CN118861873B - Intelligent construction site dust environment monitoring method, system and storage medium - Google Patents
Intelligent construction site dust environment monitoring method, system and storage medium Download PDFInfo
- Publication number
- CN118861873B CN118861873B CN202410814422.6A CN202410814422A CN118861873B CN 118861873 B CN118861873 B CN 118861873B CN 202410814422 A CN202410814422 A CN 202410814422A CN 118861873 B CN118861873 B CN 118861873B
- Authority
- CN
- China
- Prior art keywords
- monitoring information
- dust
- environmental
- early warning
- feature
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D21/00—Measuring or testing not otherwise provided for
- G01D21/02—Measuring two or more variables by means not covered by a single other subclass
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
- G06F18/15—Statistical pre-processing, e.g. techniques for normalisation or restoring missing data
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING SYSTEMS, e.g. PERSONAL CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B31/00—Predictive alarm systems characterised by extrapolation or other computation using updated historic data
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2123/00—Data types
- G06F2123/02—Data types in the time domain, e.g. time-series data
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Emergency Management (AREA)
- Business, Economics & Management (AREA)
- Computing Systems (AREA)
- Probability & Statistics with Applications (AREA)
- Testing Or Calibration Of Command Recording Devices (AREA)
Abstract
本发明公开了智慧工地扬尘环境监测方法、系统及存储介质，属于环境监测技术领域，解决了现有方法基于支持向量机或神经网络机器学习训练的危险状况模型难以提取监测指标数据的显著性特征的问题，方法包括：基于多功能监测仪采集时序性环境监测信息，获得归一化评估特征向量；预构建基于主成分分析结合随机森林的环境评估预警模型；环境评估预警模型识别分析归一化评估特征向量，计算综合污染度；本发明通过对采集到的时序性环境监测信息预处理，提取时序性环境监测信息的显著性特征，保证了基于主成分分析结合随机森林的环境评估预警模型构建和训练时能够在训练数据上学习到真正的数据分布，有效减少了模型过拟合的风险。
The present invention discloses a smart construction site dust environment monitoring method, system and storage medium, which belong to the field of environmental monitoring technology. It solves the problem that the existing method is difficult to extract the significant features of monitoring index data based on the dangerous condition model trained by support vector machine or neural network machine learning. The method comprises: collecting time-series environmental monitoring information based on a multi-functional monitoring instrument to obtain a normalized evaluation feature vector; pre-constructing an environmental assessment and early warning model based on principal component analysis combined with random forest; the environmental assessment and early warning model identifies and analyzes the normalized evaluation feature vector and calculates the comprehensive pollution degree; the present invention extracts the significant features of the time-series environmental monitoring information by pre-processing the collected time-series environmental monitoring information, thereby ensuring that the real data distribution can be learned on the training data when the environmental assessment and early warning model based on principal component analysis combined with random forest is constructed and trained, and the risk of model overfitting is effectively reduced.
Description
Technical Field
The invention belongs to the technical field of environment monitoring, and particularly relates to a method, a system and a storage medium for monitoring dust environment of an intelligent building site.
Background
Building site dust monitoring is the process of monitoring and managing dust pollution of a building site in real time by utilizing modern sensor technology, internet of things and artificial intelligence technology. With the acceleration of the urban process, building construction activities are increasingly frequent, and the dust emission problem generated by the building construction activities becomes one of important factors affecting the urban air quality. The effectual dust monitoring not only can help the management dust pollution condition, can also provide healthier living environment for constructor.
The Chinese patent CN108267170B discloses an environment monitoring method and device, which comprises the steps of acquiring environment monitoring index data obtained by monitoring the current environment by an environment sensor on a wearable device and storing the environment monitoring index data, analyzing, predicting and processing the environment monitoring index data to obtain an analysis and prediction result, and executing corresponding prompt or alarm operation on the wearable device according to the analysis and prediction result.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides an intelligent building site dust environment monitoring method, an intelligent building site dust environment monitoring system and a storage medium, and solves the problems that the existing method is difficult to extract the salient features of monitoring index data based on a dangerous condition model of Support Vector Machine (SVM) or neural network machine learning training, and the model robustness is poor when the model is trained by facing to multi-source heterogeneous detection index data.
The invention discloses a method for monitoring dust environment of an intelligent building site, which comprises the following steps:
Acquiring time-series environment monitoring information based on a multifunctional monitor, wherein the time-series environment monitoring information is multi-source heterogeneous measurement data, the time-series environment monitoring information comprises air quality data, dust content data, noise monitoring data, wind speed monitoring data and dust particle shooting data, preprocessing the acquired time-series environment monitoring information, extracting the significance characteristics of the time-series environment monitoring information, and obtaining normalized evaluation feature vectors;
Pre-constructing an environment assessment and early warning model based on principal component analysis and random forest combination, performing iterative training after the environment assessment and early warning model is improved, and outputting the environment assessment and early warning model;
the environment assessment early warning model identifies, analyzes and normalizes the assessment feature vector, calculates the comprehensive pollution degree, judges whether air pollution exists, and triggers an air pollution instruction if the air pollution exists;
and the pollution calculation network in the environment evaluation early warning model responds to the air pollution instruction, loads the normalized evaluation feature vector, calculates the air pollution diffusivity, and feeds back the air pollution diffusivity and the air pollution diffusivity.
The multifunctional monitor acquires the air pollution degree and the air pollution diffusion degree, triggers the dust fall instruction, and carries out dust fall treatment on a construction site based on the dust fall instruction.
