Radon data abnormal fluctuation monitoring method and device, medium and product

Abstract

The invention discloses a radon data abnormal fluctuation monitoring method and device, a medium and a product, and relates to the field of environmental monitoring, and the method comprises the steps: employing a continuous monitoring mode, taking the first radon concentration monitoring data of a calibrated monitoring system as a data starting point, and obtaining the historical time series data of radon concentration; determining a reference parameter based on the historical time sequence data of the radon concentration, and generating a fluctuation range based on the reference parameter; constructing a radon data abnormal fluctuation monitoring model based on a long short-term memory network; obtaining a current environment parameter, and carrying out normalization processing on the current environment parameter to obtain normalized data; inputting the normalized data into a radon data abnormal fluctuation monitoring model to obtain radon concentration prediction data; and determining an abnormal fluctuation level of the radon concentration based on the prediction data and the fluctuation range of the radon concentration. According to the invention, the real-time performance of monitoring the abnormal fluctuation of the radon concentration can be improved while accurate monitoring of the abnormal fluctuation of the radon concentration is realized.

Description

Technical Field

[0001] The present application relates to the field of environmental monitoring, and in particular to a method, equipment, medium and product for monitoring abnormal fluctuations in radon data. Background Art

[0002] Radon and its decay products are radioactive and are listed as one of the main indoor carcinogens. Underground spaces in granite surrounding rocks continuously produce natural radioactive inert gas radon. Continuous monitoring and early warning of radon are important means to avoid radiation impact on people entering. Measurement errors may occur due to aging of radon monitoring system sensors, interference from environmental factors, or improper equipment maintenance. The resulting abnormal fluctuations in radon concentration will reduce the reliability of the radon continuous monitoring system.

[0003] In order to solve the above problems, the reliability detection of existing radon monitoring systems mainly adopts the regular calibration method, which has low real-time and automation and is difficult to meet the long-term reliability monitoring needs of radon. Obviously, the reliability detection method of existing equipment lacks the ability to detect data anomalies, and cannot effectively guarantee the reliability of the radon monitoring system, and thus cannot achieve accurate monitoring of abnormal fluctuations in radon concentration. Summary of the invention

[0004] The purpose of this application is to provide a method, device, medium and product for monitoring abnormal fluctuations of radon data, which can improve the real-time performance of abnormal fluctuation monitoring of radon concentration while realizing accurate monitoring of abnormal fluctuations of radon concentration.

[0005] To achieve the above objectives, this application provides the following solutions:

[0006] In a first aspect, the present application provides a method for monitoring abnormal fluctuations in radon data, comprising:

[0007] Using continuous monitoring, the first radon concentration monitoring data of the calibrated monitoring system is taken as the data starting point to obtain the historical time series data of radon concentration;

[0008] Determine a benchmark parameter based on the historical time series data of radon concentration, and generate a fluctuation range based on the benchmark parameter; the benchmark parameter includes an average value of radon concentration and a standard deviation of radon concentration; the fluctuation range includes: a first fluctuation range, a second fluctuation range, a third fluctuation range, a fourth fluctuation range and a fifth fluctuation range;

[0009] Constructing a radon data abnormal fluctuation monitoring model based on long short-term memory network;

[0010] Acquire current environmental parameters, and perform normalization processing on the current environmental parameters to obtain normalized data; the environmental parameters include: one or more of temperature, humidity and wind speed;

[0011] Inputting the normalized data into the radon data abnormal fluctuation monitoring model to obtain the predicted data of radon concentration;

[0012] An abnormal fluctuation level of the radon concentration is determined based on the predicted data of the radon concentration and the fluctuation range.

[0013] Optionally, a radon data abnormal fluctuation monitoring model is constructed based on a long short-term memory network, including:

[0014] Extracting a plurality of radon concentration data from the historical time series data of radon concentration, and removing the radon concentration data belonging to the fourth fluctuation range or the fifth fluctuation range from the plurality of radon concentration data, to obtain sample data;

[0015] Acquiring environmental parameters at a time corresponding to the sample data, and normalizing the environmental parameters corresponding to the sample data to obtain normalized data;

[0016] generating a sample data pair based on the sample data and the normalized data;

[0017] Build an initial model based on long short-term memory network;

[0018] The normalized data in the sample data pair is used as input, and the sample data is used as output to train the initial model until the trained initial model meets the set conditions, and the trained initial model is used as the radon data abnormal fluctuation monitoring model.

