Air station operation and maintenance method and system based on big data

Abstract

The invention relates to the technical field of air quality monitoring, in particular to an air station operation and maintenance method and system based on big data. The method comprises the following steps: training an air parameter fluctuation abnormity automatic identification model of each time interval, inputting a first actual air parameter set collected by an underway vehicle into the air parameter fluctuation abnormity automatic identification model of the time interval, and obtaining a first identification result; and parameters collected by the air station are input into the model to obtain a second identification result, and then whether each sensor of the target air station needs to be calibrated or not is judged, so that comprehensive evaluation of data measured by the underway vehicle in a period of time and data monitored by the monitoring station in a period of time is realized, and the accuracy of calibration is improved. Therefore, whether the data acquired by each sensor of the monitoring station has credibility is evaluated, the problem of system misjudgment caused by setting of a single threshold value or a plurality of threshold values is avoided to a great extent, and the intelligent degree, the reliability and the usability of operation and maintenance of the air station are greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of air quality monitoring, and in particular to an air station operation and maintenance method and system based on big data. Background Art

[0002] Currently, most of the micro air monitoring stations are distributed in a grid-like manner and monitor based on the principle of electrochemical sensors. They have the characteristics of a large number of areas, great influence from the outside world, and difficulty in controlling data quality.

[0003] The following scheme is generally adopted for the calibration of existing micro air stations: factory calibration, within the range of the micro air station, the standard gas concentration is introduced, the gas concentration C is linearly correlated with the electrical signal value x, and the slope k and intercept are calculated; the specific method of factory inspection consistency is: multiple micro air stations are concentrated in an area with good environmental consistency, and linear regression analysis is performed to evaluate the consistency of the equipment. Since the current data quality control method only calibrates and adjusts the consistency of the equipment before leaving the factory, and fails to achieve data quality control throughout the entire process, it is difficult to control during daily operation and maintenance. During operation, existing air stations need to use cruising vehicles to calibrate data. In the prior art, the judgment of whether the sensors of the air station are abnormal by cruising vehicles is often based on the setting of a single or multiple thresholds. However, since the thresholds are all obtained through manual experience, there are large errors. Although the local air concentration is difficult to mutate or fluctuate due to the strong fluidity of the atmosphere, there is still the possibility of mutation. At this time, the setting of a single or multiple thresholds often leads to system misjudgment. The prior art is unable to comprehensively evaluate the data measured by the cruising vehicle over a period of time and the data monitored by the monitoring station over a period of time, and then evaluate whether the data collected by the sensors of the monitoring station is credible, thereby determining whether it is necessary to operate and maintain the sensors of the air station. Therefore, the prior art is in urgent need of a solution that can solve the above technical problems.

[0004] Therefore, the prior art needs to be further developed. Summary of the invention

[0005] The purpose of the present invention is to overcome the above-mentioned technical deficiencies and provide an air station operation and maintenance method and system based on big data to solve the problems existing in the prior art.

[0006] To achieve the above technical objectives, according to a first aspect of the present invention, the present invention provides an air station operation and maintenance method based on big data, comprising: S100, obtain the GPS positioning coordinates of the target air station, establish the target monitoring area of ​​the target air station according to the GPS positioning coordinates of the target air station and the preset radius, after the construction of the target air station is completed and before it is put into use, evenly divide the 24 hours of each day into 24 time intervals, obtain the air parameters of the preset number of times according to the first preset time interval in each time interval, record them as the first air parameter set of each time interval, obtain the first air parameter set of the preset number of days of each time interval, record them as the target air parameter set of each time interval, divide the target air parameter set of each time interval into a training data set and a verification data set according to a first preset ratio; use the training data set to train the automatic recognition model of air parameter fluctuation anomalies in each time interval, use the verification data set to optimize the trained automatic recognition model of air parameter fluctuation anomalies in each time interval, optimize the model parameters, and complete the training of the automatic recognition model of air parameter fluctuation anomalies in each time interval; S200. After the air station is put into use, the GPS positioning coordinates of the target vehicle of the target cruising vehicle equipped with the automatic electrical measuring station for ambient air quality are obtained, and then it is determined whether the target cruising vehicle has entered the target monitoring area of ​​the target air station. If so, the current time is obtained, and the air parameter fluctuation abnormality automatic recognition model of the time interval corresponding to the current time is called. The air parameters collected by the target cruising vehicle for a preset number of times are obtained according to a second preset time interval, and recorded as the first actual air parameter set of the time interval. The first actual air parameter set is input into the air parameter fluctuation abnormality automatic recognition model of the time interval to obtain a first recognition result. The air parameters collected by the target air station for a preset number of times are obtained according to the second preset time interval, and recorded as the second actual air parameter set of the time interval. The second actual air parameter set is input into the air parameter fluctuation abnormality automatic recognition model of the time interval to obtain a second recognition result. S300, judging whether the first recognition result and the second recognition result are consistent, and judging whether each sensor of the target air station needs to be calibrated according to the judgment result.

[0007] Specifically, the first recognition result and the second recognition result both include abnormal air parameter fluctuation and normal air parameter fluctuation, and determining whether the first recognition result and the second recognition result are consistent includes: If both the first recognition result and the second recognition result indicate that the air parameter fluctuation is abnormal, or both the first recognition result and the second recognition result indicate that the air parameter fluctuation is abnormal and normal, it is determined that the first recognition result and the second recognition result are consistent; If the first recognition result is that the air parameter fluctuation is abnormal, and the second recognition result is that the air parameter fluctuation is normal, it is determined that the first recognition result and the second recognition result are inconsistent; If the first recognition result is that the air parameter fluctuation is normal, and the second recognition result is that the air parameter fluctuation is abnormal, it is determined that the first recognition result and the second recognition result are inconsistent.