Preferably, the preprocessing the collected time-series environmental monitoring information specifically includes:
Acquiring time sequence environment monitoring information, and performing first pretreatment and second pretreatment on the time sequence environment monitoring information to obtain a first pretreatment set and a second pretreatment set, wherein the first pretreatment is monitoring information average value 0 pretreatment, and the second pretreatment is monitoring information root mean square error 1 pretreatment;
wherein the set of time sequence environment monitoring information is The formula for setting the monitoring information mean value to 0 is as follows:
(1)
For the centralized time-sequential environmental monitoring information, The information average value is monitored;
(2)
Wherein, For the amount of data in the set of time-series environmental monitoring information,Monitoring information for the input time sequence environment;
(3)
Wherein, The root mean square error value of the time-series environment monitoring information,Is the average value of the time sequence environment monitoring information;
acquiring centralized time-series environment monitoring information and root mean square error values, and constructing a feature decomposition matrix based on the centralized time-series environment monitoring information and the root mean square error values;
Performing dimension reduction treatment on the feature decomposition matrix to obtain a dimension-reduced feature decomposition matrix;
extracting the salient features of the time sequence environment monitoring information based on a Bayesian algorithm, and integrating to obtain a normalized evaluation feature vector;
wherein, the saliency feature extraction function is:
(4)
in order to normalize the estimated feature vector, In order to significantly transform the coefficients of the coefficients,Amplitude is transformed for significance;
(5)
Wherein, In order to significantly transform the coefficients of the coefficients,Amplitude is transformed for significance.
Preferably, the method for pre-constructing the environmental assessment early warning model based on principal component analysis and random forest specifically comprises the following steps:
obtaining a model training sample, wherein the model training sample is historical environment monitoring information, and preprocessing the model training sample;
initializing a random forest model, constructing a feature evaluator based on principal component analysis and a SimAM attention mechanism, and defining a weighting function of the feature evaluator on sample features, wherein the weighting function of the feature evaluator on the sample features is expressed as follows:
(6)
Wherein, As the weight value of the sample feature,As the number of neurons of the feature assessor,The weight deviation coefficient set for the feature judger,As the dimension of the feature vector,Evaluating feature vectors for normalization of model training samples;
Training and initializing a random forest model by using the weighted model training sample to obtain a random forest decision tree layer good at air pollution detection;
incorporating the feature judger, the random forest decision tree layer and the pollution calculation network into an initialized random forest model to obtain an environment assessment early warning model;
the environment assessment early warning model comprises an input layer, a characteristic judge, a random forest decision tree layer, a pollution calculation network and an output layer.
Preferably, the method for identifying, analyzing and normalizing the evaluation feature vector by the environment evaluation early warning model and calculating the comprehensive pollution degree specifically comprises the following steps:
Acquiring a normalized evaluation feature vector, and identifying abnormal feature factors of the normalized evaluation feature vector by a feature judge in an environment evaluation early warning model based on a K-means algorithm;
the abnormal characteristic factor extraction function is as follows:
(7);
the value of the abnormal characteristic factor is set as the evaluation value of the abnormal characteristic factor during extraction Determining the feature factor in the normalized evaluation feature vector as an abnormal feature factor by comparing the abnormal feature factor extraction value with the abnormal feature factor evaluation value, wherein,To extract the cardinality;
inputting the abnormal characteristic factor judgment value of the abnormal characteristic factor into a random forest decision tree layer, and calculating the comprehensive pollution degree;
(8)
Wherein, In order to integrate the degree of pollution,Is the bias factor.
The invention also provides an intelligent building site dust environment monitoring system, which comprises:
the multifunctional monitor is used for collecting time sequence environment monitoring information;
the data processing module is used for preprocessing the collected time-series environment monitoring information, extracting the significance characteristics of the time-series environment monitoring information and obtaining normalized evaluation characteristic vectors;
The model construction module is used for constructing an environment assessment early-warning model based on principal component analysis and random forest pre-construction, performing iterative training after the environment assessment early-warning model is improved, and outputting the environment assessment early-warning model;
The pollution comprehensive judging module is used for identifying, analyzing and normalizing the evaluation feature vector based on the environment evaluation early warning model, calculating the comprehensive pollution degree, judging whether air pollution exists, triggering an air pollution instruction if the air pollution exists, calculating the air pollution diffusivity, and feeding back the air pollution degree and the air pollution diffusivity;
and the dust fall processing module is used for carrying out dust fall processing on the construction site based on the dust fall instruction.
Preferably, the multifunctional monitor includes:
The monitoring instrument comprises a monitoring instrument main body, a monitoring instrument and a monitoring instrument, wherein the monitoring instrument main body comprises a monitoring base, a supporting rod, an equipment supporting seat and an adjusting installation seat, the supporting rod is rotatably installed on the monitoring base, the equipment supporting seat is detachably connected with the top of the supporting rod, and the adjusting installation seat is fixedly connected with the equipment supporting seat;
a multi-source data acquisition unit mounted on the equipment support bracket, the multi-source data acquisition unit being configured to acquire time-series environmental monitoring information;
The position adjusting assembly comprises a main body adjusting part and an equipment adjusting part, wherein the main body adjusting part is arranged on the monitoring base, the main body adjusting part is connected with the supporting rod, the main body adjusting part is used for adjusting the rotating angle of the supporting rod, the equipment adjusting part is arranged on the adjusting installation seat, the equipment adjusting part is connected with the multi-source data acquisition part, and the equipment adjusting part is used for adjusting the inclination angle of the multi-source data acquisition part.
Preferably, the main body adjusting part includes:
the first motor is fixedly arranged on the monitoring base;
the first gear is fixedly connected with the output end of the first motor and is rotationally connected with the upper surface of the monitoring base;
The second gear is fixedly sleeved outside the supporting rod, and meshed with the first gear for transmission;
The device adjustment section includes:
The second motor is fixedly arranged on the adjusting installation seat, the output end of the second motor is fixedly connected with a transmission fluted disc, and the transmission fluted disc is rotatably arranged on one side of the adjusting installation seat;
At least one group of third gears, the third gears are arranged on one side of the transmission fluted disc, the third gears and the transmission fluted disc are meshed for transmission, one side of each third gear is fixedly connected with a transmission screw, the transmission screw is rotationally connected with the side wall of the adjusting installation seat, one end of the transmission screw, which is far away from the third gear, is sheathed with a movable tooth seat in a threaded manner, and the movable tooth seat is in sliding connection with the equipment supporting seat through a guide block;
The fourth gear is rotatably arranged on the equipment supporting seat, is meshed with the movable tooth holder for transmission, and is also connected with the multi-source data acquisition part;
The negative pressure transmission shaft is fixedly connected with the transmission fluted disc, and at least one group of negative pressure dust extraction fans are detachably arranged at the end part of the negative pressure transmission shaft and used for assisting in cleaning the multi-source data acquisition part.