[0019] Optionally, the prediction formula of the radon data abnormal fluctuation monitoring model is generated based on the current hidden state and cell state of the long short-term memory network; the prediction formula is expressed as:

[0020]

[0021] In the formula, represents the predicted data of radon concentration at time t+1, x t-n+1 , x t-n+2 , …, x t represents normalized data, and LSTM represents long short-term memory network processing.

[0022] Optionally, the cell state includes a candidate memory cell state and a current memory cell state;

[0023] The candidate memory cell state is expressed as:

[0024]

[0025] The current memory cell state is expressed as:

[0026]

[0027] In the formula, represents the candidate memory cell state at time t, tanh() represents the hyperbolic tangent function, and w c Represents the weight matrix of cell state update, h t-1 represents the hidden state at time t-1, x t represents the normalized environmental parameters at time t, b C represents the bias term for updating the cell state, C t represents the state of the memory cell at time t, f t represents the activation value of the forget gate in the long short-term memory network, C t-1 represents the state of the memory cell at time t-1, i t Indicates the input status.

[0028] Optionally, the first fluctuation range is [C n -σ n ,C n +σ n ], the second fluctuation range is [C n -2σ n ,C n +2σ n ], the third fluctuation range is [C n -3σ n ,C n +3σ n ], the fourth fluctuation range is (C n +3σ n , +∞), the fifth fluctuation range is (-∞, C n -3σ n ), where C n represents the average radon concentration, σ n Represents the standard deviation of radon concentration.

[0029] Optionally, determining the abnormal fluctuation level of radon concentration based on the predicted data of radon concentration and the fluctuation range specifically includes:

[0030] Performing a denormalization process on the predicted data of the radon concentration to obtain final radon concentration data;

[0031] When the final radon concentration data belongs to the first fluctuation range, determining the abnormal fluctuation level of the radon concentration as normal fluctuation;

[0032] When the final radon concentration data belongs to the second fluctuation range, determining the abnormal fluctuation level of the radon concentration as a common abnormal fluctuation;

[0033] When the final radon concentration data belongs to the third fluctuation range, determining that the abnormal fluctuation level of the radon concentration is a moderate abnormal fluctuation;

[0034] When the final radon concentration data belongs to the fourth fluctuation range or the fifth fluctuation range, it is determined that the abnormal fluctuation level of the radon concentration is a serious abnormal fluctuation.

[0035] Optionally, the radon data abnormal fluctuation monitoring method further includes:

[0036] Corresponding response measures are generated based on the abnormal fluctuation level of the radon concentration.

[0037] In a second aspect, the present application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-mentioned method for monitoring abnormal fluctuations in radon data.

[0038] In a third aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-mentioned method for monitoring abnormal fluctuations in radon data.

[0039] In a fourth aspect, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of the above-mentioned radon data abnormal fluctuation monitoring method.

[0040] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0041] The present application provides a method, device, medium and product for monitoring abnormal fluctuations of radon data. By adopting a continuous monitoring method, the first radon concentration monitoring data of the calibrated monitoring system is used as the data starting point to obtain the historical time series data of the radon concentration, which can ensure the continuity of the measurement point data and provide complete time series information for abnormal fluctuation monitoring. The prediction of radon concentration is achieved by constructing a radon data abnormal fluctuation monitoring model based on a long short-term memory network, which can conduct in-depth analysis and prediction of environmental parameters and radon concentration data, determine the abnormal fluctuation level of radon concentration based on the predicted data and fluctuation range of the radon concentration, and identify and classify the abnormal fluctuation of radon concentration in advance, thereby improving the accuracy and real-time performance of abnormal monitoring of radon concentration. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0043] Figure 1This is an application environment diagram of a method for monitoring abnormal fluctuations of radon data in one embodiment of the present application;

