Aquaculture water treatment capacity monitoring and early warning system based on Internet of Things and edge calculation

Abstract

The invention discloses an aquaculture water treatment capacity monitoring and early warning system based on the Internet of Things and edge calculation, relates to the technical field of water treatment capacity monitoring and early warning, and aims to solve the problem that in the prior art, abnormal water use cannot be found in time by adopting manual regular copying. The system comprises a sensing layer, an edge computing layer, a transmission layer, a platform layer and an application layer which are in communication connection in sequence, the sensing layer comprises at least one electromagnetic flowmeter host and is used for collecting aquaculture water treatment capacity data; the edge computing layer comprises an intelligent communication gateway, and the intelligent communication gateway is in communication connection with the electromagnetic flowmeter host and used for preprocessing the collected water treatment capacity data; the transmission layer is used for encrypting and transmitting the preprocessed water treatment capacity data to the platform layer by adopting a self-adaptive encryption algorithm; the platform layer comprises a cloud platform and a data analysis platform. The system has the advantage of realizing real-time monitoring and early warning of the water treatment capacity.

Description

Technical Field

[0001] This invention relates to the field of water treatment volume monitoring and early warning technology, and more specifically, to an aquaculture water treatment volume monitoring and early warning system based on the Internet of Things and edge computing. Background Technology

[0002] In the daily operation of large-scale farms, the water treatment system is a core component to ensure the safety of the farming environment, control operating costs, and comply with environmental regulations. The accurate monitoring and dynamic management of water treatment volume directly determines the water resource utilization rate, equipment operation safety, and the effectiveness of farming cost control.

[0003] However, traditional aquaculture farms often monitor water treatment volume by manually recording master meter data periodically. This not only requires a significant investment of manpower but is also prone to problems such as omissions, errors, and data lags due to human error. For example, manual meter reading cycles are typically one day to one week, making it impossible to capture instantaneous water usage fluctuations in each zone in real time, thus preventing managers from promptly detecting abnormal water usage. Therefore, we propose an aquaculture water treatment volume monitoring and early warning system based on the Internet of Things (IoT) and edge computing. Summary of the Invention

[0004] The purpose of this invention is to provide a monitoring and early warning system for aquaculture water treatment based on the Internet of Things and edge computing, which aims to solve the problem that existing technologies that rely on manual periodic recording cannot detect abnormal water use in a timely manner.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a monitoring and early warning system for aquaculture water treatment based on the Internet of Things and edge computing, the system comprising a sensing layer, an edge computing layer, a transmission layer, a platform layer and an application layer that are connected in sequence. The sensing layer includes at least one electromagnetic flow meter host for collecting data on aquaculture water treatment volume. The edge computing layer includes an intelligent communication gateway, which is communicatively connected to the electromagnetic flowmeter host and is used to preprocess the collected water treatment volume data. The transmission layer is used to encrypt and transmit the pre-processed water treatment volume data to the platform layer using an adaptive encryption algorithm. The platform layer includes a cloud platform and a data analysis platform. The cloud platform is used to store water treatment volume data, and the data analysis platform is used to perform intelligent analysis on the water treatment volume data and generate early warning information. The application layer is used to display the early warning information data generated by the data analysis platform through intelligent analysis of water treatment volume data.

[0006] Preferably, the electromagnetic flow meter main unit is made of 316L stainless steel, with a built-in high-precision electromagnetic flow sensor and a measurement accuracy of ±0.5%. The electromagnetic flow meter main unit supports 4-20mA analog signal output and RS485 communication protocol, and has a local data caching function.

[0007] Preferably, the smart communication gateway is equipped with a microprocessor and a 4G / 5G module. The 4G / 5G module supports NB-IoT low-power mode. The smart communication gateway also provides at least one RS485 interface for connecting to the electromagnetic flowmeter host. The smart communication gateway has a built-in edge computing unit for preprocessing the collected water treatment data, including noise reduction, compression, and preliminary screening of outliers.

[0008] Preferably, the edge computing layer further includes an external antenna connected to the smart communication gateway to enhance wireless signal strength, and the smart communication gateway supports 4G+RS485 dual-mode switching.

[0009] Preferably, the cloud platform supports long-term storage of water treatment volume data and has a data synchronization function to ensure data consistency across multiple terminals.

