserted into the interior of the rotating rod, and the second end of the rotating rod is used to support the sampling probe; a gear assembly, used to provide resistance to the rotation of the rotating rod; A host, used to implement the environmental noise intelligent monitoring method described in any one of claims 1 to 7.

9. The system according to claim 8, characterized in that: The gear assembly comprises: a first gear, the first gear comprising a damping bump; a second gear, the second gear meshing with the first gear; a third gear, the third gear meshing with the second gear and fixedly connected to the rotating rod, the third gear comprising a damping slide; When the first end of the key block is clamped on the first end of the main pole and the second end of the key block is inserted into the interior of the rotating rod, the key block limits the rotating rod; when the first end of the key block is not clamped on the first end of the main pole and the second end of the key block is not inserted into the interior of the rotating rod, the rotating rod rotates around the first end of the key block, the third gear rotates with the rotating rod, and the damping protrusion of the first gear contacts the damping slope of the third gear and generates resistance during the rotation process.

10. A computer-readable medium, characterized in that A computer program is stored thereon, and when the computer program is executed by a processor, the method for intelligent monitoring of environmental noise described in any one of claims 1 to 7 is implemented.

Images