[0044] Figure 2 A flow chart of a method for monitoring abnormal fluctuations of radon data provided in one embodiment of the present application;

[0045] Figure 3 A schematic diagram of an initial model provided for an embodiment of the present application;

[0046] Figure 4 A schematic diagram of a forget gate provided in an embodiment of the present application;

[0047] Figure 5 A schematic diagram of an input gate and candidate memory cells provided in an embodiment of the present application;

[0048] Figure 6 A schematic diagram of a current memory cell provided in one embodiment of the present application;

[0049] Figure 7 A schematic diagram of an output gate and a hidden state provided in an embodiment of the present application;

[0050] Figure 8 A schematic diagram of the overall implementation process of a method for monitoring abnormal fluctuations of radon data provided in one embodiment of the present application;

[0051] Fig. 9 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0052] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0053] In order to make the purpose, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0054] The radon data abnormal fluctuation monitoring method provided in the embodiment of the present application can be applied to Figure 1In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be set up separately, integrated on the server 104, or placed on the cloud or other servers. The terminal 102 can send the historical time series data of radon concentration and the current environmental parameters to the server 104. After the server 104 receives the historical time series data of radon concentration and the current environmental parameters, the server 104 determines the benchmark parameters based on the historical time series data of radon concentration, and generates a fluctuation range based on the benchmark parameters. The current environmental parameters are normalized to obtain normalized data. The normalized data is input into the radon data abnormal fluctuation monitoring model to obtain the predicted data of radon concentration. The abnormal fluctuation level of radon concentration is determined based on the predicted data and fluctuation range of radon concentration. Among them, the radon data abnormal fluctuation monitoring model is implanted in the server 104, which is a neural network model constructed based on the long short-term memory network. The server 104 can feedback the obtained abnormal fluctuation level of radon concentration to the terminal 102. In addition, in some embodiments, the method for monitoring abnormal fluctuations in radon data can also be implemented independently by the server 104 or the terminal 102. For example, the terminal 102 can directly perform abnormal fluctuation monitoring on the historical time series data of radon concentration and the current environmental parameters, or the server 104 can obtain the historical time series data of radon concentration and the current environmental parameters from the data storage system, and perform abnormal fluctuation monitoring on the historical time series data of radon concentration and the current environmental parameters.

[0055] The terminal 102 may be, but is not limited to, various desktop computers, laptop computers, smart phones, tablet computers, IoT devices, and portable wearable devices. The IoT devices may be smart speakers, smart TVs, smart air conditioners, smart vehicle-mounted devices, etc. The portable wearable devices may be smart watches, smart bracelets, head-mounted devices, etc. The server 104 may be implemented as an independent server or a server cluster consisting of multiple servers, or may be a cloud server.

[0056] In an exemplary embodiment, Figure 2 As shown, a method for monitoring abnormal fluctuations in radon data is provided. The method is executed by a computer device, and can be executed by a computer device such as a terminal or a server alone, or by a terminal and a server together. In the embodiment of the present application, the method is applied to Figure 1 The server 104 in the example is used as an example to illustrate, including the following steps 200 to 205. Among them:

[0057] Step 200: Using continuous monitoring, take the first radon concentration monitoring data of the calibrated monitoring system as the data starting point to obtain the historical time series data of radon concentration. The purpose of implementing this step is to ensure the continuity of the measurement point data and provide complete time series information for abnormal analysis. For example, the historical time series data of radon concentration is recorded as {C i}. i Represents the i-th radon concentration value.

[0058] Step 201: Determine a benchmark parameter based on the historical time series data of radon concentration, and generate a fluctuation range based on the benchmark parameter. The benchmark parameter includes the average value of radon concentration and the standard deviation of radon concentration. The fluctuation range includes: a first fluctuation range, a second fluctuation range, a third fluctuation range, a fourth fluctuation range, and a fifth fluctuation range.

[0059] Step 202: Construct a radon data abnormal fluctuation monitoring model based on the long short-term memory network.

[0060] Step 203: Acquire current environmental parameters and perform normalization processing on the current environmental parameters to obtain normalized data. Environmental parameters include: one or more of temperature, humidity and wind speed.

