A meteorological data processing method and device and related equipment thereof

Abstract

The application provides a meteorological data processing method and device and related equipment thereof. The method comprises the following steps: monitoring the working state of a first sensor used for collecting a first meteorological parameter; when it is determined that the first sensor is faulty according to the working state, starting a second sensor corresponding to the same monitoring point as the first sensor to replace the first sensor to collect the first meteorological parameter; in the process of replacing the first sensor by the second sensor to collect data, obtaining a first historical data sequence of the first sensor before the fault, and obtaining first real-time data collected by the second sensor in real time; and performing fusion processing on the first historical data sequence and the first real-time data by using a preset smooth transition algorithm to generate smooth transition period meteorological data.

Description

Technical Field

[0001] This application relates to the fields of Internet of Things and artificial intelligence technologies, and more specifically, to a meteorological data processing method, apparatus and related equipment. Background Technology

[0002] Currently, meteorological monitoring systems are widely used in environmental monitoring, smart agriculture, and smart cities. Existing technologies include patent application CN115964675A, which discloses a power meteorological data fusion method based on multi-source weather forecasts, focusing on fusing data from multiple numerical weather forecast sources to improve forecast accuracy. Another example is patent application CN206684329U, which discloses a multifunctional meteorological monitoring system that integrates various sensors and transmits data wirelessly, enabling the collection of basic meteorological elements.

[0003] However, the existing technical solutions still have the following shortcomings: First, the monitoring systems mostly rely on a single sensor. Once the sensor fails, data monitoring will be directly interrupted, posing a single point of failure risk. Second, although multi-source data fusion technology exists, it is mostly for forecast data. For real-time sensor data, especially during the switching between primary and backup sensors, there is a lack of an effective data smoothing mechanism, which can easily cause data jumps and affect data quality. Therefore, how to build a meteorological monitoring system with high reliability, high accuracy, and strong data fault tolerance remains an urgent technical problem to be solved. Summary of the Invention

[0004] This application provides a meteorological data processing method, apparatus, and related equipment for constructing a meteorological monitoring system with high reliability, high accuracy, and strong data fault tolerance.

[0005] Firstly, a meteorological data processing method is provided, including: Monitor the working status of the first sensor, which is used to collect the first meteorological parameter; When the first sensor is determined to be faulty based on the working status, the second sensor at the same monitoring point as the first sensor is activated to take over the collection of the first meteorological parameters from the first sensor. During the process of the second sensor taking over data collection from the first sensor, the first historical data sequence of the first sensor before the failure is obtained, and the first real-time data collected by the second sensor in real time is obtained. The first historical data sequence and the first real-time data are fused using a preset smooth transition algorithm to generate smooth transition meteorological data.

[0006] Secondly, a meteorological data processing device is provided, comprising: The status monitoring module is used to monitor the working status of the first sensor, which is used to collect the first meteorological parameters. The sensor activation module is used to activate a second sensor at the same monitoring point as the first sensor when the first sensor is determined to be faulty based on the working status, so as to take over the collection of the first meteorological parameters from the first sensor. The data acquisition module is used to acquire the first historical data sequence of the first sensor before the failure during the process of the second sensor taking over the data acquisition of the first sensor, and to acquire the first real-time data collected by the second sensor in real time. The data processing module is used to fuse the first historical data sequence and the first real-time data using a preset smooth transition algorithm to generate smooth transition meteorological data.

[0007] Thirdly, an electronic device is provided, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, performs the steps of the method described in the first aspect.

[0008] Fourthly, a computer-readable storage medium is provided having a computer program or instructions stored thereon, which, when executed by a processor, implement the steps of the method described in the first aspect.

[0009] Fifthly, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of the method described in the first aspect.

[0010] The meteorological data processing method provided in this application solves the problem of data interruption caused by a single sensor failure by monitoring the working status of the first sensor and activating the second sensor at the same monitoring point to take over the data collection when a fault is detected. On this basis, by acquiring the historical data sequence before the fault and the real-time data of the backup sensor, and using a smooth transition algorithm to fuse the two, the data during the switching between the primary and backup sensors can be smoothly connected, avoiding data jumps caused by individual differences or sudden changes in the state of the sensors. Thus, even when a sensor fails, it can still output continuous and smooth meteorological data to the monitoring platform, ensuring the integrity and reliability of meteorological monitoring data. Attached Figure Description

[0011] Figure 1 This is a schematic diagram illustrating the implementation flow of a meteorological data processing method provided in an exemplary embodiment of this application; Figure 2 This is a schematic diagram of the system architecture of the meteorological monitoring system provided in the embodiments of this application; Figure 3 This is a schematic diagram of the sensor redundancy switching process provided in an embodiment of this application; Figure 4 This is a schematic diagram of the data fusion and optimization process provided in the embodiments of this application; Figure 5 This is a schematic diagram of the structure of a meteorological data processing device provided in an exemplary embodiment of this application; Figure 6 This is a schematic diagram of the structure of an electronic device provided in an exemplary embodiment of this application. Detailed Implementation

[0012] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0013] The following is a description of the terms used in this application: Multi-source sensors refer to various types of environmental sensors deployed in meteorological monitoring systems, including temperature and humidity sensors, wind speed and direction sensors, atmospheric pressure sensors, and rainfall sensors. These sensors are used to collect data on the same or different environmental parameters from multiple sources, providing the system with a rich source of basic data.