[0010] Preferably, the data analysis platform includes a smart eye big data platform model and an AI early warning module; The aforementioned Insight Big Data Platform model is used to integrate water treatment volume data to achieve standardized data processing and multi-dimensional aggregation analysis. The AI ​​early warning module constructs a dynamic water consumption model based on historical water treatment data stored in the cloud platform, using the following formula: Calculate the dynamic early warning threshold when When this occurs, it is determined to be an abnormal state and an alert is triggered, where, for Dynamic warning threshold at any time, for The time corresponds to the average water treatment volume for the same historical period. This is a seasonal correction factor. This is a correction factor for the breeding cycle. , , These are the weighting coefficients. for Real-time water treatment volume at any given moment This is the early warning sensitivity coefficient.

[0011] Preferably, the data analysis platform also includes a big data analysis engine, which uses time series analysis algorithms to mine the long-term variation patterns of water treatment volume data stored in the cloud platform, and provides iterative optimization for weighting coefficients and early warning sensitivity coefficients.

[0012] Preferably, the application layer includes mobile devices and PCs.

[0013] Preferably, the data analysis platform further includes an equipment health assessment module, which calculates the current deviation rate by collecting the three-phase current parameters of the water treatment equipment in real time. ,when If the percentage is greater than 15%, the equipment is considered to be malfunctioning.

[0014] Preferably, the adaptive encryption algorithm used in the transport layer is either the AES-256 algorithm or the Chinese national standard SM4 algorithm.

[0015] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention collects aquaculture water treatment volume data in real time through the electromagnetic flowmeter host in the sensing layer. After the data is preprocessed by the edge computing layer, it is encrypted and transmitted to the platform layer by the transmission layer. The data analysis platform in the platform layer uses the AI ​​early warning module to build a dynamic water consumption model based on historical water treatment volume data. The dynamic early warning threshold is calculated by formula. Once the deviation between the real-time water treatment volume and the threshold exceeds the set range of the early warning sensitivity coefficient, it can be immediately judged as an abnormal state and an early warning is triggered. This realizes the full-process automation and real-time nature of water treatment volume data collection, analysis and abnormal early warning.

[0016] 2. In this invention, the equipment health assessment module of the platform layer will collect the three-phase current parameters of the water treatment equipment in real time, and judge the equipment operation status by calculating the current deviation rate. When the current deviation rate exceeds 15%, the abnormal operation of the equipment can be determined in time. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the system architecture of the present invention. Detailed Implementation

[0018] Example 1 This embodiment includes a perception layer, an edge computing layer, a transport layer, a platform layer, and an application layer that are connected in sequence.

[0019] in: The sensing layer includes at least one electromagnetic flow meter host for collecting aquaculture water treatment volume data. The electromagnetic flow meter host is made of 316L stainless steel and has a built-in high-precision electromagnetic flow sensor with a measurement accuracy of ±0.5%. The electromagnetic flow meter host supports 4-20mA analog signal output and RS485 communication protocol, and has a local data caching function.

[0020] in: The edge computing layer includes a smart communication gateway and an external antenna. The smart communication gateway is equipped with a microprocessor and a 4G / 5G module. The 4G / 5G module supports NB-IoT low-power mode. The smart communication gateway also provides at least one RS485 interface and supports 4G+RS485 dual-mode switching for connecting to the electromagnetic flowmeter host. The smart communication gateway has a built-in edge computing unit for preprocessing the collected water treatment data, including noise reduction, compression, and preliminary screening of outliers. The external antenna is connected to the smart communication gateway to enhance the wireless signal strength.

[0021] In this embodiment, noise reduction, compression, and preliminary outlier screening are implemented in the following manner; I. Noise Reduction: Noise type identification: The edge computing unit first performs short-term fluctuation analysis on the raw data every 10 seconds to identify two typical types of noise—impulse noise and random noise; Impulse noise: The data suddenly shows an instantaneous value that is far beyond the normal range, which is manifested as a single data point deviating from the mean of 10 adjacent data points by more than 50%; Random noise: Data fluctuates irregularly within the normal range, characterized by the standard deviation of 20 consecutive data points exceeding the normal fluctuation threshold (the normal fluctuation threshold is set based on historical data and is usually [value missing]). 0.2m 3 / h).