Preferably, the multi-source data acquisition section includes:
The acquisition instrument controller is fixedly connected with the fourth gear;
The collecting swing seat is arranged on the collecting instrument controller, a plurality of groups of atomizing spray ports are embedded in the collecting swing seat, the atomizing spray ports are connected with a water storage tank through a telescopic water guide pipe, and the atomizing spray ports carry out dust fall treatment on a construction site based on a dust fall instruction;
The CCD camera is arranged on the acquisition swing seat and is used for acquiring dust particle shooting data;
the dust sensor is used for collecting dust content data;
The noise collector is used for collecting noise monitoring data;
the air quality monitor is used for collecting air quality data;
the wind speed collector is used for collecting wind speed monitoring data;
The acquisition instrument controller is electrically connected with the CCD camera, the dust sensor, the noise acquisition device, the air quality monitor and the wind speed acquisition device respectively.
On the other hand, the invention also provides a computer readable storage medium, and the computer readable storage medium stores computer program instructions which are executed to realize the intelligent building site dust environment monitoring method.
Compared with the prior art, the embodiment of the application has the following main beneficial effects:
according to the embodiment of the invention, the acquired time-series environment monitoring information is preprocessed, the salient features of the time-series environment monitoring information are extracted, the normalized evaluation feature vector is obtained, and the acquired multi-source heterogeneous data can be normalized, so that the salient features of the data are accurately extracted, the real data distribution can be learned on training data when the environment evaluation early warning model based on the principal component analysis and the random forest is combined for construction and training, the risk of model fitting is effectively reduced, and the accuracy of comprehensive pollution degree evaluation is ensured.
According to the embodiment of the invention, an environment assessment early warning model based on principal component analysis and combining with a random forest is provided, the environment assessment early warning model is an improved random forest model, a SimAM attention mechanism is introduced in model construction to construct a feature judging device, and meanwhile, the principal component analysis and the random forest are combined to combine the advantages of the principal component analysis in the aspect of significant feature extraction and the advantages of the random forest in the aspect of long-distance modeling, so that the features of multi-source heterogeneous data are better extracted, and the accuracy of a random forest decision tree layer and the feature judging device is improved.
In the embodiment of the invention, an intelligent building site dust environment monitoring system is provided, which consists of a multifunctional monitor, a data processing module, a model construction module, a pollution comprehensive judging module and a dust fall processing module, wherein the multifunctional monitor can accurately acquire time sequence environment monitoring information, the acquisition coverage area is larger, the accuracy is higher, and the multifunctional monitor and the dust fall processing module can optimize a dust fall spraying path based on air pollution diffusivity, so that the energy consumption of dust fall operation is reduced, and the intelligent green environment-friendly construction is realized.
In the embodiment of the invention, the multifunctional monitor is arranged, and the multifunctional monitor realizes the accurate acquisition of the environmental monitoring data of different areas of the construction site through the multi-source data acquisition part, so that the data support is provided for the analysis and evaluation of the construction environment by the environmental evaluation early warning model, and meanwhile, the equipment adjustment part can synchronously adjust the positions of a plurality of groups of multi-source data acquisition parts, thereby not only facilitating the acquisition of data, but also facilitating the spraying and dust removing operation on the construction site.
Drawings
FIG. 1 is a schematic diagram of an implementation flow of the intelligent building site dust environment monitoring method provided by the invention.
Fig. 2 is a schematic diagram of a flow chart for preprocessing collected time-series environmental monitoring information.
Fig. 3 is a schematic flow chart of implementation of the method for pre-constructing the environmental assessment early warning model based on principal component analysis and random forest.
Fig. 4 is a schematic flow chart of the method for calculating the comprehensive pollution degree by identifying, analyzing and normalizing the evaluation feature vector by using the environmental evaluation early warning model.
FIG. 5 is a control framework diagram of the intelligent building site dust environment monitoring system provided by the invention.
Fig. 6 is a schematic structural diagram of the multifunctional monitor provided by the invention.
Fig. 7 is a schematic three-dimensional structure of the multifunctional monitor provided by the invention.
Fig. 8 is a schematic structural diagram of a multi-source data acquisition unit according to the present invention.
Fig. 9 is an isometric view of a multi-source data acquisition unit provided by the present invention.
Fig. 10 is a front view of a multi-source data acquisition unit provided by the present invention.
In the figure, a 10-multifunctional monitor, a 11-monitoring base, a 12-supporting rod, a 13-equipment supporting seat, a 14-adjusting installation seat, a 2-position adjusting component, a 21-main body adjusting part, a 211-first motor, a 212-first gear, a 213-second gear, a 22-equipment adjusting part, a 221-second motor, a 222-transmission fluted disc, a 223-third gear, a 224-transmission screw, a 225-movable tooth seat, a 226-guide block, a 227-fourth gear, a 228-negative pressure transmission shaft, a 229-negative pressure dust suction fan, a 3-multi-source data acquisition part, a 31-acquisition instrument controller, a 32-acquisition swing seat, a 33-atomization spray opening, a 34-telescopic water guide pipe, a 35-CCD camera, a 36-noise collector, a 37-dust sensor, a 38-air quality monitor, a 39-wind speed collector, a 20-data processing module, a 30-model building module, a 40-pollution comprehensive judging module and a 50-dust fall processing module are shown.
Detailed Description
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs, the terms used in the description of this application are for the purpose of describing particular embodiments only and are not intended to be limiting of the application, and the terms "comprising" and "having" and any variations thereof in the description of this application and the claims and the above description of the drawings are intended to cover non-exclusive inclusions. The terms first, second and the like in the description and in the claims or in the above-described figures, are used for distinguishing between different objects and not necessarily for describing a sequential or chronological order.
Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the application. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of skill in the art will explicitly and implicitly appreciate that the embodiments described herein may be combined with other embodiments.