[0014] Redundant sensors refer to backup sensors configured for the same meteorological parameters at the same monitoring point. When the main sensor fails, the system automatically activates the backup sensor to take over the data acquisition work, thereby eliminating the risk of single-point failure and ensuring continuous acquisition of meteorological data.

[0015] The sliding window algorithm is a method for real-time data stream analysis. It maintains a fixed-size data window over a time series and calculates the mean of the data within the window to dynamically monitor data trends and determine whether sensor data shows continuous anomalies.

[0016] Kalman filtering, an optimal estimation recursive algorithm, uses system state equations and observation data to predict historical data sequences. In sensor switching scenarios, it is used to predict data change trends after the failure of the main sensor, achieving a smooth transition between old and new data sources.

[0017] Exponentially Weighted Moving Average (EWMA) is a time series smoothing algorithm that uses weighted averages to assign different weights to historical and current data, with recent data having higher weights and older data having lower weights. In sensor switching, it is used to fuse Kalman filter prediction data with real-time data from backup sensors to avoid data abrupt changes.

[0018] Bayesian weighted fusion is a multi-source data fusion algorithm based on probability statistics. It dynamically adjusts the weight of each sensor in the fusion calculation according to the historical accuracy and reliability of each sensor. The higher the accuracy of the sensor, the greater the weight, thereby calculating the optimal fusion value and improving the accuracy of meteorological data.

[0019] Z-score standardization is an outlier detection method that calculates the ratio of the deviation of a data point from the mean to the standard deviation to obtain a standardized score. When the standardized score exceeds a preset threshold, the data point is determined to be an outlier. It is used to identify and remove outliers in meteorological data.

[0020] A circular buffer is a fixed-size memory cache structure that uses a first-in-first-out circular storage method. When the buffer is full, new data automatically overwrites the oldest data. It is used on the data acquisition device to store all meteorological data collected in the past 72 hours to prevent data loss in the event of network failure.

[0021] The dynamic bandwidth adaptive algorithm is an algorithm that dynamically adjusts data transmission parameters based on the current network conditions. When network bandwidth is low, it reduces the data upload frequency or the data packet size, and when network bandwidth is sufficient, it restores the normal upload strategy to ensure the stability and reliability of data transmission.

[0022] LSTM prediction models, or Long Short-Term Memory networks, are a special type of recurrent neural network structure. By introducing a gating mechanism, they address the long-term dependency problem of traditional RNNs. In meteorological monitoring systems, they are used to predict and fill in missing data over long periods, generating complete data for missing time periods by utilizing historical data patterns.

[0023] As described in the background section, most existing meteorological monitoring systems rely on a single sensor for data acquisition. When a sensor malfunctions due to harsh environments or equipment aging, data monitoring may be interrupted or lost, resulting in a high risk of single-point failure. Furthermore, even when some systems are equipped with backup sensors, there is a lack of effective data smoothing mechanisms. Switching between primary and backup sensors can easily cause data jumps due to individual differences, affecting the continuity and reliability of the data. This application addresses these problems by proposing a meteorological data processing method. This method monitors the operating status of a first sensor and, upon determining its failure, activates a second sensor at the same monitoring point to take over data acquisition. A smoothing transition algorithm is used to fuse the historical data sequence before the failure with the real-time data from the backup sensor, generating smooth transition meteorological data. This ensures that even when a sensor fails, continuous and smooth meteorological data can still be output, guaranteeing the integrity and reliability of meteorological monitoring data.

[0024] The technical solutions provided by the various embodiments of this application are described in detail below with reference to the accompanying drawings.

[0025] Figure 1 This is a flowchart illustrating a meteorological data processing method provided as an exemplary embodiment of this application. Figure 1 As shown, the method includes: Step 110: Monitor the working status of the first sensor, which is used to collect the first meteorological parameters.

[0026] The meteorological data processing method provided in this application embodiment can be applied to... Figure 2 The image shows a meteorological monitoring system based on multi-source sensor access. Figure 2The system architecture of the meteorological monitoring system provided in this application embodiment is illustrated. The system adopts an architecture combining distributed data acquisition and centralized intelligent processing, mainly comprising the following layers: First layer, a multi-type sensor layer, which is the front-end data acquisition layer, configured with various meteorological sensors, including at least one of temperature and humidity sensors, wind speed and direction sensors, atmospheric pressure sensors, and rainfall sensors. Each sensor group corresponds to a meteorological parameter, and each sensor group includes at least one main sensor and at least one backup sensor for redundant acquisition. Second layer, a data transmission layer, which is responsible for transmitting the data collected by the front-end sensors to the intelligent processing center via wired or wireless networks. Specifically, it supports wired transmission methods such as RS485 and wired MODBUS, as well as wireless transmission methods such as GPRS, LoRa, and NB-IoT, meeting the remote data acquisition and transmission needs under different deployment scenarios. Third layer, an intelligent processing center, which utilizes built-in AI algorithms and data fusion algorithms to perform real-time verification, anomaly detection, and predictive correction on the received data. It also implements redundancy management, dynamically judging sensor status and automatically switching backup sensors when necessary, executing redundant data processing methods. Fourth layer, a data storage layer, which implements dual storage of data, including local caching and cloud storage. A circular buffer design is adopted at the data acquisition device end to store all data collected within the past 72 hours, preventing data loss in the event of network failure; at the same time, the data is backed up to the cloud server to ensure data security. The fifth layer is the user terminal, which provides a user-visual interface to realize functions such as data display, alarm management, and historical data query, allowing maintenance personnel or end users to view meteorological monitoring data and system operating status in real time.