[0008] Specifically, the determining whether the first recognition result and the second recognition result are consistent, and determining whether each sensor of the target air station needs to be calibrated according to the determination result, includes: If the first recognition result and the second recognition result are consistent, it is determined that the sensors of the target air station do not need to be calibrated, and a prompt signal is outputted indicating that the sensors of the target air station do not need to be calibrated; If the first recognition result and the second recognition result are inconsistent, it is determined that the sensors of the target air station need to be calibrated, and an alarm signal indicating that the sensors of the target air station need to be calibrated is output.

[0009] Specifically, the method further includes: If it is determined that the sensors of the target air station need to be calibrated, the target air parameter set of the time interval is used to perform zero point calibration or span calibration on the sensors of the target air station.

[0010] Specifically, the method further includes: If the target cruising vehicle does not enter the target monitoring area of ​​the target air station, the current time is not obtained, and the air parameter fluctuation anomaly automatic recognition model for the time interval corresponding to the current time is not called.

[0011] Specifically, the method of using the training data set to train the air parameter fluctuation abnormal automatic recognition model for each time interval, using the verification data set to optimize the trained air parameter fluctuation abnormal automatic recognition model for each time interval, optimizing the model parameters, and completing the air parameter fluctuation abnormal automatic recognition model training for each time interval includes: The training data set is input into the preset network layer in batches for training, and the preset network layer includes a Transformer network layer, and the Transformer network layer is used to predict the air at the next time point of a certain time interval according to the forward propagation of the air parameters at the current time point of the time interval, and then obtain the predicted loss value, and the time interval between the current time point and the next time point is the first preset time interval; the loss value of the preset network layer is calculated and input into the optimizer for optimization, and the direction in which the parameter gradient of the automatic identification model of air parameter fluctuation anomalies in each time interval is determined to be the fastest decreasing direction; the automatic identification model of air parameter fluctuation anomalies in each time interval is back-propagated according to the loss value and the parameter gradient of the model, and the parameters of the automatic identification model of air parameter fluctuation anomalies in each time interval are optimized.

[0012] Specifically, the method uses the training data set to train the automatic recognition model for abnormal air parameter fluctuations in each time interval, uses the verification data set to optimize the trained automatic recognition model for abnormal air parameter fluctuations in each time interval, optimizes the model parameters, and completes the training of the automatic recognition model for abnormal air parameter fluctuations in each time interval, and also includes: After each training, the verification data set is input into the preset network layer of the previous training in batches for model parameter verification, and cyclic training is performed, with the total number of training rounds set to the first preset round; the loss value of the preset network layer is recorded, and it is determined whether the loss value meets the first preset condition, and whether to end the training and output the model parameters based on the judgment result.

[0013] Specifically, the determining whether the loss value satisfies the first preset condition, and determining whether to end the training and output the model parameters according to the determination result, includes: If the loss value meets the first preset condition, the training is terminated and the parameters of the current air parameter fluctuation abnormal automatic identification model are output; if the loss value does not meet the first preset condition, the training continues.

[0014] Specifically, the first preset condition is: After the second preset round of training, the loss values ​​obtained in the next round of training are greater than or equal to the loss values ​​that occurred during the second preset round of training.

[0015] According to a second aspect of the present invention, there is provided an air station operation and maintenance system based on big data, comprising: An acquisition module is used to acquire the GPS positioning coordinates of a target air station; after the air station is put into use, it is used to acquire the GPS positioning coordinates of a target vehicle carrying an automatic electrical measurement station for ambient air quality; A control module is used to establish a target monitoring area of ​​the target air station according to the GPS positioning coordinates and a preset radius of the target air station. After the construction of the target air station is completed and before it is put into use, the 24 hours of each day are evenly divided into 24 time intervals. The air parameters of the preset number of times obtained at the first preset time interval in each time interval are recorded as the first air parameter set of each time interval, and the first air parameter set of the preset number of days in each time interval is obtained, which is recorded as the target air parameter set of each time interval. The target air parameter set of each time interval is divided into a training data set and a verification data set according to a first preset ratio; the training data set is used to train an automatic recognition model for abnormal air parameter fluctuations in each time interval, and the verification data set is used to optimize the trained automatic recognition model for abnormal air parameter fluctuations in each time interval, optimize the model parameters, and complete the automatic recognition of abnormal air parameter fluctuations in each time interval. Model training; used to determine whether the target cruising vehicle enters the target monitoring area of ​​the target air station. If so, obtain the current time, call the automatic recognition model for abnormal air parameter fluctuations in the time interval corresponding to the current time, obtain the air parameters collected by the target cruising vehicle for a preset number of times according to a second preset time interval, record it as the first actual air parameter set for the time interval, input the first actual air parameter set into the automatic recognition model for abnormal air parameter fluctuations in the time interval, and obtain a first recognition result; obtain the air parameters collected by the target air station for a preset number of times according to the second preset time interval, record it as the second actual air parameter set for the time interval, input the second actual air parameter set into the automatic recognition model for abnormal air parameter fluctuations in the time interval, and obtain a second recognition result; used to determine whether the first recognition result and the second recognition result are consistent, and determine whether the sensors of the target air station need to be calibrated based on the judgment result.