[0022] Layered noise reduction processing: For impulse noise: a median filtering method is used, taking the current data and four data points before and after it, for a total of nine data points, sorting them, and replacing the current data with the median value. If three or more impulse noises occur consecutively, a temporary buffering mechanism is triggered to pause data output until five consecutive data points return to the normal fluctuation range.

[0023] For random noise: a moving average filtering method is used, with a sliding window size of 5 (i.e., calculating the average of the current data and the previous 4 data each time). The calculation formula is: after filtering... ,in for Raw time data, filtered for The data is denoised at all times, while limiting the maximum deviation of the data within the sliding window. If the deviation of a data point from the mean within the window exceeds a certain threshold... 0.3m 3 If the value is / h, then remove that data and recalculate the average.

[0024] II. Compression: Data format conversion: Convert the raw data output by the electromagnetic flowmeter main unit into an integer and decimal separation format, where the integer part retains 3 digits (maximum representing 999 m³ / h, meeting the maximum daily water treatment capacity requirement of the aquaculture farm), and the decimal part retains 2 digits (accuracy 0.01 m³ / h, balancing accuracy and compression efficiency), discarding the last redundant data; Differential encoding compression: A differential encoding strategy is used, with a compression cycle of 10 minutes. The specific operation is as follows: Record the first data point (baseline data) within the recording period, preserving its integer and decimal information in its entirety; Each subsequent data point only records the difference between it and the previous data point. If the absolute value of the difference is ≤ 5.00 m³ / h, it is represented by 1 byte (where the highest bit is the sign bit and the remaining 7 bits represent the value, with a precision of 0.01 m³ / h and a range of -6.39 to +6.39 m³ / h, covering normal fluctuations). If the absolute value of the difference is > 5.00 m³ / h, it is determined to be a data mutation, and the complete data is re-recorded as a new baseline. A 1-byte checksum is added to the end of each compression cycle for data integrity verification after transmission.

[0025] Data encapsulation after compression: The data of each compression cycle is encapsulated in the format of "cycle identifier (2 bytes) + base data (4 bytes) + difference sequence (n bytes) + check code (1 byte)," where n is the number of other data in the cycle besides the base data (approximately 60 data per 10 minutes, n=59, the difference sequence is approximately 59 bytes, and the total data volume of a single cycle is approximately 66 bytes).

[0026] III. Preliminary screening and processing of outliers: Based on historical water treatment data from the aquaculture farm, three types of static thresholds are preset: Measuring range threshold: Based on the electromagnetic flowmeter's measuring range, set a lower limit (when there is no water flow) and an upper limit (to avoid sensor over-range error). Data exceeding this range is judged as "measuring range abnormality". Rate threshold: Sets the maximum rate of change of water treatment volume per unit time. If the rate of change of two consecutive data exceeds this value, it is judged as "abnormal rate". Time period thresholds: Based on the production patterns of the farm, normal ranges are set for different time periods (e.g., 0-6 am is the equipment maintenance period, with a normal range of 0-50 m³ / h; 8-18 am is the active breeding period, with a normal range of 80-300 m³ / h). Data exceeding the corresponding time period range is judged as "time period abnormal".

[0027] The transport layer is used to encrypt the pre-processed water treatment data to the platform layer using an adaptive encryption algorithm. In this embodiment, the AES-256 algorithm is used, with a fixed block length of 128 bits (i.e., encrypting 16 bytes of data each time) and a key length of 256 bits (32 bytes). The encryption process requires 14 rounds of "round transformation" operations. Each round includes four steps: byte substitution (SubBytes), row shifting (ShiftRows), column mixing (MixColumns), and round key addition (AddRoundKey). The encryption complexity is higher, making it more difficult to attack even in the face of potential threats from quantum computing, which meets the secure transmission requirements of aquaculture water treatment data (involving key parameters of aquaculture production).