[0027] The first sensor refers to the sensor used to collect the first meteorological parameter, which can be any one of a temperature and humidity sensor, a wind speed and direction sensor, an atmospheric pressure sensor, or a rainfall sensor. The operating status refers to the sensor's healthy operational status, including whether the sensor's voltage is stable, whether the response time is normal, and whether the collected data is continuous and stable.

[0028] As an example, taking a temperature sensor as an example, the voltage value and current temperature data of the sensor are collected every 10 seconds. The hardware status is judged by monitoring whether the sensor voltage is stable within the range of 3.3V-3.6V and whether the response time is within 100ms. At the same time, a sliding window containing the most recent 10 temperature data is maintained, and the average temperature within the window is calculated. If the average value of the window exceeds the preset normal temperature change threshold for 3 consecutive times, the temperature sensor is judged to be faulty.

[0029] In some exemplary embodiments, monitoring the operating status of the first sensor includes: The operating status parameters of the first sensor are collected, including at least one of voltage and response time; And / or, The sliding window mean of the real-time data collected by the first sensor is calculated to obtain the data change trend; When the operating status parameters are abnormal, or the data change trend exceeds the preset abnormal threshold multiple times in a row, the first sensor is determined to be faulty.

[0030] Specifically, sliding window mean calculation involves maintaining a fixed-size data window, calculating the mean of the collected data within the window, and determining whether the sensor data is abnormal by analyzing the trend of the mean changes across multiple consecutive windows. For example, if the mean of five consecutive sliding windows exceeds a preset normal range threshold, the sensor is considered to be faulty. This dual-judgment mechanism combines hardware status monitoring and data quality analysis, enabling accurate identification of sensor faults and avoiding misjudgments.

[0031] Step 120: When the first sensor is determined to be faulty based on its working status, the second sensor at the same monitoring point as the first sensor is activated to take over the collection of the first meteorological parameters from the first sensor.

[0032] Specifically, when the first sensor is determined to be faulty based on its operating status, a second sensor at the same monitoring point as the first sensor is activated to take over the collection of the first meteorological parameter. The second sensor refers to a backup sensor at the same monitoring point as the first sensor; that is, a redundant sensor deployed in the same physical location for the same meteorological parameter. "Same monitoring point" means that the primary and backup sensors are deployed in the same geographical location to monitor the same environmental parameter. "Take over" means that the backup sensor takes over the data collection task and becomes the currently active data source.

[0033] As an example, at a wind speed monitoring point of a meteorological monitoring station, a primary wind speed sensor and a backup wind speed sensor are deployed simultaneously. When the system determines through step 110 that the primary wind speed sensor has failed, it immediately activates the backup wind speed sensor, which then begins collecting wind speed data, ensuring that wind speed monitoring data is not interrupted due to the failure of the primary sensor.

[0034] Step 130: During the process of the second sensor taking over data collection from the first sensor, the first historical data sequence of the first sensor before the failure is obtained, and the first real-time data collected by the second sensor in real time is obtained.

[0035] Specifically, during the process of the second sensor taking over data collection from the first sensor, the first historical data sequence of the first sensor prior to the failure is acquired, and the first real-time data collected by the second sensor is also acquired. The first historical data sequence refers to the set of historical data points collected by the first sensor within a certain period before the failure, reflecting the trend of meteorological parameter changes before the failure. The first real-time data refers to the latest data points collected by the second sensor after its activation. The takeover process refers to the transition period from the confirmation of the primary sensor failure to the stable operation of the backup sensor.

[0036] As an example, when the main wind speed sensor is determined to be faulty at time t0, the system acquires 60 wind speed data points collected by the sensor within 10 minutes before time t0 to form a historical data sequence; at the same time, the system acquires the first wind speed data point collected by the backup wind speed sensor in real time after time t0 as real-time data.

[0037] Step 140: The first historical data sequence and the first real-time data are fused using a preset smooth transition algorithm to generate smooth transition meteorological data.

[0038] A pre-defined smoothing transition algorithm is used to fuse the first historical data sequence and the first real-time data to generate smooth transitional meteorological data. The pre-defined smoothing transition algorithm is a pre-set data processing algorithm used to fuse the old and new data sources, aiming to avoid data jumps during the switch between primary and backup sensors. Smooth transitional meteorological data refers to the continuous and smooth meteorological data sequence output after algorithm processing during the transition period.

[0039] In some exemplary embodiments, the preset smooth transition algorithm includes the Kalman filter algorithm and the exponentially weighted moving average algorithm; The first historical data sequence and the first real-time data are fused using a preset smooth transition algorithm, including: The Kalman filter algorithm is used to predict the first historical data sequence to obtain the predicted data; An exponentially weighted moving average algorithm is used to fuse the predicted data with the first real-time data to generate transitional meteorological data.