[0016] Beneficial effects: The present invention evenly divides 24 hours a day into 24 time intervals. After the target air station is built and before it is put into use, an automatic recognition model for abnormal air parameter fluctuations in each time interval is trained. After the air station is put into use, the target vehicle GPS positioning coordinates and the current time interval of the target cruising vehicle equipped with an automatic electrical measuring station for ambient air quality are obtained, and then it is determined whether the target cruising vehicle enters the target monitoring area of ​​the target air station. If so, a first actual air parameter set collected by the cruising vehicle is input into the automatic recognition model for abnormal air parameter fluctuations in the time interval to obtain a first recognition result; the air parameters collected by the target air station for a preset number of times are obtained at a second preset time interval, which is recorded as the first actual air parameter set collected by the target air station. A second actual air parameter set for the time interval is input into the automatic recognition model for air parameter fluctuation anomalies in the time interval to obtain a second recognition result. By judging whether the first recognition result is consistent with the second recognition result, and judging whether the sensors of the target air station need to be calibrated according to the judgment result, a comprehensive evaluation is achieved on the data measured by the cruising vehicle over a period of time and the data monitored by the monitoring station over a period of time, and then whether the data collected by the sensors of the monitoring station are credible is evaluated. This avoids the problem that the setting of a single or multiple thresholds often leads to system misjudgment to a large extent, and greatly improves the intelligence, reliability and availability of air station operation and maintenance. BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 It is a flowchart of an air station operation and maintenance method based on big data provided in a specific embodiment of the present invention; Figure 2 It is a schematic diagram of the system composition of the air station operation and maintenance system based on big data provided in a specific embodiment of the present invention; Figure 3 It is a structural schematic diagram of an automatic identification model for abnormal air parameter fluctuations provided in a specific embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to enable those skilled in the art to better understand the technical solution of the present invention, the technical solution of the present invention is clearly and completely described below in conjunction with the accompanying drawings of the present invention. Based on the embodiments in this application, other similar embodiments obtained by ordinary technicians in this field without making creative work should all fall within the scope of protection of this application. In addition, the directional words mentioned in the following embodiments, such as "up", "down", "left", "right", etc., are only reference to the directions of the accompanying drawings. Therefore, the directional words used are used to illustrate rather than limit the invention.

[0018] The present invention will be further described below in conjunction with the accompanying drawings and preferred embodiments.

[0019] See also Figure 1 and Figure 3The present invention provides an air station operation and maintenance method based on big data, comprising: S100. Obtain the GPS positioning coordinates of the target air station, establish the target monitoring area of ​​the target air station according to the GPS positioning coordinates of the target air station and the preset radius, and after the construction of the target air station is completed and before it is put into use, evenly divide the 24 hours of each day into 24 time intervals, obtain the air parameters of a preset number of times at a first preset time interval in each time interval, record them as the first air parameter set of each time interval, obtain the first air parameter set of a preset number of days in each time interval, record them as the target air parameter set of each time interval, and divide the target air parameter set of each time interval into a training data set and a verification data set according to a first preset ratio; use the training data set to train the automatic recognition model for air parameter fluctuation anomalies in each time interval, use the verification data set to optimize the trained automatic recognition model for air parameter fluctuation anomalies in each time interval, optimize the model parameters, and complete the training of the automatic recognition model for air parameter fluctuation anomalies in each time interval.

[0020] Specifically, the method of using the training data set to train the automatic recognition model for abnormal air parameter fluctuations in each time interval, using the verification data set to optimize the trained automatic recognition model for abnormal air parameter fluctuations in each time interval, optimizing the model parameters, and completing the training of the automatic recognition model for abnormal air parameter fluctuations in each time interval includes: The training data set is input into the preset network layer in batches for training, and the preset network layer includes a Transformer network layer, and the Transformer network layer is used to predict the air at the next time point of a certain time interval according to the forward propagation of the air parameters at the current time point of the time interval, and then obtain the predicted loss value, and the time interval between the current time point and the next time point is the first preset time interval; the loss value of the preset network layer is calculated and input into the optimizer for optimization, and the direction in which the parameter gradient of the automatic identification model of air parameter fluctuation anomalies in each time interval is determined to be the fastest decreasing direction; the automatic identification model of air parameter fluctuation anomalies in each time interval is back-propagated according to the loss value and the parameter gradient of the model, and the parameters of the automatic identification model of air parameter fluctuation anomalies in each time interval are optimized.

[0021] It can be understood that Transformer is a deep learning model based on the self-attention mechanism, which consists of a multi-head attention mechanism and a feedforward neural network. It can process sequence data and capture long-distance dependencies. The Transformer model does not contain traditional convolutional layers, but processes each element in the input sequence through an attention layer. The typical structure of Transformer includes an encoder and a decoder, wherein the encoder is composed of multiple identical layers stacked together, and the decoder adds a self-attention layer on the basis of the encoder to process sequence generation tasks. The present invention uses Transformer to supplement global context information, enhances the model's capabilities in feature extraction and global understanding, greatly improves the performance of the model, and greatly reduces the computational cost.

[0022] Specifically, the preset network layer also includes a ResNet network layer, which is used to classify and predict the output results according to the target air parameter sets of each time interval collected in the training data set and the verification data set, and the classification results include abnormal air parameter fluctuations and normal air parameter fluctuations.

[0023] It should be noted here that ResNet is a deep convolutional neural network that solves the gradient vanishing problem in deep network training by introducing residual connections. The core component of ResNet is the residual block, which usually contains two or three convolutional layers and a jump connection, which allows the gradient to bypass these convolutional layers directly. The network structure of ResNet can be very deep. These networks extract the features of pressure data by stacking multiple residual blocks, and classify and predict the output results through global average pooling and fully connected layers. The classification results include abnormal air parameter fluctuations and normal air parameter fluctuations.

[0024] See also Figure 3 The present invention combines the ResNet network layer and the Transformer network layer. ResNet is good at extracting local features of air parameters in the training data set, while Transformer can supplement the global context information. This combination can enhance the model's capabilities in feature extraction and global understanding, greatly improve the performance of the model, and greatly reduce the computational cost.