The existing method is based on a Support Vector Machine (SVM) or a neural network machine learning training dangerous condition model, the significance characteristics of monitoring index data are difficult to extract, the robustness of the model is poor when the model is trained on multi-source heterogeneous detection index data, aiming at the problems, an intelligent building site dust environment monitoring method, system and storage medium are provided, in short, the method comprises the steps of collecting time sequence environment monitoring information based on a multifunctional monitor 10, pre-constructing an environment assessment early-warning model based on principal component analysis and combining with random forests, identifying and analyzing normalized assessment feature vectors by the environment assessment early-warning model, calculating comprehensive pollution degree, judging whether air pollution exists or not, loading the normalized assessment feature vectors by a pollution calculation network in the environment assessment early-warning model in response to an air pollution instruction, and calculating the air pollution diffuseness. According to the embodiment of the invention, the acquired time-series environment monitoring information is preprocessed, the salient features of the time-series environment monitoring information are extracted, the normalized evaluation feature vector is obtained, and the acquired multi-source heterogeneous data can be normalized, so that the salient features of the data are accurately extracted, the real data distribution can be learned on training data when the environment evaluation early warning model based on the principal component analysis and the random forest is combined for construction and training, the risk of model fitting is effectively reduced, and the accuracy of comprehensive pollution degree evaluation is ensured.
The embodiment of the invention provides a method for monitoring dust environment of an intelligent building site, wherein fig. 1 shows a schematic flow chart of implementation of the method for monitoring dust environment of the intelligent building site, and the method for monitoring dust environment of the intelligent building site specifically comprises the following steps:
Step S10, collecting time-series environment monitoring information based on the multifunctional monitor 10, wherein the time-series environment monitoring information is multi-source heterogeneous measurement data, the time-series environment monitoring information comprises air quality data, dust content data, noise monitoring data, wind speed monitoring data and dust particle shooting data, preprocessing the collected time-series environment monitoring information, extracting the significance characteristics of the time-series environment monitoring information, and obtaining normalized evaluation characteristic vectors;
It should be noted that the time-series environmental monitoring information is multi-source heterogeneous measurement data, the type of the multi-source heterogeneous measurement data may be structured data, semi-structured data and unstructured data, wherein the unstructured data includes but is not limited to text, picture, audio and video data, and meanwhile, the time-series environmental monitoring information acquisition period may be 30min, 60min, 2h and 4h.
S20, pre-constructing an environment assessment and early warning model based on principal component analysis and random forest combination, and performing iterative training after improving the environment assessment and early warning model to output the environment assessment and early warning model;
In this embodiment, the number of iterative training times of the environmental assessment and early warning model may be 100-2000 times, and the ultra-parameter optimization mode of the environmental assessment and early warning model may be grid search (GRIDSEARCHCV) and random search (RandomizedSearchCV).
S30, identifying and analyzing normalized evaluation feature vectors by using an environment evaluation early warning model, and calculating the comprehensive pollution degree;
step S40, judging whether air pollution exists or not;
it should be noted that, unlike the present step of judging the air pollution level by using a single judgment index, in this embodiment, the air environment of the current construction environment is comprehensively evaluated by the comprehensive pollution level, and the basis for judging whether the air pollution exists is whether the calculated comprehensive pollution level exceeds a preset pollution threshold, and the preset pollution threshold may be set to 0.5-0.8.
Step S50, if air pollution exists, triggering an air pollution instruction;
If no air pollution exists, the current comprehensive pollution degree is output, and the multifunctional monitor 10 continues to periodically acquire time-sequence environment monitoring information.
And S60, loading a normalized evaluation feature vector by a pollution calculation network in the environment evaluation early warning model in response to the air pollution instruction, calculating the air pollution diffuseness, and feeding back the air pollution diffuseness and the air pollution diffuseness.
In step S70, the multifunctional monitor 10 acquires the air pollution level and the air pollution diffusion level, triggers a dust fall instruction, and performs dust fall treatment on the construction site based on the dust fall instruction.
According to the embodiment of the invention, the acquired time-series environment monitoring information is preprocessed, the salient features of the time-series environment monitoring information are extracted, the normalized evaluation feature vector is obtained, and the acquired multi-source heterogeneous data can be normalized, so that the salient features of the data are accurately extracted, the real data distribution can be learned on training data when the environment evaluation early warning model based on the principal component analysis and the random forest is combined for construction and training, the risk of model fitting is effectively reduced, and the accuracy of comprehensive pollution degree evaluation is ensured.
The embodiment of the invention provides a method for preprocessing acquired time-series environment monitoring information, and fig. 2 shows a schematic flow chart for implementing the preprocessing of the acquired time-series environment monitoring information, wherein the preprocessing of the acquired time-series environment monitoring information specifically comprises the following steps:
Step S101, acquiring time-sequence environmental monitoring information, and performing first pretreatment and second pretreatment on the time-sequence environmental monitoring information to obtain a first pretreatment set and a second pretreatment set, wherein the first pretreatment is monitoring information mean value 0 pretreatment, and the second pretreatment is monitoring information root mean square error 1 pretreatment;
According to the embodiment of the invention, the data centering can be realized by carrying out the first preprocessing and the second preprocessing on the time-sequence environment monitoring information, so that the influence on the data analysis result caused by different feature sizes is eliminated, and the data is more suitable for comprehensive comparison and evaluation.
Wherein the set of time sequence environment monitoring information isThe formula for setting the monitoring information mean value to 0 is as follows:
(1)
For the centralized time-sequential environmental monitoring information, The information average value is monitored;
(2)
Wherein, For the amount of data in the set of time-series environmental monitoring information,Monitoring information for the input time sequence environment;
(3)
Wherein, The root mean square error value of the time-series environment monitoring information,Is the average value of the time sequence environment monitoring information.
Step S102, acquiring centralized time-series environment monitoring information and root mean square error values, and constructing a feature decomposition matrix based on the centralized time-series environment monitoring information and the root mean square error values;
In the embodiment, the feature decomposition matrix is constructed based on the centralized time-sequential environment monitoring information and the root mean square error value, so that on one hand, the calculation process of the model can be simplified, the algorithm efficiency is improved, and on the other hand, the balance between the centering of the model and the root mean square error value can be found by constructing the feature decomposition matrix, and therefore the risk of over-fitting is reduced, and the prediction performance of the model on new data is improved.
It should be noted that, the method for constructing the feature decomposition matrix based on the centralized time-series environment monitoring information and the root mean square error value may be to find the feature values and the feature vectors of the centralized time-series environment monitoring information and the root mean square error value. Then, a feature vector corresponding to the feature value is found by solving a linear equation set, normalized into a unit vector, and then a feature decomposition matrix is constructed.