[0040] The Kalman filter algorithm can predict the expected value at the current moment based on historical data sequences, while the exponentially weighted moving average algorithm weights and fuses the predicted value and the real-time measurement value, with recent data having a higher weight, thus generating smooth transition data.

[0041] As an example, suppose the historical wind speed data sequence for the 10 minutes prior to the main anemometer failure is [3.2, 3.3, 3.1, 3.4, 3.2, 3.5, 3.3, 3.6, 3.4, 3.5] m / s. The Kalman filter algorithm predicts the current wind speed to be 3.45 m / s based on this historical data. The first wind speed data collected in real time by the backup anemometer is 3.8 m / s. Using an exponentially weighted moving average algorithm, with weighting coefficients of 0.7 (predicted value weight) and 0.3 (measured value weight), the resulting transitional wind speed data is 3.45 × 0.7 + 3.8 × 0.3 = 3.555 m / s. Subsequent data points are then fused sequentially to generate a complete transitional wind speed data sequence, avoiding a direct jump from 3.5 m / s to 3.8 m / s.

[0042] Historical data is predicted using the Kalman filter algorithm, and then the predicted value is fused with the real-time data from the backup sensor using the exponentially weighted moving average algorithm. This enables a smooth transition of data during the switchover between the primary and backup sensors, avoiding data jumps caused by individual sensor differences or sudden changes in sensor status, and ensuring the smoothness and reliability of meteorological monitoring data.

[0043] Figure 3 The sensor redundancy switching process provided in an embodiment of this application is illustrated. The process includes: Step 301: Sensor Data Acquisition. Meteorological data is acquired in real time through a first sensor. The first sensor is used to collect first meteorological parameters, which include at least one of temperature and humidity, wind speed and direction, atmospheric pressure, and rainfall.

[0044] Step 302: Health Status Monitoring. Monitor the operating status of the first sensor, specifically including collecting the operating status parameters of the first sensor, which include at least one of voltage, response time, and data stability; simultaneously, calculate the sliding window mean of the real-time data collected by the first sensor to obtain the data change trend.

[0045] Step 303: Check for data anomalies. Based on the monitoring results of Step 302, determine whether the first sensor is abnormal. Specifically, if the operating status parameters are abnormal (such as unstable voltage or response timeout), or if the data change trend shows abnormal fluctuations, an anomaly is initially determined; if no anomaly is detected, return to Step 301 to continue normal data acquisition.

[0046] Step 304: Anomaly Accumulation Judgment. When an anomaly is detected in step 303, it is not immediately determined to be a fault, but rather an anomaly accumulation judgment is performed. Specifically, a sliding window algorithm is used to accumulate and count multiple consecutive anomaly events. Only when the data change trend exceeds a preset anomaly threshold multiple times consecutively is the first sensor finally determined to be faulty. Through the anomaly accumulation mechanism, erroneous switching caused by single, occasional anomalies is avoided.

[0047] Step 305: Fault Alarm and Recording. When a fault is detected in the first sensor, the system automatically sends an alarm to the operation and maintenance platform and records the fault tracing information, including the fault occurrence time, faulty sensor identifier, and anomaly type, to facilitate subsequent maintenance and fault analysis.

[0048] Step 306: Activate the backup sensor. Automatically activate the second sensor at the same monitoring point as the first sensor to take over the collection of the first meteorological parameters.

[0049] Step 307: Data Smoothing Processing. During the process of the second sensor taking over data collection from the first sensor, the historical data sequence of the first sensor before the failure is obtained, and the real-time data collected by the second sensor is also obtained. The historical data sequence is processed using a Kalman filter algorithm to obtain predicted data. The predicted data is then fused with the real-time data using an exponentially weighted moving average algorithm to generate smooth transitional meteorological data.

[0050] Step 308: Output transition period meteorological data. Output the generated smooth transition period meteorological data to subsequent processing stages (such as data fusion, storage, or uploading) to complete sensor redundancy switching and ensure the continuity and smoothness of meteorological data.

[0051] In some exemplary embodiments, the method provided in this application further includes a data fusion and optimization step. Specifically, the method provided in this application further includes: Acquire raw data from multiple sensors for the same meteorological parameter; A Bayesian weighted fusion algorithm is adopted to dynamically adjust the corresponding weight of each sensor according to the historical accuracy of each sensor, and to perform fusion calculation on multiple raw data to obtain the first fused data; Perform outlier detection on the first fused data and remove data points that are identified as outliers; For data points that have been removed or are missing, a data completion strategy is used to fill them in, generating optimized meteorological data.

[0052] Raw data collected from multiple sensors for the same meteorological parameter is acquired. A Bayesian weighted fusion algorithm is used to dynamically adjust the weights of each sensor based on its historical accuracy, and then the multiple raw data are fused to obtain the first fused data. The "Bayesian weighted fusion algorithm" is a probabilistic statistical multi-source data fusion method that dynamically assigns weights to each sensor based on its historical accuracy and reliability, with higher-accuracy sensors receiving larger weights, thereby calculating the optimal fused value.