[0025] It can be understood that the ResNet network layer is used to classify and predict the output results through the target air parameter sets of each time interval collected in the training data set and the validation data set. The specific steps are: (1) First, put the training data set into the Backbones network layer of the model. Backbone refers to a series of convolutional layers that constitute the skeleton of the neural network. The main function of these layers is to extract the features of the input data. The Backbone network is usually composed of multiple convolutional layers, pooling layers, and activation functions. It can extract meaningful feature representations from the original data. Backbones is composed of multiple Backbones, and the core network layer in Backbone is the Resnet network layer. After the data set is input into the model, the target air parameter set will be input into Backbones for training; (2) When constructing Backbone, the target air station code and Resnet network layer were added. After the training data set was input, Resnet began to train layer by layer to extract features from the target air parameter set, assisting Transformer in predicting whether there is an abnormality in the air parameter fluctuation. The extracted features were used with the help of the target air station code to obtain the code of the target air station, which facilitated maintenance by the staff.

[0026] Specifically, the method uses the training data set to train the automatic recognition model for abnormal air parameter fluctuations in each time interval, uses the verification data set to optimize the trained automatic recognition model for abnormal air parameter fluctuations in each time interval, optimizes the model parameters, and completes the training of the automatic recognition model for abnormal air parameter fluctuations in each time interval, and also includes: After each training, the verification data set is input into the preset network layer of the previous training in batches for model parameter verification, and cyclic training is performed, with the total number of training rounds set to the first preset round; the loss value of the preset network layer is recorded, and it is determined whether the loss value meets the first preset condition, and whether to end the training and output the model parameters based on the judgment result.

[0027] Specifically, after each training, the verification data set is input into the preset network layer of the previous training in batches to verify the model parameters, and the cyclic training is performed, including the following methods: When the number of training times is not enough and the model needs to be optimized, the training can only be performed from the beginning. Therefore, the present invention develops a method for continuing the following training based on the previously trained model parameter file; (1) Provide custom parameters, which can be set according to user needs to continue training based on any training parameters; (2) Load the parameters of the custom model file; (3) Put the verification data set into the custom parameter network layer for verification, obtain the loss value and record the model parameters of the model corresponding to the minimum loss value during the training process; (4) Based on this set of model parameters, forward propagation training and back propagation are performed, and the optimizer optimizes the parameters to obtain a new set of data; (5) Put the verification data into the new data network layer for inference to obtain the loss value; (6) Perform cyclic training to obtain the best model and complete model optimization.

[0028] It is understandable that the training method adopted by the present invention is batch training, which means that each time the model parameters are updated, only a part of the samples in the validation data set is used, which is called a batch. The advantages of batch training are that it can reduce memory consumption, speed up training, increase randomness, and facilitate the generalization of the model.

[0029] It should be noted here that, in the present invention, batch_size=2, and the batch_size is the batch size. For example, when batch_size=2, it means that two target air parameter sets are selected from the verification data set each time and put into the model for verification.

[0030] Specifically, the determining whether the loss value satisfies the first preset condition, and determining whether to end the training and output the model parameters according to the determination result, includes: If the loss value meets the first preset condition, the training is terminated and the parameters of the current air parameter fluctuation abnormal automatic identification model are output; if the loss value does not meet the first preset condition, the training continues.

[0031] Specifically, the first preset condition is: After the second preset round of training, the loss values ​​obtained in the next round of training are greater than or equal to the loss values ​​that occurred during the second preset round of training.

[0032] It can be understood that before step S100, the following steps are included: A preset radius, a preset number of times, a preset number of days, a first preset time interval, a second preset time interval, a first preset ratio, a first preset round, and a second preset round are preset in the control module.

[0033] It can be understood that the preset radius, preset number of times, preset number of days, first preset time interval, second preset time interval, first preset ratio, first preset round, and second preset round can be specifically set according to the actual needs of the user of the present invention. The present invention does not limit the specific values ​​of the above parameters, and there is no regularity in setting the specific values ​​of the above parameters, as long as it is applicable to the air station operation and maintenance method based on big data proposed in the present invention.

[0034] Preferably, the present invention sets the preset radius to 500 meters, the present invention sets the preset number of times to 60 times, the present invention sets the first preset time interval to 1 minute, the present invention sets the second preset time interval to 10 seconds, the present invention sets the first preset ratio to 8:2, the present invention sets the first preset rounds to 30,000 times, and the present invention sets the second preset rounds to 300. The above settings can further improve the performance of the model and the efficiency of model training, and greatly improve the accuracy and reliability of the prediction of the automatic identification model of air parameter fluctuation anomalies in each time interval of the present invention.

[0035] It can be understood that when the sensors of the target air station are normal, due to the strong fluidity of the atmosphere, the local air concentration is less likely to change suddenly or fluctuate. The air parameters collected by the sensors of the target air station should have a certain coupling law within a period of time. For example, the air parameters should have a certain dynamic equilibrium relationship within a period of time. When the sensors of the target air station are abnormal, this equilibrium relationship will be broken. Based on this, the present invention designs an automatic identification model for abnormal air parameter fluctuations in each time interval.

[0036] It should be noted here that the air parameters include: SO2 concentration, CO concentration, O3 concentration, PM2.5, PM10 and multiple pollutant parameters.

[0037] It is understandable that since the working hours of the cruising vehicle may be relatively tight, the present invention sets the preset number of times to 60 times for matching the input dimension of the model, and sets the first preset time interval to 1 minute. The present invention sets the second preset time interval to 10 seconds, which greatly saves the detection time of the cruising vehicle and greatly improves the detection efficiency of each sensor of the target air station.