Step S103, performing dimension reduction processing on the feature decomposition matrix to obtain a dimension-reduced feature decomposition matrix;
step S104, extracting the salient features of the time sequence environment monitoring information based on a Bayesian algorithm, and integrating to obtain a normalized evaluation feature vector;
wherein, the saliency feature extraction function is:
(4)
in order to normalize the estimated feature vector, For the significant conversion coefficient, in the present embodiment, the significant conversion coefficient may be set to 0.01 to 0.02,Amplitude is transformed for significance;
(5)
Wherein, In order to significantly transform the coefficients of the coefficients,Amplitude is transformed for significance.
The embodiment of the invention provides a method for pre-constructing an environmental assessment and early warning model based on principal component analysis and random forest, and fig. 3 shows a schematic flow chart of implementation of the method for pre-constructing the environmental assessment and early warning model based on principal component analysis and random forest, wherein the method for pre-constructing the environmental assessment and early warning model based on principal component analysis and random forest specifically comprises the following steps:
step S201, a model training sample is obtained, wherein the model training sample is historical environment monitoring information, and the model training sample is preprocessed;
It should be noted that, the model training samples are also heterogeneous environmental monitoring information, the model training samples may be obtained from the multifunctional monitor 10, or may be obtained from a shared platform, and the mode of preprocessing the model training samples is the same as preprocessing the acquired time-series environmental monitoring information.
Step S202, initializing a random forest model, constructing a feature judgment device based on the principle component analysis and the integration of SimAM attention mechanisms, and defining a weighting function of the feature judgment device on sample features, wherein the weighting function of the feature judgment device on the sample features is expressed as:
(6)
Wherein, As the weight value of the sample feature,The number of neurons that are feature assessors may be 10-100,The weight deviation coefficient set for the feature evaluator, in this embodiment, the weight deviation coefficient may be set to 0.1 or 0.2,As the dimension of the feature vector, in this embodiment, the feature vector is a normalized evaluation feature vector, and the feature dimension of the normalized evaluation feature vector may be 2 or 3,Evaluating feature vectors for normalization of model training samples;
Step S203, training and initializing a random forest model by using the weighted model training sample to obtain a random forest decision tree layer good at air pollution detection;
and S204, incorporating the feature judger, the random forest decision tree layer and the pollution calculation network into the initialized random forest model to obtain an environment assessment early warning model.
In this embodiment, the pollution computing network may be a yolov model backbone network incorporating a K-means algorithm, the input layer is connected to a feature evaluator, the feature evaluator is connected to a random forest decision tree layer, and the pollution computing network is connected to an output layer
The environment assessment early warning model comprises an input layer, a characteristic judge, a random forest decision tree layer, a pollution calculation network and an output layer.
According to the embodiment of the invention, an environment assessment early warning model based on principal component analysis and combining with a random forest is provided, the environment assessment early warning model is an improved random forest model, a SimAM attention mechanism is introduced in model construction to construct a feature judging device, and meanwhile, the principal component analysis and the random forest are combined to combine the advantages of the principal component analysis in the aspect of significant feature extraction and the advantages of the random forest in the aspect of long-distance modeling, so that the features of multi-source heterogeneous data are better extracted, and the accuracy of a random forest decision tree layer and the feature judging device is improved.
The embodiment of the invention provides a method for identifying, analyzing and normalizing an evaluation feature vector and calculating comprehensive pollution degree by an environment evaluation early warning model, wherein fig. 4 shows a schematic flow chart of the method for identifying, analyzing and normalizing the evaluation feature vector and calculating the comprehensive pollution degree by the environment evaluation early warning model, and the method specifically comprises the following steps:
step S301, a normalized evaluation feature vector is obtained, and a feature judge in an environment evaluation early warning model identifies abnormal feature factors of the normalized evaluation feature vector based on a K-means algorithm;
the abnormal characteristic factor extraction function is as follows:
(7);
the value of the abnormal characteristic factor is set as the evaluation value of the abnormal characteristic factor during extraction Determining the feature factor in the normalized evaluation feature vector as an abnormal feature factor by comparing the abnormal feature factor extraction value with the abnormal feature factor evaluation value, wherein,For extraction radix, in this embodiment, the extraction radix may be 0.3-0.8;
step S302, inputting an abnormal characteristic factor judgment value of an abnormal characteristic factor into a random forest decision tree layer, and calculating the comprehensive pollution degree;
(8)
Wherein, In order to integrate the degree of pollution,As the bias coefficient, in this embodiment, the bias coefficient may be updated by an optimization algorithm such as gradient descent, so as to minimize the loss function, and the bias coefficient may be set to 1-5.
In this embodiment, the comprehensive pollution level is used to characterize the environmental pollution level, and the environmental assessment early warning model combines various pollution factors such as dust pollution, noise pollution, air quality pollution and the like during calculation, so as to implement comprehensive assessment of environmental pollution.
Step S303, calculating air pollution diffusivity based on the comprehensive pollution degree;
wherein, the air pollution diffusivity is calculated by the formula (9):
(9)
Wherein, For the air pollution diffusivity, the air pollution diffusivity is obtained,To integrate the pollution degreeFor the amount of data in the set of time-series environmental monitoring information,For the wind speed feature vector in the set of time-series environmental monitoring information,Wind vector feature vectors are collected for time-series environmental monitoring information.
On the other hand, the embodiment of the invention also provides an intelligent building site dust environment monitoring system, and fig. 5 shows a control frame diagram of the intelligent building site dust environment monitoring system, where the intelligent building site dust environment monitoring system specifically includes:
A multifunctional monitor 10, wherein the multifunctional monitor 10 is used for collecting time-series environment monitoring information;
The data processing module 20 is used for preprocessing the collected time-series environment monitoring information, extracting the significance characteristics of the time-series environment monitoring information and obtaining normalized evaluation characteristic vectors;
the model construction module 30 is used for constructing an environment assessment early-warning model based on principal component analysis and random forest pre-construction, performing iterative training after the environment assessment early-warning model is improved, and outputting the environment assessment early-warning model;
The pollution comprehensive judging module 40 is used for identifying, analyzing and normalizing the evaluation feature vector based on the environment evaluation early warning model, calculating the comprehensive pollution degree, judging whether air pollution exists, triggering an air pollution instruction if the air pollution exists, calculating the air pollution diffusion degree, and feeding back the air pollution degree and the air pollution diffusion degree;
The dust fall processing module 50, the dust fall processing module 50 performs dust fall processing on the construction site based on the dust fall instruction.