[0053] In some exemplary embodiments, outlier detection is performed on the first fused data, including: The Z-score normalization method was used to calculate the standardized score of each data point in the first fused dataset. Data points whose standardized scores exceed a preset threshold are identified as outliers.

[0054] Specifically, outlier detection is performed on the first fused data, and data points identified as outliers are removed. Specifically, the Z-score standardization method is used to calculate the standardized score for each data point in the first fused data; data points with standardized scores exceeding a preset threshold are identified as outliers. Z-score standardization refers to calculating the deviation of a data point x from the mean μ and standard deviation σ using the formula Z=(x-μ) / σ. When the absolute value of Z exceeds a preset threshold (usually 3), the data point is identified as an outlier.

[0055] In some exemplary embodiments, the data completion strategy includes: For data segments with missing durations less than the preset duration, a linear interpolation algorithm is used to complete them; For data segments with missing durations greater than or equal to the preset duration, a pre-trained Long Short-Term Memory (LSTM) prediction model is used to impute the missing data.

[0056] For data points that have been removed or are missing, a data completion strategy is employed to fill in the gaps and generate optimized meteorological data. Specifically, for data segments with missing durations shorter than a preset duration, a linear interpolation algorithm is used for completion; for data segments with missing durations greater than or equal to the preset duration, a pre-trained Long Short-Term Memory (LSTM) prediction model is used for data completion. The linear interpolation algorithm uses valid data points before and after the missing point for linear estimation; the LSTM prediction model is a special type of recurrent neural network that learns the changing patterns of meteorological parameters by training on historical data, enabling it to predict and fill in data gaps over longer periods.

[0057] As an example, a monitoring point is equipped with three temperature sensors with historical accuracies of 98%, 95%, and 92%, respectively. At time t, the temperature values ​​collected by the three sensors are 23.5℃, 23.8℃, and 24.2℃, respectively. A Bayesian weighted fusion algorithm assigns weights of 0.35, 0.33, and 0.32 based on the historical accuracy, calculating the fused value as 23.5×0.35 + 23.8×0.33 + 24.2×0.32 ≈ 23.8℃. Subsequently, the system uses the Z-score normalization method to examine the fused data sequence. It finds that the 23.8℃ at a certain time differs significantly from the preceding and following data, with a normalization score of 3.2, exceeding the threshold of 3, and is therefore identified as an outlier and removed. For data gaps after removal, if the missing time is less than 1 hour, linear interpolation is used to fill in the gaps; if the missing time reaches 2 hours, a pre-trained LSTM model is invoked, inputting historical temperature data from the previous 24 hours to predict and output the temperature value for the missing period.

[0058] By dynamically adjusting the weights of multi-source data using a Bayesian weighted fusion algorithm, the accuracy of the fused data was improved; outliers were identified and removed using the Z-score standardization method, thus improving data quality; and a hierarchical data completion strategy (linear interpolation for short-term missing data and LSTM model for long-term missing data) was employed to achieve high-precision data completion while ensuring computational efficiency, thereby effectively improving the integrity and reliability of meteorological data.

[0059] Figure 4 The data fusion and optimization process provided in this application embodiment is illustrated. The process includes the following steps: Step 401: Multi-source data acquisition. Acquire raw data from multiple sensors for the same meteorological parameter. These sensors can be deployed at the same monitoring point or different monitoring points to collect meteorological parameters such as temperature and humidity, wind speed and direction, atmospheric pressure, and rainfall.

[0060] Step 402: Data Preprocessing. The multi-source raw data collected in Step 401 is processed for format unification and time synchronization. Different data formats used by different sensors (such as RS485, MODBUS and other protocol formats) are converted into a unified data format, and the acquisition time of each sensor is aligned to eliminate time deviations caused by inconsistent sampling frequencies.

[0061] Step 403: Outlier Detection and Removal. Outlier detection is performed on the preprocessed data. Specifically, the Z-score standardization method is used to calculate the standardized score of each data point. Data points with standardized scores exceeding a preset threshold are identified as outliers and removed.

[0062] Step 404: Data Fusion. A Bayesian weighted fusion algorithm is used to dynamically adjust the corresponding weight of each sensor based on its historical accuracy. The weighted fusion calculation is then performed on multiple raw data sets after outliers are removed to obtain the first fused data.

[0063] Step 405: Data Imputation. For data points that were removed or were originally missing, a data imputation strategy is used to fill them in. Specifically, for data segments with a missing duration less than a preset duration, a linear interpolation algorithm is used for imputation; for data segments with a missing duration greater than or equal to the preset duration, a pre-trained Long Short-Term Memory (LSTM) prediction model is used for data imputation.

[0064] Step 406: Intelligent Data Correction. The Kalman filter algorithm is used to predictively correct the completed data. Through the optimal estimation recursive algorithm of the Kalman filter, combined with historical data sequences and current observations, the meteorological data is smoothed and corrected, further improving the accuracy and stability of the data.

[0065] Step 407: Generate optimized data. Output optimized meteorological data after fusion, anomaly removal, completion, and intelligent correction for subsequent applications.