[0038] It can be understood that the present invention preferably sets the maximum round of cyclic training to 30,000 times, which can effectively ensure the accuracy of the model prediction of the present invention and record the loss value of the model training. The present invention sets the second preset round to 300, that is, when the loss value of the training no longer decreases within 300 rounds of training, the round with the smallest loss value in the 300 rounds of training is saved as the best verification round parameter to generate the best model file. The above setting makes the training rounds no longer rely on manual judgment, but uses the deep learning model for automatic training, which improves the accuracy of the model prediction, and automatically ends the training when the accuracy meets the requirements, which greatly saves the training time, and can also effectively prevent overfitting, which greatly improves the intelligence level of the present invention and the efficiency of model training.

[0039] Furthermore, the specific process of the air parameter fluctuation abnormality automatic identification model in each time interval of the present invention is as follows: (1) Obtain the potential vector mean mu and logarithmic variance logvar of the air parameter fluctuation anomaly automatic recognition model in each time interval during the forward propagation process; (2) Calculate a reasonable loss value representing the training effect: Calculate the divergence: k1=-0.5*(1+logvar-mu^2-(e^logvar)); Calculate the loss between the air parameters in the validation dataset and the predicted values: loss=L1+k1*k1_weight; Among them, loss is the loss value, k1 is the divergence, k1_weight is the proportion of k1, the value is 10, L1 is the absolute difference between the predicted value and the true value, and the calculation formula of L1 is as follows: ; Among them, A is the L1 loss in pytorch, which calculates the average or sum of the absolute differences between two tensors; A has a shape of N*C*H*W, which indicates that it is a four-dimensional tensor, N represents the number of samples (Batch Size), indicating that there are N data points, C represents the number of channels (Channels), H and W represent the height and width respectively, which are usually the two-dimensional spatial dimensions of the input features; P represents the padding value, which is a Boolean type, and the shape of P is c*h*w*1. The dimensions of P are the same as those of A in terms of channel, height, and width, but there is no sample number N, which means that P is a mask or padding information for a feature map of each sample, and the Boolean value of P may be used to mark whether the elements in A are filled (True) or valid (False).

[0040] (3) The loss value when using the validation data set to validate the current model network; (4) If the validation loss of this round is less than that of the previous round, the best model parameters are replaced with the training model of this round; (5) Put the rounds and loss values ​​into the dictionary one by one; (6) Compare the loss value of the current round with the loss values ​​of the 300 rounds before this round (parameters are provided and can be set according to actual needs, the default is 300), that is, traverse the loss values ​​from epoch-300 to epoch-1 (epoch is the training round, and comparison is performed when epoch is greater than 300). If the loss values ​​of the current round of epoch are all greater than the loss values ​​from (epoch-300) to (epoch-1), then stop training.

[0041] (7) Save the best model parameters in the cyclic verification as a ckpt model file.

[0042] Specifically, the method further includes: After the target air station is installed and before it is put into use, it is necessary to test the installed target air station to ensure that there are no abnormalities in the sensors of the target air station.

[0043] It is understandable that how to ensure that there are no abnormalities in the sensors of the target air station can be checked by relevant staff according to industry standards, and the present invention will not go into details here.

[0044] S200. After the air station is put into use, the GPS positioning coordinates of the target vehicle of the target cruising vehicle equipped with the automatic electrical measuring station for ambient air quality are obtained, and then it is determined whether the target cruising vehicle has entered the target monitoring area of ​​the target air station. If so, the current time is obtained, and the automatic recognition model for abnormal air parameter fluctuations in the time interval corresponding to the current time is called. The air parameters collected by the target cruising vehicle for a preset number of times are obtained at a second preset time interval, recorded as a first actual air parameter set for the time interval, and the first actual air parameter set is input into the automatic recognition model for abnormal air parameter fluctuations in the time interval to obtain a first recognition result. The air parameters collected by the target air station for a preset number of times are obtained at a second preset time interval, recorded as a second actual air parameter set for the time interval, and the second actual air parameter set is input into the automatic recognition model for abnormal air parameter fluctuations in the time interval to obtain a second recognition result.

[0045] Specifically, the method further includes: If the target cruising vehicle does not enter the target monitoring area of ​​the target air station, the current time is not obtained, and the air parameter fluctuation anomaly automatic recognition model for the time interval corresponding to the current time is not called.

[0046] It should be noted here that in order to ensure the accuracy of the data, a cruising vehicle equipped with an automatic ambient air quality monitoring station that meets the "HJ / T193-2005 Technical Specifications for Automatic Monitoring of Ambient Air Quality" is used to collect air parameters.

[0047] S300, judging whether the first recognition result and the second recognition result are consistent, and judging whether each sensor of the target air station needs to be calibrated according to the judgment result.

[0048] Specifically, the first recognition result and the second recognition result both include abnormal air parameter fluctuation and normal air parameter fluctuation, and determining whether the first recognition result and the second recognition result are consistent includes: If both the first recognition result and the second recognition result indicate that the air parameter fluctuation is abnormal, or both the first recognition result and the second recognition result indicate that the air parameter fluctuation is abnormal and normal, it is determined that the first recognition result and the second recognition result are consistent; If the first recognition result is that the air parameter fluctuation is abnormal, and the second recognition result is that the air parameter fluctuation is normal, it is determined that the first recognition result and the second recognition result are inconsistent; If the first recognition result is that the air parameter fluctuation is normal, and the second recognition result is that the air parameter fluctuation is abnormal, it is determined that the first recognition result and the second recognition result are inconsistent.

[0049] Specifically, the determining whether the first recognition result and the second recognition result are consistent, and determining whether each sensor of the target air station needs to be calibrated according to the determination result, includes: If the first recognition result and the second recognition result are consistent, it is determined that the sensors of the target air station do not need to be calibrated, and a prompt signal is outputted indicating that the sensors of the target air station do not need to be calibrated; If the first recognition result and the second recognition result are inconsistent, it is determined that the sensors of the target air station need to be calibrated, and an alarm signal indicating that the sensors of the target air station need to be calibrated is output.