It should be noted that, the multifunctional monitor 10, the data processing module 20, the model building module 30, the pollution comprehensive judging module 40, and the dust settling processing module 50 implement data interaction by using bluetooth or a local area network. In the embodiment of the invention, an intelligent building site dust environment monitoring system is provided, the intelligent building site dust environment monitoring system is composed of a multifunctional monitor 10, a data processing module 20, a model construction module 30, a pollution comprehensive judging module 40 and a dust fall processing module 50, the multifunctional monitor 10 can accurately acquire time-series environment monitoring information, the acquisition coverage area is larger, the accuracy is higher, and the multifunctional monitor 10 and the dust fall processing module 50 can optimize a dust fall spraying path based on air pollution diffusivity, so that the energy consumption of dust fall operation is reduced, and intelligent green environment-friendly construction is realized.
The embodiment of the invention provides a multifunctional monitor 10, fig. 6-7 show schematic structural diagrams of the multifunctional monitor 10, and the multifunctional monitor 10 specifically includes:
the monitor main body comprises a monitor base 11, a support rod 12, an equipment supporting seat 13 and an adjusting installation seat 14, wherein the support rod 12 is rotatably installed on the monitor base 11, the equipment supporting seat 13 is detachably connected with the top of the support rod 12, and the adjusting installation seat 14 is fixedly connected with the equipment supporting seat 13;
It should be noted that, the monitoring base 11 may be a rectangular base or a circular base, in order to facilitate moving the multifunctional monitor 10, a plurality of groups of moving rollers may be installed at the bottom of the monitoring base 11, and the top of the device supporting seat 13 and the top of the supporting rod 12 may be connected in a threaded or plug-in manner, the device supporting seat 13 and the adjusting mounting seat 14 may be hollow circular bases or circular rings, and the device supporting seat 13 and the adjusting mounting seat 14 are fixedly connected through a buckle or a bolt.
A multi-source data acquisition unit 3 mounted on the equipment support bracket 13, the multi-source data acquisition unit 3 being configured to acquire time-series environmental monitoring information;
The position adjusting assembly 2, the position adjusting assembly 2 includes a main body adjusting portion 21 and an equipment adjusting portion 22, the main body adjusting portion 21 is installed on the monitoring base 11, the main body adjusting portion 21 is connected with the supporting rod 12, the main body adjusting portion 21 is used for adjusting the rotation angle of the supporting rod 12, the equipment adjusting portion 22 is installed on the adjusting installation seat 14, the equipment adjusting portion 22 is connected with the multi-source data obtaining portion 3, and the equipment adjusting portion 22 is used for adjusting the inclination angle of the multi-source data obtaining portion 3.
In the embodiment of the invention, the multifunctional monitor 10 is arranged, and the multifunctional monitor 10 realizes accurate acquisition of environment monitoring data of different areas of a construction site through the multi-source data acquisition part 3, so that data support is provided for analysis and evaluation of construction environments of an environment evaluation early warning model, and meanwhile, the equipment adjustment part 22 can synchronously adjust the positions of a plurality of groups of the multi-source data acquisition parts 3, thereby not only facilitating data acquisition, but also facilitating spraying and dust removing operation on the construction site.
In a further preferred embodiment of the present invention, as shown in fig. 6, the main body adjusting part 21 includes:
a first motor 211, the first motor 211 being fixedly mounted on the monitoring base 11;
The first gear 212 is fixedly connected with the output end of the first motor 211, and the first gear 212 is rotationally connected with the upper surface of the monitoring base 11;
the second gear 213 is fixedly sleeved outside the support rod 12, and the second gear 213 and the first gear 212 are in meshed transmission.
It should be noted that, the first motor 211 may be a servo motor, and the first motor 211 is fixedly mounted on the monitoring base 11 through a fastening bolt, an output end of the first motor 211 is fixedly connected with the first gear 212 in an interference fit manner, and the second gear 213 is fixedly mounted on an outer wall of the support rod 12 in a joggle or plug manner.
When the position and the angle of the multi-source data acquisition part 3 need to be adjusted during operation, the first motor 211 is started, and the first motor 211 is started to drive the first gear 212 to rotate, so that the first gear 212 drives the second gear 213 and the support rod 12 to rotate, and the position and the angle of the multi-source data acquisition part 3 are adjusted.
In the present embodiment, the device adjusting section 22 includes:
The second motor 221 is fixedly arranged on the adjusting installation seat 14, the output end of the second motor 221 is fixedly connected with a transmission fluted disc 222, and the transmission fluted disc 222 is rotatably arranged on one side of the adjusting installation seat 14;
It should be noted that, the second motor 221 may be a single-phase asynchronous motor, and the output end of the second motor 221 is fixedly connected with the driving fluted disc 222 in an interference fit manner, and the driving fluted disc 222 is rotatably connected with the side wall of the adjusting installation seat 14 through a roller or a bearing.
At least one group of third gears 223, the third gears 223 are arranged on one side of the transmission fluted disc 222, the third gears 223 and the transmission fluted disc 222 are in meshed transmission, one side of each third gear 223 is fixedly connected with a transmission screw 224, the transmission screw 224 is rotationally connected with the side wall of the adjusting installation seat 14, one end of the transmission screw 224 far away from the third gears 223 is sheathed with a movable tooth seat 225 in a threaded manner, and the movable tooth seat 225 is in sliding connection with the equipment supporting seat 13 through a guide block 226;
A fourth gear 227, the fourth gear 227 is rotatably installed on the equipment supporting seat 13, the fourth gear 227 is in meshed transmission with the movable tooth holder 225, and the fourth gear 227 is also connected with the multi-source data acquisition part 3;
In this embodiment, the number of the third gears 223 may be three or four, and the third gears 223 are riveted or welded with the end of the driving screw 224, the driving screw 224 is rotatably connected with the side wall of the adjusting installation seat 14 through a bearing, the moving tooth seat 225 is a hollow screw seat, the side wall of the moving tooth seat 225 is provided with a guide groove, the guide groove is slidably connected with a guide block 226, the guide block 226 is fixedly connected with the equipment supporting seat 13 through a supporting rod or a fastening bolt, and the fourth gear 227 is rotatably connected with the equipment supporting seat 13 through a bearing or a rotating shaft.