[0061] Among them, temperature, humidity and wind speed are three key environmental parameters that affect radon concentration. These parameters can be measured efficiently and accurately through on-site monitoring equipment. The implementation cost is low and the operability is strong. It can greatly improve the accuracy and reliability of radon anomaly monitoring.

[0062] Step 204: input the normalized data into the radon data abnormal fluctuation monitoring model to obtain the predicted data of radon concentration.

[0063] Step 205: Determine the abnormal fluctuation level of radon concentration based on the predicted data and fluctuation range of radon concentration.

[0064] Implementing the above steps 201 to 205 can improve the accuracy and real-time performance of abnormal monitoring of radon concentration. In addition, the present application constructs a radon data abnormal fluctuation monitoring model based on a long short-term memory network (LSTM). Compared with the traditional BP neural network algorithm, the present application exhibits stronger long-term dependency learning ability and robustness when processing time series data.

[0065] In another exemplary embodiment of the present application, in order to improve the accuracy and reliability of monitoring data under the normal operation of the radon monitoring system and normal fluctuations in the measuring point environment, and to provide judgment parameters for the construction of a monitoring model for abnormal fluctuations in radon data, this embodiment adopts a statistical analysis method to determine the baseline parameters for abnormal fluctuations in the radon monitoring system (such as a radon detector). For example, the first n radon concentration data in the historical time series data are taken for statistical analysis, and their average value and standard deviation are calculated.

[0066] Among them, the average value of the first n radon concentration data is expressed as:

[0067] The standard deviation of the first n radon concentration data is expressed as:

[0068] At this time, the calculated average radon concentration C n and the standard deviation of radon concentration σ n as a benchmark parameter.

[0069] Based on this, the first fluctuation range is [C n -σ n ,C n +σ n ], the second fluctuation range is [C n -2σ n ,C n +2σ n ], the third fluctuation range is [C n -3σ n ,C n +3σ n ], the fourth fluctuation range is (C n +3σ n , +∞), the fifth fluctuation range is (-∞, C n -3σ n ).

[0070] In another exemplary embodiment of the present application, compared with the BP neural network model in the traditional radon concentration detection, when processing continuous monitoring data, in order to more accurately predict and identify abnormal fluctuations in radon concentration, thereby significantly improving the real-time performance of monitoring, the implementation process of the above step 202 can be described as follows:

[0071] Step 1: extract multiple radon concentration data from the historical time series data of radon concentration, and remove the radon concentration data belonging to the fourth fluctuation range or the fifth fluctuation range from the multiple radon concentration data to obtain sample data.

[0072] For example, the improved three-times standard deviation method is used to pre-judge abnormal radon fluctuations. The radon concentration value to be judged is C p , using C n and σ nPreliminary judgment C p Is it an abnormal fluctuation data point? If it satisfies: C p >C n +3σ n or C p <C n -3σ n , then it is judged as C p It is an abnormal fluctuation data point.

[0073] This step can provide a data set after removing outliers for the model training phase, ensuring the reliability of subsequent model training data, thereby enhancing the accuracy of subsequent LSTM algorithm model predictions.

[0074] Step 2: Acquire environmental parameters corresponding to the sample data, and perform normalization processing on the environmental parameters corresponding to the sample data to obtain normalized data.

[0075] For example, take the first m (m>3n) radon concentration data, use the outlier elimination method in step (1) to eliminate the abnormal fluctuation values ​​in the extracted multiple radon concentration data, and obtain a clean radon concentration data set sorted by time after screening, denoted as {C′ i For the clean radon concentration dataset {C′ i}, collect the temperature T at the corresponding time i 、Humidity i , wind speed W i And other environmental parameters, and construct feature data set X i =[T i , H i , W i ] The feature dataset X i =[T i , H i , W i ] to normalize the data to meet the input requirements of the subsequent analysis model and ensure the accuracy and efficiency of the model analysis. The normalization formula is:

[0076]

[0077] Where, X i ′ represents normalized data, X max Represents the maximum value of the feature data set, X min The minimum value of the feature dataset.