[0028] in: The platform layer includes a cloud platform and a data analysis platform. The cloud platform is used to store water treatment volume data and supports long-term storage of water treatment volume data. It also has a data synchronization function to ensure data consistency across multiple terminals. The data analysis platform is used to perform intelligent analysis on water treatment volume data and generate early warning information. The data analysis platform includes the Insight Big Data Platform Model, data dashboard, AI early warning module, big data analysis engine, and equipment health assessment module; The Huiyan Big Data Platform model is used to integrate water treatment volume data to achieve standardized data processing and multi-dimensional aggregation analysis. The AI ​​early warning module constructs a dynamic water consumption model based on historical water treatment data stored in the cloud platform, using the following formula: Calculate the dynamic early warning threshold when When this occurs, it is determined to be an abnormal state and an alert is triggered, where, for Dynamic warning threshold at any time, for The time corresponds to the average water treatment volume for the same historical period. The seasonal correction factor is 1.2-1.5 for summer, 0.8-1.0 for winter, and 1.0 for spring and autumn. The correction factor for the breeding cycle is 0.7-0.9 for the juvenile stage, 1.0-1.2 for the growth stage, and 1.1-1.3 for the mature stage. , , These are the weighting coefficients (the sum of the three equals 1). for Real-time water treatment volume at any given moment The early warning sensitivity coefficient (with a value range of 0.1-0.3). The big data analytics engine uses time series analysis algorithms (ARIMA model in this embodiment) to mine the long-term variation patterns (trend, periodic, and fluctuation characteristics) of water treatment data stored in the cloud platform, providing iterative optimization for the weight coefficients and early warning sensitivity coefficients of the dynamic water consumption model; The equipment health assessment module calculates the current deviation rate by collecting the three-phase current parameters of the water treatment equipment in real time. ,when If the percentage is greater than 15%, the equipment is considered to be malfunctioning.

[0029] The application layer includes mobile and PC terminals, which are used to display the early warning information data generated by the data analysis platform through intelligent analysis of water treatment volume data.

[0030] In this embodiment, the entire process of aquaculture water treatment volume data from collection, preprocessing, encrypted transmission to intelligent analysis and early warning display can be efficiently operated. The sensing layer adopts an electromagnetic flow meter host made of 316L stainless steel and has a built-in high-precision electromagnetic flow sensor. It accurately collects aquaculture water treatment volume data with a measurement accuracy of ±0.5%. At the same time, it supports multiple signal outputs and local data caching, laying a high-quality foundation for subsequent data processing. The edge computing layer, equipped with a microprocessor, a 4G / 5G module (supporting NB-IoT low-power mode) and an intelligent communication gateway with an RS485 interface, enables flexible access to the electromagnetic flowmeter host. Its built-in edge computing unit can perform noise reduction, compression and preliminary screening of outliers on the collected data. Combined with an external antenna to enhance the wireless signal, it ensures the efficiency of data preprocessing and the stability of data transmission. The transport layer uses the AES-256 algorithm to ensure the security of preprocessed data during transmission to the platform layer, preventing data leakage or tampering. The cloud platform at the platform layer enables long-term storage of water treatment volume data and synchronization of data across multiple terminals, ensuring data integrity and consistency. The data analysis platform integrates data through the Huiyan Big Data Platform model and achieves standardized processing and multi-dimensional aggregation analysis. The AI ​​early warning module builds a dynamic water consumption model based on historical data, and calculates dynamic early warning thresholds by combining seasonal, aquaculture cycle correction coefficients and weighting coefficients. It promptly identifies data anomalies and triggers early warnings, facilitating relevant personnel to monitor aquaculture water treatment volume and equipment operating status in real time.

[0031] Example 2 This embodiment is basically the same as Embodiment 1. The difference is that when the transmission layer encrypts the pre-processed water treatment data to the platform layer using an adaptive encryption algorithm, it uses the national cryptographic SM4 algorithm. Compared with Embodiment 1, the encryption process requires 32 rounds of "round transformation". Each round includes three steps: S-box substitution, linear transformation, and round key addition. The number of rounds is much greater than that in Embodiment 1. The difference in key length is compensated by increasing the computational complexity. The overall logic is relatively simple, without complex column mixing operations, which can achieve efficient encryption and reduce hardware resource consumption.

[0032] This embodiment specifically includes a perception layer, an edge computing layer, a transmission layer, a platform layer, and an application layer that are connected in sequence via communication.

[0033] in: The sensing layer includes at least one electromagnetic flow meter host for collecting aquaculture water treatment volume data. The electromagnetic flow meter host is made of 316L stainless steel and has a built-in high-precision electromagnetic flow sensor with a measurement accuracy of ±0.5%. The electromagnetic flow meter host supports 4-20mA analog signal output and RS485 communication protocol, and has a local data caching function.