[0066] In some exemplary embodiments, the method provided in this application further includes a reliable transmission step. Specifically, the method provided in this application further includes: The collected meteorological data is written into the local circular buffer of the acquisition device in real time; Monitor the current network status; Based on network conditions, the data upload frequency and data packet size are adjusted using a dynamic bandwidth adaptive algorithm, and the data in the circular buffer is uploaded to the cloud server. When the network is detected to have recovered from an abnormal state to a normal state, the data cached in the circular buffer during the network anomaly will be retransmitted to the cloud server.

[0067] The collected meteorological data is written to the local circular buffer of the acquisition device in real time. The "circular buffer" is a fixed-size memory cache structure that uses a first-in, first-out circular storage method. When the buffer is full, new data automatically overwrites the oldest data, and it is used to temporarily store data in the event of network failure.

[0068] Monitor the current network conditions; based on these conditions, adjust the data upload frequency and packet size using a dynamic bandwidth adaptive algorithm, and upload the data in the circular buffer to the cloud server. The dynamic bandwidth adaptive algorithm refers to a mechanism that dynamically adjusts data transmission parameters based on the current network bandwidth conditions. When network bandwidth is low, it reduces the upload frequency or packet size; when network bandwidth is sufficient, it restores the normal upload strategy.

[0069] When the network is detected to have recovered from an abnormal state, the data cached in the circular buffer during the network outage is retransmitted to the cloud server. This mechanism ensures that locally cached data is not lost during network interruption and is automatically retransmitted after the network is restored.

[0070] In some exemplary embodiments, the size of the circular buffer is set to be able to store at least 72 hours of meteorological data.

[0071] Since the 72-hour data cache capacity can cover most network failure scenarios, it can ensure that the data can still be completely preserved in the event of a long network interruption.

[0072] As an example, the data acquisition device allocates a 100MB circular buffer in its local RAM, writing meteorological data at a rate of one record per minute. This buffer can store approximately 72 hours of data. The system monitors the connection status with the cloud server via heartbeat packets. When a network interruption is detected, it automatically adjusts the upload frequency from once per minute to once every 5 minutes and reduces the packet size. Once the network is restored, the system automatically reads the approximately 2,000 data records stored during the network interruption from the circular buffer and re-uploads them to the cloud server one by one, achieving seamless data recovery.

[0073] By caching data in a local circular buffer, data will not be lost during network anomalies; by using a dynamic bandwidth adaptive algorithm to adjust the upload strategy according to network conditions, the transmission success rate in weak network environments is improved; and by using an automatic retransmission mechanism after network recovery, reliable data transmission from the acquisition end to the cloud is achieved, solving the problem of data loss caused by network instability and ensuring the integrity of meteorological monitoring data.

[0074] The meteorological data processing method provided in this application solves the problem of data interruption caused by a single sensor failure by monitoring the working status of the first sensor and activating the second sensor at the same monitoring point to take over the data collection when a fault is detected. On this basis, by acquiring the historical data sequence before the fault and the real-time data of the backup sensor, and using a smooth transition algorithm to fuse the two, the data during the switching between the primary and backup sensors can be smoothly connected, avoiding data jumps caused by individual differences or sudden changes in the state of the sensors. Thus, even when a sensor fails, it can still output continuous and smooth meteorological data to the monitoring platform, ensuring the integrity and reliability of meteorological monitoring data.

[0075] Figure 5 This is a schematic diagram of the structure of a meteorological data processing device 500 provided for an exemplary embodiment of this application. Figure 5As shown, the device 500 includes: a status monitoring module 510, a sensor activation module 520, a data acquisition module 530, and a data processing module 540, wherein: The status monitoring module 510 is used to monitor the working status of the first sensor, which is used to collect the first meteorological parameters; The sensor activation module 520 is used to activate a second sensor at the same monitoring point as the first sensor when the first sensor is determined to be faulty based on the working status, so as to take over the collection of the first meteorological parameters from the first sensor. The data acquisition module 530 is used to acquire the first historical data sequence of the first sensor before the failure during the process of the second sensor taking over the data acquisition of the first sensor, and to acquire the first real-time data collected by the second sensor in real time. The data processing module 540 is used to fuse the first historical data sequence and the first real-time data using a preset smooth transition algorithm to generate smooth transition meteorological data.

[0076] The meteorological data processing device 500 provided in this application embodiment solves the problem of data interruption caused by a single sensor failure by monitoring the working status of the first sensor and activating the second sensor at the same monitoring point to take over data collection when a fault is determined. On this basis, by acquiring the historical data sequence before the fault and the real-time data of the backup sensor, and using a smooth transition algorithm to fuse the two, the data during the switching between the main and backup sensors can be smoothly connected, avoiding data jumps caused by individual differences or sudden changes in the state of the sensors. Thus, even when a sensor fails, it can still output continuous and smooth meteorological data to the monitoring platform, ensuring the integrity and reliability of meteorological monitoring data.

[0077] Optionally, the status monitoring module 510 is specifically used for: The operating status parameters of the first sensor are collected, and the operating status parameters include at least one of voltage and response time; And / or, The sliding window mean of the real-time data collected by the first sensor is calculated to obtain the data change trend; When the operating status parameters are abnormal, or when the data change trend exceeds the preset abnormal threshold multiple times consecutively, the first sensor is determined to be faulty.