[0078] Step 3: Generate sample data pairs based on the sample data and normalized data.

[0079] Step 4: Build an initial model based on the long short-term memory network.

[0080] Step 5: Use the normalized data in the sample data pair as input and the sample data as output to train the initial model until the trained initial model meets the set conditions, and then use the trained initial model as the radon data abnormal fluctuation monitoring model.

[0081] In the actual application process, before the sample data is input into the initial model, it is necessary to convert the format to fit the input initial model. Based on this, an input sequence (i.e. normalized data) containing the past t steps and the corresponding output sequence (i.e. sample data) are created. The input features are temperature, humidity, wind speed, and the time step is p.

[0082] In another exemplary embodiment of the present application, the implementation process of constructing the initial model based on the long short-term memory network in the above step 4 can be described as:

[0083] 1) Construct input and output sequences, where the input sequence is expressed as:

[0084]

[0085] The output sequence is represented as: Output t =y t .

[0086] In the formula, Input t represents the input data at time t, X′ t represents the normalized data at time t, T* represents the temperature at time *, H* represents the humidity at time *, and W* represents the wind speed at time *.

[0087] 2) Construct a neural network containing several LSTM layers (i.e., LSTM units) and fully connected layers. The initial model constructed is as follows Figure 3 As shown, Figure 3 A represents the LSTM unit.

[0088] (1) The forget gate is represented by: f t =s(w f ·[h t-1 , x t ]+b f ). The forget gate is Figure 4 shown.

[0089] Among them, s is the sigmoid function, w f and b f are weights and biases, h t-1 is the hidden state at time t-1, x t Represents the environmental parameters at time t.

[0090] The forget gate determines which information is discarded from the current cell state and is used to process continuous radon concentration data while ignoring noise or irrelevant changes in a short period of time.

[0091] (2) The input gate is represented as: i t =s(w i ·[h t-1 , x t ]+b i ). The input gate and candidate memory cells are as follows Figure 5 shown.

[0092] (3) The state of candidate memory cells is expressed as:

[0093] In the formula, represents the candidate memory cell state at time t, which is obtained by transforming the input data of the current time step through the tanh function, and represents the new information written into the cell state. i Represents the weight matrix of the input gate, reflecting the degree of opening and closing of the input gate adjusted by radon concentration data and other environmental parameters. t Represents the bias term of the input gate. tanh() is a hyperbolic tangent function that converts the input value to the interval (-1,1). c The weight matrix used as the cell state update determines the current input and the cell state of the previous time step to update the current cell state. C Represents the bias term for cell state update.

[0094] (4) The current memory cell state is expressed as: Current memory cells Figure 6 shown.

[0095] In the formula, C t represents the state of the memory cell at time t. t Represents the activation value of the forget gate in the long short-term memory network, which determines the cell state information before the current time step is forgotten. C t-1 Represents the state of the memory cell at time t-1, including the memory information of the previous time step. t Indicates the input status.

[0096] (5) The output gate is represented by: o t =s(w o ·[h t-1 , x t ]+b o ).

[0097] The current hidden state is represented as: h t =o t tanh(C t ). The output gate and the current hidden state are as follows Figure 7shown.

[0098] In the formula, o t Represents the activation value of the output gate, which determines how much cell state information is output at the current time step. o As the weight matrix of the output gate, it indicates the degree of opening and closing of the output gate adjusted by radon concentration data and other environmental parameters. o As the bias term of the output gate. t Represents the hidden state at time t, represents the output of LSTM at the current time step, and is used to predict the radon concentration at the next time step.

[0099] (6) The loss function is expressed as: N represents the total number of radon concentration data.

[0100] Based on the initial model constructed above, by adjusting its parameters, the radon concentration can be better predicted, thereby improving the accuracy and reliability of the monitoring system.

[0101] In another exemplary embodiment of the present application, in the above step 5, the normalized time series data is input, and the LSTM will learn the patterns in the data and predict the future radon concentration value. Among them, the prediction formula of the radon data abnormal fluctuation monitoring model is generated based on the current hidden state and cell state of the long short-term memory network. The prediction formula is expressed as:

[0102]

[0103] In the formula, represents the predicted data of radon concentration at time t+1, x t-n+1 , x t-n+2 , …, x t represents normalized data, and LSTM represents long short-term memory network processing.