[0050] Specifically, the method further includes: If it is determined that the sensors of the target air station need to be calibrated, the target air parameter set of the time interval is used to perform zero point calibration or span calibration on the sensors of the target air station.

[0051] It can be understood that the present invention evenly divides the 24 hours of each day into 24 time intervals. After the target air station is built and before it is put into use, an automatic recognition model for abnormal air parameter fluctuations in each time interval is trained. After the air station is put into use, the target vehicle GPS positioning coordinates and the current time interval of the target cruising vehicle equipped with an automatic electrical measuring station for ambient air quality are obtained, and then it is determined whether the target cruising vehicle enters the target monitoring area of ​​the target air station. If so, the first actual air parameter set collected by the cruising vehicle is input into the automatic recognition model for abnormal air parameter fluctuations in the time interval to obtain a first recognition result; the air parameters collected by the target air station for a preset number of times are obtained at a second preset time interval. , recorded as the second actual air parameter set of the time interval, the second actual air parameter set is input into the automatic recognition model of air parameter fluctuation anomalies in the time interval to obtain a second recognition result, and by judging whether the first recognition result is consistent with the second recognition result, and judging whether the sensors of the target air station need to be calibrated according to the judgment result, a comprehensive evaluation is achieved on the data measured by the cruising vehicle over a period of time and the data monitored by the monitoring station over a period of time, and then whether the data collected by the sensors of the monitoring station are credible is evaluated, which largely avoids the problem that the setting of a single or multiple thresholds often leads to system misjudgment, and greatly improves the intelligence, reliability and availability of air station operation and maintenance. See also Figure 2The present invention provides another embodiment, which provides an air station operation and maintenance system based on big data, and the air station operation and maintenance system based on big data includes: The acquisition module 100 is used to acquire the GPS positioning coordinates of the target air station; after the air station is put into use, the target vehicle GPS positioning coordinates of the target vehicle equipped with the automatic electrical measurement station of the ambient air quality are acquired; The control module 200 is used to establish a target monitoring area of ​​the target air station according to the GPS positioning coordinates and a preset radius of the target air station. After the construction of the target air station is completed and before it is put into use, the 24 hours of each day are evenly divided into 24 time intervals. The air parameters of a preset number of times obtained at a first preset time interval in each time interval are recorded as the first air parameter set of each time interval, and the first air parameter set of a preset number of days in each time interval is obtained, which is recorded as the target air parameter set of each time interval. The target air parameter set of each time interval is divided into a training data set and a verification data set according to a first preset ratio; the training data set is used to train an automatic recognition model for abnormal air parameter fluctuations in each time interval, and the verification data set is used to optimize the trained automatic recognition model for abnormal air parameter fluctuations in each time interval, optimize the model parameters, and complete the automatic recognition model for abnormal air parameter fluctuations in each time interval. Identification model training; used to determine whether the target cruising vehicle enters the target monitoring area of ​​the target air station, if so, obtain the current time, call the air parameter fluctuation abnormal automatic identification model of the time interval corresponding to the current time, obtain the air parameters collected by the target cruising vehicle for a preset number of times according to a second preset time interval, record it as the first actual air parameter set of the time interval, input the first actual air parameter set into the air parameter fluctuation abnormal automatic identification model of the time interval, and obtain a first identification result; obtain the air parameters collected by the target air station for a preset number of times according to the second preset time interval, record it as the second actual air parameter set of the time interval, input the second actual air parameter set into the air parameter fluctuation abnormal automatic identification model of the time interval, and obtain a second identification result; used to determine whether the first identification result and the second identification result are consistent, and determine whether the sensors of the target air station need to be calibrated according to the judgment result.

[0052] It should be noted here that the present invention evenly divides the 24 hours of each day into 24 time intervals. After the target air station is built and before it is put into use, an automatic recognition model for abnormal air parameter fluctuations in each time interval is trained. After the air station is put into use, the target vehicle GPS positioning coordinates and the current time interval of the target cruising vehicle equipped with an automatic electrical measuring station for ambient air quality are obtained, and then it is determined whether the target cruising vehicle enters the target monitoring area of ​​the target air station. If so, the first actual air parameter set collected by the cruising vehicle is input into the automatic recognition model for abnormal air parameter fluctuations in the time interval to obtain a first recognition result; the air parameters collected by the target air station for a preset number of times are obtained at a second preset time interval. The second actual air parameter set is input into the air parameter fluctuation anomaly automatic recognition model of the time interval to obtain a second recognition result. By judging whether the first recognition result is consistent with the second recognition result, and judging whether the sensors of the target air station need to be calibrated according to the judgment result, a comprehensive evaluation is made on the data measured by the cruising vehicle over a period of time and the data monitored by the monitoring station over a period of time, and then whether the data collected by the sensors of the monitoring station are credible is evaluated. This avoids the problem that the setting of a single or multiple thresholds often leads to system misjudgment to a large extent, and greatly improves the intelligence, reliability and availability of air station operation and maintenance. In a preferred embodiment, the present application further provides an electronic device, the electronic device comprising: A memory; and a processor, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the air station operation and maintenance method based on big data is implemented. The computer device can be broadly a server, a terminal, or any other electronic device with necessary computing and / or processing capabilities. In one embodiment, the computer device may include a processor, a memory, a network interface, a communication interface, etc. connected via a system bus. The processor of the computer device can be used to provide necessary computing, processing and / or control capabilities. The memory of the computer device may include a non-volatile storage medium and an internal memory. An operating system, a computer program, etc. may be stored in or on the non-volatile storage medium. The internal memory can provide an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface and the communication interface of the computer device can be used to connect and communicate with external devices through a network. When the computer program is executed by the processor, the steps of the method of the present invention are performed.