[0078] Optionally, the preset smooth transition algorithm includes the Kalman filter algorithm and the exponentially weighted moving average algorithm; The data processing module 540 is specifically used for: The Kalman filter algorithm is used to perform prediction processing on the first historical data sequence to obtain predicted data; The predicted data is fused with the first real-time data using the exponentially weighted moving average algorithm to generate the transitional meteorological data.

[0079] Optionally, the device further includes a data generation module, specifically used for: Acquire raw data from multiple sensors for the same meteorological parameter; A Bayesian weighted fusion algorithm is used to dynamically adjust the corresponding weight of each sensor based on its historical accuracy, and to perform fusion calculation on the multiple original data to obtain the first fused data; Anomaly detection is performed on the first fused data, and data points identified as anomalies are removed; For data points that have been removed or are missing, a data completion strategy is used to fill them in, generating optimized meteorological data.

[0080] Optionally, when the data generation module performs outlier detection on the first fused data, it is specifically used for: The Z-score normalization method was used to calculate the standardized score of each data point in the first fused data. Data points whose standardized scores exceed a preset threshold are identified as outliers.

[0081] Optionally, the data completion strategy includes: For data segments with missing durations less than the preset duration, a linear interpolation algorithm is used to complete them; For data segments with missing durations greater than or equal to the preset duration, a pre-trained Long Short-Term Memory (LSTM) prediction model is used to fill in the missing data.

[0082] Optionally, the device further includes a data retransmission module for: The collected meteorological data is written into the local circular buffer of the acquisition device in real time; Monitor the current network status; Based on the network conditions, the data upload frequency and data packet size are adjusted using a dynamic bandwidth adaptive algorithm, and the data in the circular buffer is uploaded to the cloud server. When the network is detected to have recovered from an abnormal state to a normal state, the data cached in the circular buffer during the network abnormality period is retransmitted to the cloud server.

[0083] The meteorological data processing device 500 can achieve Figures 1-4 For details of the method implementation examples, please refer to [link / reference]. Figures 1-4 The meteorological data processing method shown in the embodiment will not be described in detail here.

[0084] Figure 6This is a schematic diagram of the structure of an electronic device provided as an exemplary embodiment of this application. For example... Figure 6 As shown, the device includes a memory 61 and a processor 62.

[0085] Memory 61 is used to store computer programs and can be configured to store various other data to support operation on the computing device. Examples of this data include instructions for any application or method used to operate on the computing device, contact data, phone book data, messages, images, videos, etc.

[0086] The processor 62, coupled to the memory 61, is used to execute the computer program in the memory 61 for: monitoring the operating status of a first sensor, which is used to collect first meteorological parameters; when the operating status determines that the first sensor is faulty, activating a second sensor at the same monitoring point as the first sensor to take over the collection of the first meteorological parameters; during the process of the second sensor taking over the data collection from the first sensor, acquiring a first historical data sequence of the first sensor before the fault, and acquiring first real-time data collected by the second sensor in real time; and using a preset smooth transition algorithm to fuse the first historical data sequence and the first real-time data to generate smooth transition meteorological data.

[0087] The electronic device provided in this application solves the problem of data interruption caused by a single sensor failure by monitoring the working status of the first sensor and activating the second sensor at the same monitoring point to take over the data collection when a fault is detected. On this basis, by acquiring the historical data sequence before the fault and the real-time data of the backup sensor, and using a smooth transition algorithm to fuse the two, the data during the switching between the primary and backup sensors can be smoothly connected, avoiding data jumps caused by individual differences or sudden changes in the state of the sensors. Thus, even when a sensor fails, it can still output continuous and smooth meteorological data to the monitoring platform, ensuring the integrity and reliability of meteorological monitoring data.

[0088] Furthermore, such as Figure 6 As shown, the electronic device also includes other components such as a communication component 63, a display 64, a power supply component 65, and an audio component 66. Figure 6 The diagram only shows some components and does not mean that the electronic device includes only these components. Figure 6 The components shown. Additionally, depending on the implementation of the traffic playback device, Figure 6 The components within the dashed box are optional, not mandatory. For example, when an electronic device is implemented as a terminal device such as a smartphone, tablet, or desktop computer, it may include... Figure 6The components within the dashed box; when the electronic device is implemented as a server-side device such as a conventional server, cloud server, data center, or server array, it may be excluded. Figure 6 The component within the dashed box.

[0089] Accordingly, embodiments of this application also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, enables the processor to implement the steps in the above-described meteorological data processing method embodiments.

[0090] Accordingly, this application also provides a computer program product, which stores instructions that, when executed by a computer, cause the computer to implement the steps in the meteorological data processing method embodiment provided in this application.

[0091] The above Figure 6 The communication component is configured to facilitate wired or wireless communication between the device containing the communication component and other devices. The device containing the communication component can access wireless networks based on communication standards, such as WiFi, 2G, or 3G, or combinations thereof. In one exemplary embodiment, the communication component receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, the communication component may further include a Near Field Communication (NFC) module, Radio Frequency Identification (RFID) technology, Infrared Data Association (IrDA) technology, Ultra Wideband (UWB) technology, Bluetooth (BT) technology, etc.