[0104] Based on the above description, LSTM provides an accurate prediction basis for real-time monitoring of radon concentration and abnormal self-detection and identification, ensuring the accuracy of subsequent abnormal level classification.

[0105] In another exemplary embodiment of the present application, after the predicted data of radon concentration is obtained, in order to facilitate comparison with the actual monitoring value and more accurately determine the abnormal fluctuation level, in this embodiment, the implementation process of the above step 205 can be:

[0106] Step 1: Denormalize the predicted data of radon concentration to obtain the final radon concentration data. The denormalization formula is:

[0107]

[0108] In the formula, represents the predicted data of radon concentration, and y represents the final radon concentration data.

[0109] Step 2: When the final radon concentration data belongs to the first fluctuation range (i.e. C n -σ n ≤y i ≤C n +σ n ), the abnormal fluctuation level of radon concentration is determined to be normal fluctuation.

[0110] Step 3: When the final radon concentration data belongs to the second fluctuation range (i.e. C n -2σ n ≤y i ≤C n +2σ n ), the abnormal fluctuation level of radon concentration is determined to be ordinary abnormal fluctuation.

[0111] Step 4: When the final radon concentration data belongs to the third fluctuation range (i.e. C n -3σ n ≤y i ≤C n +3σ n ), the abnormal fluctuation level of radon concentration is determined to be moderate abnormal fluctuation.

[0112] Step 5: When the final radon concentration data belongs to the fourth fluctuation range or the fifth fluctuation range (i.e. i >C n +3σ n or i <C n -3σ n ), the abnormal fluctuation level of radon concentration is determined to be severe abnormal fluctuation. i Represents the i-th final radon concentration data.

[0113] In another exemplary embodiment of the present application, corresponding response measures can also be generated based on the abnormal fluctuation level of radon concentration. For example, when it is considered as a serious abnormal fluctuation level, the corresponding response measures are automatically triggered and relevant personnel are notified to ensure that the response strategy is taken in time.

[0114] In summary, the overall implementation process of the radon data abnormal fluctuation monitoring method provided by this application is as follows: Figure 8As shown. Compared with traditional technologies (such as the method of using BP neural network to monitor radon concentration), this application realizes real-time continuous monitoring of radon concentration and automatic anomaly detection. By adopting long short-term memory network (LSTM), this application can conduct in-depth analysis and prediction of environmental parameters and radon concentration data, so as to identify and classify abnormal fluctuations in advance, so that this application has significant beneficial effects in improving monitoring efficiency, reducing labor costs, enhancing data reliability and optimizing environmental health management (that is, it can classify the abnormal fluctuation level of radon concentration results obtained by real-time monitoring, and automatically trigger corresponding response measures according to the level, thereby reducing labor costs while enhancing the reliability of monitoring data and optimizing environmental health management), providing a more scientific and efficient solution for the field of environmental monitoring.

[0115] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Fig. 9 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store radon abnormal fluctuation monitoring data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for monitoring abnormal fluctuations in radon data is implemented.

[0116] Those skilled in the art will understand that Fig. 9 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components. In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0117] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0118] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0119] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0120] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0121] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0122] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0123] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A method for monitoring abnormal fluctuations in radon data, characterized in that: The radon data abnormal fluctuation monitoring method comprises: Using continuous monitoring, the first radon concentration monitoring data of the calibrated monitoring system is taken as the data starting point to obtain the historical time series data of radon concentration; Determine a benchmark parameter based on the historical time series data of radon concentration, and generate a fluctuation range based on the benchmark parameter; the benchmark parameter includes an average value of radon concentration and a standard deviation of radon concentration; the fluctuation range includes: a first fluctuation range, a second fluctuation range, a third fluctuation range, a fourth fluctuation range and a fifth fluctuation range; Constructing a radon data abnormal fluctuation monitoring model based on long short-term memory network; Acquire current environmental parameters, and perform normalization processing on the current environmental parameters to obtain normalized data; the environmental parameters include: one or more of temperature, humidity and wind speed; Inputting the normalized data into the radon data abnormal fluctuation monitoring model to obtain the predicted data of radon concentration; An abnormal fluctuation level of the radon concentration is determined based on the predicted data of the radon concentration and the fluctuation range.