[0053] The present invention may be implemented as a computer-readable storage medium having a computer program stored thereon, which causes the steps of the method of an embodiment of the present invention to be executed when executed by a processor. In one embodiment, the computer program is distributed on a plurality of computer devices or processors coupled to a network so that the computer program is stored, accessed and executed in a distributed manner by one or more computer devices or processors. A single method step / operation, or two or more method steps / operations, may be performed by a single computer device or processor or by two or more computer devices or processors. One or more method steps / operations may be performed by one or more computer devices or processors, and one or more other method steps / operations may be performed by one or more other computer devices or processors. One or more computer devices or processors may perform a single method step / operation, or perform two or more method steps / operations.

[0054] It will be appreciated by a person skilled in the art that the method steps of the present invention may be performed by instructing related hardware such as a computer device or a processor through a computer program, and the computer program may be stored in a non-temporary computer-readable storage medium, which causes the steps of the present invention to be performed when the computer program is executed. Depending on the circumstances, any reference to memory, storage, database, or other media herein may include non-volatile and / or volatile memory. Examples of non-volatile memory include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, magnetic tape, floppy disk, magneto-optical data storage device, optical data storage device, hard disk, solid state disk, etc. Examples of volatile memory include random access memory (RAM), external cache memory, etc.

[0055] It can be understood that the present invention evenly divides the 24 hours of each day into 24 time intervals. After the target air station is built and before it is put into use, an automatic recognition model for abnormal air parameter fluctuations in each time interval is trained. After the air station is put into use, the target vehicle GPS positioning coordinates and the current time interval of the target cruising vehicle equipped with an automatic electrical measuring station for ambient air quality are obtained, and then it is determined whether the target cruising vehicle enters the target monitoring area of ​​the target air station. If so, the first actual air parameter set collected by the cruising vehicle is input into the automatic recognition model for abnormal air parameter fluctuations in the time interval to obtain a first recognition result; the air parameters collected by the target air station for a preset number of times are obtained at a second preset time interval. , recorded as the second actual air parameter set of the time interval, the second actual air parameter set is input into the automatic recognition model of air parameter fluctuation anomalies in the time interval to obtain a second recognition result, and by judging whether the first recognition result is consistent with the second recognition result, and judging whether the sensors of the target air station need to be calibrated according to the judgment result, a comprehensive evaluation is achieved on the data measured by the cruising vehicle over a period of time and the data monitored by the monitoring station over a period of time, and then whether the data collected by the sensors of the monitoring station are credible is evaluated, which largely avoids the problem that the setting of a single or multiple thresholds often leads to system misjudgment, and greatly improves the intelligence, reliability and availability of air station operation and maintenance. The various technical features described above can be combined arbitrarily. Although all possible combinations of these technical features are not described, any combination of these technical features should be considered to be covered by this specification as long as there is no contradiction in such combination.

[0056] The specific implementation of the present invention described above does not constitute a limitation on the protection scope of the present invention. Any other corresponding changes and modifications made based on the technical concept of the present invention should be included in the protection scope of the claims of the present invention.

Claims

1. An air station operation and maintenance method based on big data, characterized in that: The method comprises: S100, obtain the GPS positioning coordinates of the target air station, establish the target monitoring area of ​​the target air station according to the GPS positioning coordinates of the target air station and the preset radius, after the construction of the target air station is completed and before it is put into use, evenly divide the 24 hours of each day into 24 time intervals, obtain the air parameters of the preset number of times according to the first preset time interval in each time interval, record them as the first air parameter set of each time interval, obtain the first air parameter set of the preset number of days of each time interval, record them as the target air parameter set of each time interval, divide the target air parameter set of each time interval into a training data set and a verification data set according to a first preset ratio; use the training data set to train the automatic recognition model of air parameter fluctuation anomalies in each time interval, use the verification data set to optimize the trained automatic recognition model of air parameter fluctuation anomalies in each time interval, optimize the model parameters, and complete the training of the automatic recognition model of air parameter fluctuation anomalies in each time interval; S200. After the air station is put into use, the GPS positioning coordinates of the target vehicle of the target cruising vehicle equipped with the automatic electrical measuring station for ambient air quality are obtained, and then it is determined whether the target cruising vehicle has entered the target monitoring area of ​​the target air station. If so, the current time is obtained, and the air parameter fluctuation abnormality automatic recognition model of the time interval corresponding to the current time is called. The air parameters collected by the target cruising vehicle for a preset number of times are obtained according to a second preset time interval, and recorded as the first actual air parameter set of the time interval. The first actual air parameter set is input into the air parameter fluctuation abnormality automatic recognition model of the time interval to obtain a first recognition result. The air parameters collected by the target air station for a preset number of times are obtained according to the second preset time interval, and recorded as the second actual air parameter set of the time interval. The second actual air parameter set is input into the air parameter fluctuation abnormality automatic recognition model of the time interval to obtain a second recognition result. S300, judging whether the first recognition result and the second recognition result are consistent, and judging whether each sensor of the target air station needs to be calibrated according to the judgment result.

2. The air station operation and maintenance method based on big data according to claim 1 is characterized in that: The first recognition result and the second recognition result both include abnormal air parameter fluctuation and normal air parameter fluctuation, and the determining whether the first recognition result and the second recognition result are consistent includes: If both the first recognition result and the second recognition result indicate that the air parameter fluctuation is abnormal, or both the first recognition result and the second recognition result indicate that the air parameter fluctuation is abnormal and normal, it is determined that the first recognition result and the second recognition result are consistent; If the first recognition result is that the air parameter fluctuation is abnormal, and the second recognition result is that the air parameter fluctuation is normal, it is determined that the first recognition result and the second recognition result are inconsistent; If the first recognition result is that the air parameter fluctuation is normal, and the second recognition result is that the air parameter fluctuation is abnormal, it is determined that the first recognition result and the second recognition result are inconsistent.