The negative pressure transmission shaft 228, the negative pressure transmission shaft 228 is fixedly connected with the transmission fluted disc 222, and at least one group of negative pressure dust extraction fans 229 are detachably mounted at the end part of the negative pressure transmission shaft 228, and the negative pressure dust extraction fans 229 are used for assisting in cleaning the multi-source data acquisition part 3.
In this embodiment, when the spraying angle or the data collection angle needs to be adjusted, the second motor 221 is turned on, and the second motor 221 is turned on to drive the driving fluted disc 222 to rotate, so that the driving fluted disc 222 synchronously drives multiple groups of the third gears 223 to rotate, the third gears 223 drive the driving screw 224 to rotate, and the driving screw 224 can drive the movable toothholder 225 to move when rotating, so that the movable toothholder 225 drives the fourth gear 227 to rotate, and the fourth gear 227 rotates to drive the multi-source data acquisition part 3 to swing, thereby facilitating adjustment of the spraying angle and the data collection angle.
In a further preferred embodiment of the present invention, as shown in fig. 8 to 10, the multi-source data acquisition section 3 includes:
the collector controller 31, the collector controller 31 is fixedly connected with the fourth gear 227;
The collecting swing seat 32 is arranged on the collecting instrument controller 31, a plurality of groups of atomizing spray ports 33 are embedded in the collecting swing seat 32, the atomizing spray ports 33 are connected with a water storage tank through a telescopic water guide pipe 34, and the atomizing spray ports 33 carry out dust fall treatment on a construction site based on a dust fall instruction.
It should be noted that, the collector controller 31 may be a PLC controller, and the collector controller 31 is further electrically connected to the first motor 211 and the second motor 221, where the collector controller 31 is configured to control the first motor 211 and the second motor 221 to be turned on or turned off, and the collector controller 31 is further communicatively connected to the data processing module 20, the model building module 30, the pollution comprehensive judging module 40, and the dust settling processing module 50.
In this embodiment, the atomizing spray opening 33 is a matrix or is linearly arranged on the side wall of the collecting swing seat 32, a high-pressure atomizing spray nozzle is arranged in the atomizing spray opening 33, and the setting of the atomizing spray opening 33 and the high-pressure atomizing spray nozzle facilitates dust fall treatment on the construction environment, the collecting swing seat 32 can be a rectangular seat or a T-shaped seat structure with hollow inside, and one end of the collecting swing seat 32 is fixedly connected with the side wall of the collecting instrument controller 31 by clamping or fastening bolts.
And the CCD camera 35 is arranged on the acquisition swinging seat 32, and the CCD camera 35 is used for acquiring dust particle shooting data, wherein the CCD camera 35 directly captures light change in an image through a photoelectric element on a sensor of the CCD camera, converts an optical signal into an electric signal and finally forms a digital image. When capturing dust and dirt, the light reflected or scattered by these particles is captured and recorded by the CCD sensor, thereby showing the presence of dust and dirt in the image.
A dust sensor 37, the dust sensor 37 being configured to collect dust content data;
The noise collector 36 is configured to collect noise monitoring data, and in this embodiment, the noise collector 36 may be a rectangular plate or a fan-shaped plate structure, and a microphone is built in the noise collector 36, and can convert received sound waves into an electrical signal. The conversion is based on vibration of a film in the microphone caused by sound waves, so that voltage change is generated, and conversion from sound signals to electric signals is realized.
An air quality monitor 38, the air quality monitor 38 is configured to collect air quality data, and in this embodiment, the air quality monitor 38 may measure the concentration of pollutants such as sulfur dioxide, nitrogen dioxide, PM10, PM2.5, etc.;
the wind speed collector 39 is used for collecting wind speed monitoring data, the output mode of the wind speed collector 39 is RS485, and the wind speed collector 39 can be mechanical or ultrasonic;
The collector controller 31 is electrically connected with the CCD camera 35, the dust sensor 37, the noise collector 36, the air quality monitor 38 and the wind speed collector 39 respectively.
In this embodiment, the CCD camera 35, the dust sensor 37, the noise collector 36, the air quality monitor 38, and the wind speed collector 39 are mounted at the end of the collecting swing seat 32 by means of a snap or fastening bolt.
In another aspect, embodiments of the present invention also provide a computer-readable storage medium storing computer program instructions executable by a processor. Which when executed, performs the method of any of the embodiments described above.
The computer-readable storage medium (e.g., memory) herein may be volatile memory or nonvolatile memory, or may include both volatile and nonvolatile memory. By way of example, and not limitation, nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM), which acts as external cache memory. By way of example, and not limitation, RAM may be available in a variety of forms such as synchronous RAM (DRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), and Direct Rambus RAM (DRRAM).
In summary, the embodiment of the invention provides the smart site dust environment monitoring method, the smart site dust environment monitoring system and the storage medium, wherein the acquired time-series environment monitoring information is preprocessed, the salient features of the time-series environment monitoring information are extracted to obtain the normalized evaluation feature vector, and the acquired multi-source heterogeneous data can be normalized, so that the salient features of the data can be accurately extracted, the real data distribution can be learned on training data when the environment evaluation early warning model based on the principal component analysis and the random forest is constructed and trained, the risk of model overfitting is effectively reduced, and the accuracy of comprehensive pollution degree evaluation is ensured.
The above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the scope of the present invention. It will be apparent that the described embodiments are merely some, but not all, embodiments of the invention. Based on these embodiments, all other embodiments that may be obtained by one of ordinary skill in the art without inventive effort are within the scope of the invention. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art may still combine, add or delete features of the embodiments of the present invention or make other adjustments according to circumstances without any conflict, so as to obtain different technical solutions without substantially departing from the spirit of the present invention, which also falls within the scope of the present invention.