[0092] The above Figure 6 The memory in the memory can be implemented by any class of volatile or non-volatile storage devices or combinations thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0093] The above Figure 6 The display includes a screen, which may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen can be implemented as a touchscreen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors can sense not only the boundaries of the touch or swipe action, but also the duration and pressure associated with the touch or swipe operation.

[0094] The above Figure 6The power supply component provides power to the various components of the device in which it resides. The power supply component may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to the device in which it resides.

[0095] The above Figure 6 The audio component can be configured to output and / or input audio signals. For example, the audio component includes a microphone (MIC) configured to receive external audio signals when the device containing the audio component is in an operating mode, such as call mode, recording mode, or voice recognition mode. The received audio signals can be further stored in memory or transmitted via a communication component. In some embodiments, the audio component also includes a speaker for outputting audio signals.

[0096] Accordingly, embodiments of this application also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, enables the processor to implement the steps in the above-described meteorological data processing method embodiments.

[0097] Accordingly, this application also provides a computer program product, which stores instructions that, when executed by a computer, cause the computer to implement the steps in the meteorological data processing method embodiment provided in this application.

[0098] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0099] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0100] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0101] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0102] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0103] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0104] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other classes of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0105] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0106] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method of processing meteorological data, characterized by, include: Monitor the working status of the first sensor, which is used to collect the first meteorological parameter; When the first sensor is determined to be faulty based on the working status, the second sensor at the same monitoring point as the first sensor is activated to take over the collection of the first meteorological parameters from the first sensor. During the process of the second sensor taking over data collection from the first sensor, the first historical data sequence of the first sensor before the failure is obtained, and the first real-time data collected by the second sensor in real time is obtained. The first historical data sequence and the first real-time data are fused using a preset smooth transition algorithm to generate smooth transition meteorological data.

2. The method according to claim 1, characterized in that, The monitoring of the operating status of the first sensor includes: The operating status parameters of the first sensor are collected, and the operating status parameters include at least one of voltage and response time; And / or, The sliding window mean of the real-time data collected by the first sensor is calculated to obtain the data change trend; When the operating status parameters are abnormal, or when the data change trend exceeds the preset abnormal threshold multiple times consecutively, the first sensor is determined to be faulty.

3. The method according to claim 1 or 2, characterized in that, The preset smooth transition algorithm includes the Kalman filter algorithm and the exponentially weighted moving average algorithm; The process of fusing the first historical data sequence and the first real-time data using a preset smooth transition algorithm includes: The Kalman filter algorithm is used to perform prediction processing on the first historical data sequence to obtain predicted data; The predicted data is fused with the first real-time data using the exponentially weighted moving average algorithm to generate the transitional meteorological data.

4. The method according to claim 1, characterized in that, Also includes: Acquire raw data from multiple sensors for the same meteorological parameter; A Bayesian weighted fusion algorithm is used to dynamically adjust the corresponding weight of each sensor based on its historical accuracy, and to perform fusion calculation on the multiple original data to obtain the first fused data; Anomaly detection is performed on the first fused data, and data points identified as anomalies are removed; For data points that have been removed or are missing, a data completion strategy is used to fill them in, generating optimized meteorological data.

5. The method according to claim 4, characterized in that, The outlier detection of the first fused data includes: The Z-score normalization method was used to calculate the standardized score of each data point in the first fused data. Data points whose standardized scores exceed a preset threshold are identified as outliers.

6. The method according to claim 4 or 5, characterized in that, The data completion strategy includes: For data segments with missing durations less than the preset duration, a linear interpolation algorithm is used to complete them; For data segments with missing durations greater than or equal to the preset duration, a pre-trained Long Short-Term Memory (LSTM) prediction model is used to fill in the missing data.

7. The method according to claim 1, characterized in that, The method further includes: The collected meteorological data is written into the local circular buffer of the acquisition device in real time; Monitor the current network status; Based on the network conditions, the data upload frequency and data packet size are adjusted using a dynamic bandwidth adaptive algorithm, and the data in the circular buffer is uploaded to the cloud server. When the network is detected to have recovered from an abnormal state to a normal state, the data cached in the circular buffer during the network abnormality period is retransmitted to the cloud server.

8. A meteorological data processing device, characterized in that, include: The status monitoring module is used to monitor the working status of the first sensor, which is used to collect the first meteorological parameters. The sensor activation module is used to activate a second sensor at the same monitoring point as the first sensor when the first sensor is determined to be faulty based on the working status, so as to take over the collection of the first meteorological parameters from the first sensor. The data acquisition module is used to acquire the first historical data sequence of the first sensor before the failure during the process of the second sensor taking over the data acquisition of the first sensor, and to acquire the first real-time data collected by the second sensor in real time. The data processing module is used to fuse the first historical data sequence and the first real-time data using a preset smooth transition algorithm to generate smooth transition meteorological data.

9. A computer-readable storage medium having a computer program or instructions stored thereon, characterized in that, When the computer program or instructions are executed by a processor, they implement the steps of the method as described in any one of claims 1 to 7.

10. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the steps of the method according to any one of claims 1 to 7.

Images