3. The air station operation and maintenance method based on big data according to claim 1 is characterized in that: The determining whether the first recognition result and the second recognition result are consistent, and determining whether each sensor of the target air station needs to be calibrated according to the determination result, includes: If the first recognition result and the second recognition result are consistent, it is determined that the sensors of the target air station do not need to be calibrated, and a prompt signal is outputted indicating that the sensors of the target air station do not need to be calibrated; If the first recognition result and the second recognition result are inconsistent, it is determined that the sensors of the target air station need to be calibrated, and an alarm signal indicating that the sensors of the target air station need to be calibrated is output.

4. The air station operation and maintenance method based on big data according to claim 3 is characterized in that: The method further comprises: If it is determined that the sensors of the target air station need to be calibrated, the target air parameter set of the time interval is used to perform zero point calibration or span calibration on the sensors of the target air station.

5. The air station operation and maintenance method based on big data according to claim 1 is characterized in that: The method further comprises: If the target cruising vehicle does not enter the target monitoring area of ​​the target air station, the current time is not obtained, and the air parameter fluctuation anomaly automatic recognition model for the time interval corresponding to the current time is not called.

6. The air station operation and maintenance method based on big data according to claim 1 is characterized in that: The method of using the training data set to train the air parameter fluctuation abnormal automatic recognition model for each time interval, using the verification data set to optimize the trained air parameter fluctuation abnormal automatic recognition model for each time interval, optimizing the model parameters, and completing the air parameter fluctuation abnormal automatic recognition model training for each time interval includes: The training data set is input into the preset network layer in batches for training, and the preset network layer includes a Transformer network layer, and the Transformer network layer is used to predict the air at the next time point of a certain time interval according to the forward propagation of the air parameters at the current time point of the time interval, and then obtain the predicted loss value, and the time interval between the current time point and the next time point is the first preset time interval; the loss value of the preset network layer is calculated and input into the optimizer for optimization, and the direction in which the parameter gradient of the automatic identification model of air parameter fluctuation anomalies in each time interval is determined to be the fastest decreasing direction; the automatic identification model of air parameter fluctuation anomalies in each time interval is back-propagated according to the loss value and the parameter gradient of the model, and the parameters of the automatic identification model of air parameter fluctuation anomalies in each time interval are optimized.

7. The air station operation and maintenance method based on big data according to claim 1 is characterized in that: The method of using the training data set to train the air parameter fluctuation abnormal automatic recognition model for each time interval, using the verification data set to optimize the trained air parameter fluctuation abnormal automatic recognition model for each time interval, optimizing the model parameters, and completing the air parameter fluctuation abnormal automatic recognition model training for each time interval, further includes: After each training, the verification data set is input into the preset network layer of the previous training in batches for model parameter verification, and cyclic training is performed, with the total number of training rounds set to the first preset round; the loss value of the preset network layer is recorded, and it is determined whether the loss value meets the first preset condition, and whether to end the training and output the model parameters based on the judgment result.

8. The air station operation and maintenance method based on big data according to claim 7 is characterized in that: The determining whether the loss value satisfies the first preset condition, and determining whether to end the training and output the model parameters according to the determination result, includes: If the loss value meets the first preset condition, the training is terminated and the parameters of the current air parameter fluctuation abnormal automatic identification model are output; if the loss value does not meet the first preset condition, the training continues.

9. The air station operation and maintenance method based on big data according to claim 8 is characterized in that: The first preset condition is: After the second preset round of training, the loss values ​​obtained in the next round of training are greater than or equal to the loss values ​​that occurred during the second preset round of training.

10. An air station operation and maintenance system based on big data, characterized in that: include: An acquisition module is used to acquire the GPS positioning coordinates of a target air station; after the air station is put into use, it is used to acquire the GPS positioning coordinates of a target vehicle carrying an automatic electrical measurement station for ambient air quality; The control module is used to establish a target monitoring area of ​​the target air station according to the GPS positioning coordinates and the preset radius of the target air station. After the construction of the target air station is completed and before it is put into use, the 24 hours of each day are evenly divided into 24 time intervals. The air parameters of the preset number of times obtained at the first preset time interval in each time interval are recorded as the first air parameter set of each time interval, and the first air parameter set of the preset number of days of each time interval is obtained, which is recorded as the target air parameter set of each time interval. The target air parameter set of each time interval is divided into a training data set and a verification data set according to a first preset ratio; the training data set is used to train the air parameter fluctuation abnormality automatic detection system of each time interval. A dynamic recognition model is used to optimize the automatic recognition model for abnormal air parameter fluctuations in each time interval after training using a verification data set, optimize the model parameters, and complete the training of the automatic recognition model for abnormal air parameter fluctuations in each time interval; it is used to determine whether the target cruising vehicle has entered the target monitoring area of ​​the target air station, and if so, obtain the current time, call the automatic recognition model for abnormal air parameter fluctuations in the time interval corresponding to the current time, obtain the air parameters collected by the target cruising vehicle for a preset number of times according to a second preset time interval, record them as the first actual air parameter set for the time interval, input the first actual air parameter set into the automatic recognition model for abnormal air parameter fluctuations in the time interval, and obtain a first recognition result; Acquire the air parameters collected by the target air station for a preset number of times according to a second preset time interval, record them as a second actual air parameter set in the time interval, input the second actual air parameter set into the air parameter fluctuation abnormal automatic recognition model in the time interval, and obtain a second recognition result; It is used to judge whether the first recognition result and the second recognition result are consistent, and to judge whether the sensors of the target air station need to be calibrated according to the judgment result.

Images