Claims (6)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410814422.6A CN118861873B (en) | 2024-06-24 | 2024-06-24 | Intelligent construction site dust environment monitoring method, system and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410814422.6A CN118861873B (en) | 2024-06-24 | 2024-06-24 | Intelligent construction site dust environment monitoring method, system and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118861873A CN118861873A (en) | 2024-10-29 |
| CN118861873B true CN118861873B (en) | 2025-04-18 |
Family
ID=93178241
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410814422.6A Active CN118861873B (en) | 2024-06-24 | 2024-06-24 | Intelligent construction site dust environment monitoring method, system and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118861873B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN120538591B (en) * | 2025-05-22 | 2026-05-08 | 浙江省环保集团生态环保研究院有限公司 | A method and system for online integrated environmental monitoring |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114440980A (en) * | 2022-02-07 | 2022-05-06 | 西北大学 | Ecological environment monitoring system based on remote sensing and geographic information system |
| CN115840793A (en) * | 2022-12-12 | 2023-03-24 | 四川大学 | Meteorological space normalization method and system based on random forest |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106779505B (en) * | 2017-02-28 | 2021-04-02 | 中国南方电网有限责任公司 | A transmission line fault early warning method and system driven by big data |
| CN111445110B (en) * | 2020-03-05 | 2023-03-03 | 深圳供电局有限公司 | Environmental risk decision-making method, device, and computer equipment based on cable trench |
| CN116132081B (en) * | 2022-09-05 | 2025-02-07 | 桂林电子科技大学 | Collaborative defense method for DDOS attacks in software-defined networks based on ensemble learning |
| CN116306197A (en) * | 2022-09-07 | 2023-06-23 | 贵州大学 | Complex Manufacturing Process Modeling and Prediction Method Based on Functional Wide Learning Network |
| CN117164103B (en) * | 2023-07-03 | 2024-08-09 | 广西智碧达智慧环境科技有限公司 | Intelligent control method, terminal and system of domestic sewage treatment system |
| CN117521873A (en) * | 2023-09-18 | 2024-02-06 | 重庆邮电大学 | PM2.5 inversion method and system based on comprehensive data processing and random forest |
| CN117689516A (en) * | 2023-12-07 | 2024-03-12 | 四川大学 | Chemical industry garden water, soil and gas environment monitoring system |
-
2024
- 2024-06-24 CN CN202410814422.6A patent/CN118861873B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114440980A (en) * | 2022-02-07 | 2022-05-06 | 西北大学 | Ecological environment monitoring system based on remote sensing and geographic information system |
| CN115840793A (en) * | 2022-12-12 | 2023-03-24 | 四川大学 | Meteorological space normalization method and system based on random forest |
Non-Patent Citations (1)
| Title |
|---|
| "基于随机森林模型的武汉市城区大气PM2.5来源解析";张志豪等;《环境科学》;20220331;第43卷(第3期);第1151-1156页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118861873A (en) | 2024-10-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109886396A (en) | A kind of transmission line galloping on-line prediction system and method | |
| CN118861873A (en) | Intelligent construction site dust environment monitoring method, system and storage medium | |
| CN111878324A (en) | Wind power plant tower drum vortex-induced vibration early warning method and early warning system | |
| CN117804405A (en) | A method and system for monitoring deformation of transmission line poles and towers | |
| CN111414871A (en) | Intelligent safety protection detection system based on big data | |
| CN119559769B (en) | A voice warning system and method for preventing campus bullying | |
| CN119446186A (en) | Multi-channel audio acquisition and intelligent fusion coal mill fault voiceprint recognition system and method | |
| US10460431B2 (en) | Part recognition and damage characterization using deep learning | |
| CN117990162B (en) | Inspection well monitoring device and method based on convolutional neural network | |
| CN120126328A (en) | Highway environment monitoring terminal based on the Internet of Things | |
| CN120738896A (en) | Abnormality detection method for fabric processing equipment, electronic equipment and fabric processing equipment | |
| CN115526205A (en) | A method and system for intelligent recognition of excavator voiceprint based on convolutional neural network | |
| CN119827709A (en) | Full-automatic carbon emission monitoring system and carbon emission monitoring method | |
| CN121049272B (en) | Visual audio multi-mode monitoring system and method for identifying defects of fan blades | |
| CN116151042B (en) | Crane monitoring and maintaining method and system based on multidimensional data analysis | |
| CN118155135A (en) | A method for identifying trash racks in hydropower plants based on improved YOLOv8n | |
| CN116937819A (en) | Remote power distribution cabinet monitoring system with learning function | |
| CN116278896A (en) | Charging pile safety monitoring system based on multiple perception analysis | |
| CN120629370A (en) | A fully automatic wall surface acquisition device and intelligent wall hollowing detection method | |
| CN108009514A (en) | Level of material for ball mill Forecasting Methodology | |
| CN119778158A (en) | A yaw control method and system based on meteorological data | |
| CN120048066A (en) | TinyML-based industrial scene personnel state abnormality detection and early warning system and method | |
| CN119226663A (en) | Multi-view human posture data acquisition device and data processing method for embodied intelligence | |
| CN115166872B (en) | Snow concentration detection method, detection device and snow protection system | |
| CN121834134A (en) | Outdoor near-surface high-frequency wind, sand and dust collaborative observation system and monitoring method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP03 | Change of name, title or address | ||
| CP03 | Change of name, title or address |
Address after: 233400 Anhui Province Bengbu City China (Anhui) Free Trade Pilot Zone Bengbu Area Yuhui District Gaoxin Road South Xinghua Road East Electronic Information Industrial Park No. 1 Comprehensive Office Building 205 Patentee after: Anhui Construction Engineering Water Resources Development Investment Group Ecological Technology Co.,Ltd. Country or region after: China Patentee after: Anhui Water Resources Development Co.,Ltd. Address before: 233050 Anhui Province, Bengbu City, China (Anhui) Free Trade Pilot Zone, Yuhui District, South of Gaoxin Road, East of Xinghua Road, in the Information Electronics Industrial Park, No. 1 Comprehensive Office Building, 205th Floor Patentee before: Anhui Construction Engineering Ecological Technology Co.,Ltd. Country or region before: China Patentee before: Anhui Water Resources Development Co.,Ltd. |