2. The method for monitoring abnormal fluctuations of radon data according to claim 1, characterized in that: A radon data abnormal fluctuation monitoring model is constructed based on long short-term memory network, including: Extracting a plurality of radon concentration data from the historical time series data of radon concentration, and removing the radon concentration data belonging to the fourth fluctuation range or the fifth fluctuation range from the plurality of radon concentration data, to obtain sample data; Acquiring environmental parameters at a time corresponding to the sample data, and normalizing the environmental parameters corresponding to the sample data to obtain normalized data; generating a sample data pair based on the sample data and the normalized data; Build an initial model based on long short-term memory network; The normalized data in the sample data pair is used as input, and the sample data is used as output to train the initial model until the trained initial model meets the set conditions, and the trained initial model is used as the radon data abnormal fluctuation monitoring model.

3. The method for monitoring abnormal fluctuation of radon data according to claim 2, characterized in that: The prediction formula of the radon data abnormal fluctuation monitoring model is generated based on the current hidden state and cell state of the long short-term memory network; the prediction formula is expressed as: In the formula, represents the predicted data of radon concentration at time t+1, x t-n+1 , x t-n+2 , …, x t represents normalized data, and LSTM represents long short-term memory network processing.

4. The method for monitoring abnormal fluctuation of radon data according to claim 3, characterized in that: The cell state includes the candidate memory cell state and the current memory cell state; The candidate memory cell state is expressed as: The current memory cell state is expressed as: In the formula, represents the candidate memory cell state at time t, tanh() represents the hyperbolic tangent function, and w c Represents the weight matrix of cell state update, h t-1 represents the hidden state at time t-1, x t represents the normalized environmental parameters at time t, b C represents the bias term for updating the cell state, C t represents the state of the memory cell at time t, f t represents the activation value of the forget gate in the long short-term memory network, C t-1 represents the state of the memory cell at time t-1, i t Indicates the input status.

5. The method for monitoring abnormal fluctuation of radon data according to claim 1, characterized in that: The first fluctuation range is [C n -σ n ,C n +σ n ], the second fluctuation range is [C n -2σ n ,C n +2σ n ], the third fluctuation range is [C n -3σ n ,C n +3σ n ], the fourth fluctuation range is (C n +3σ n , +∞), the fifth fluctuation range is (-∞, C n -3σ n ), where C n represents the average radon concentration, σ n Represents the standard deviation of radon concentration.

6. The method for monitoring abnormal fluctuation of radon data according to claim 1, characterized in that: Determining the abnormal fluctuation level of radon concentration based on the predicted data of radon concentration and the fluctuation range specifically includes: Performing a denormalization process on the predicted data of the radon concentration to obtain final radon concentration data; When the final radon concentration data belongs to the first fluctuation range, determining the abnormal fluctuation level of the radon concentration as normal fluctuation; When the final radon concentration data belongs to the second fluctuation range, determining the abnormal fluctuation level of the radon concentration as a common abnormal fluctuation; When the final radon concentration data belongs to the third fluctuation range, determining that the abnormal fluctuation level of the radon concentration is a moderate abnormal fluctuation; When the final radon concentration data belongs to the fourth fluctuation range or the fifth fluctuation range, it is determined that the abnormal fluctuation level of the radon concentration is a serious abnormal fluctuation.

7. The method for monitoring abnormal fluctuation of radon data according to claim 6, characterized in that: The radon data abnormal fluctuation monitoring method also includes: Corresponding response measures are generated based on the abnormal fluctuation level of the radon concentration.

8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method for monitoring abnormal fluctuations of radon data according to any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for monitoring abnormal fluctuation of radon data described in any one of claims 1 to 7 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for monitoring abnormal fluctuations of radon data described in any one of claims 1 to 7 is implemented.

Images