[0034] in: The edge computing layer includes a smart communication gateway and an external antenna. The smart communication gateway is equipped with a microprocessor and a 4G / 5G module. The 4G / 5G module supports NB-IoT low-power mode. The smart communication gateway also provides at least one RS485 interface and supports 4G+RS485 dual-mode switching for connecting to the electromagnetic flowmeter host. The smart communication gateway has a built-in edge computing unit for preprocessing the collected water treatment data, including noise reduction, compression, and preliminary screening of outliers. The external antenna is connected to the smart communication gateway to enhance the wireless signal strength.

[0035] The transport layer is used to encrypt the pre-processed water treatment volume data and transmit it to the platform layer using an adaptive encryption algorithm. In this embodiment, the national cryptographic algorithm SM4 is used.

[0036] in: The platform layer includes a cloud platform and a data analysis platform. The cloud platform is used to store water treatment volume data and supports long-term storage of water treatment volume data. It also has a data synchronization function to ensure data consistency across multiple terminals. The data analysis platform is used to perform intelligent analysis on water treatment volume data and generate early warning information. The data analysis platform includes the Insight Big Data Platform Model, data dashboard, AI early warning module, big data analysis engine, and equipment health assessment module; The Huiyan Big Data Platform model is used to integrate water treatment volume data to achieve standardized data processing and multi-dimensional aggregation analysis. The AI ​​early warning module constructs a dynamic water consumption model based on historical water treatment data stored in the cloud platform, using the following formula: Calculate the dynamic early warning threshold when When this occurs, it is determined to be an abnormal state and an alert is triggered, where, for Dynamic warning threshold at any time, for The time corresponds to the average water treatment volume for the same historical period. The seasonal correction factor is 1.2-1.5 for summer, 0.8-1.0 for winter, and 1.0 for spring and autumn. The correction factor for the breeding cycle is 0.7-0.9 for the juvenile stage, 1.0-1.2 for the growth stage, and 1.1-1.3 for the mature stage. , , These are the weighting coefficients (the sum of the three equals 1). for Real-time water treatment volume at any given moment The early warning sensitivity coefficient (with a value range of 0.1-0.3). The big data analytics engine uses time series analysis algorithms (ARIMA model in this embodiment) to mine the long-term variation patterns (trend, periodic, and fluctuation characteristics) of water treatment data stored in the cloud platform, providing iterative optimization for the weight coefficients and early warning sensitivity coefficients of the dynamic water consumption model; The equipment health assessment module calculates the current deviation rate by collecting the three-phase current parameters of the water treatment equipment in real time. ,when If the percentage is greater than 15%, the equipment is considered to be malfunctioning.

[0037] The application layer includes mobile and PC terminals, which are used to display the early warning information data generated by the data analysis platform through intelligent analysis of water treatment volume data.

[0038] The embodiments disclosed in this invention are preferred embodiments, but are not limited thereto. Those skilled in the art can easily understand the spirit of this invention based on the above embodiments and make different extensions and variations, but as long as they do not depart from the spirit of this invention, they are all within the protection scope of this invention.

Claims

1. A monitoring and early warning system for aquaculture water treatment capacity based on the Internet of Things and edge computing, characterized in that, The system comprises a perception layer, an edge computing layer, a transmission layer, a platform layer, and an application layer that are connected in sequence via communication. The sensing layer includes at least one electromagnetic flow meter host for collecting data on aquaculture water treatment volume. The edge computing layer includes an intelligent communication gateway, which is communicatively connected to the electromagnetic flowmeter host and is used to preprocess the collected water treatment volume data. The transmission layer is used to encrypt and transmit the pre-processed water treatment volume data to the platform layer using an adaptive encryption algorithm. The platform layer includes a cloud platform and a data analysis platform. The cloud platform is used to store water treatment volume data, and the data analysis platform is used to perform intelligent analysis on the water treatment volume data and generate early warning information. The application layer is used to display the early warning information data generated by the data analysis platform through intelligent analysis of water treatment volume data.

2. The aquaculture water treatment capacity monitoring and early warning system based on the Internet of Things and edge computing according to claim 1, characterized in that, The electromagnetic flow meter main unit is made of 316L stainless steel and has a built-in high-precision electromagnetic flow sensor with a measurement accuracy of ±0.5%. The electromagnetic flow meter main unit supports 4-20mA analog signal output and RS485 communication protocol, and has a local data caching function.

3. The aquaculture water treatment capacity monitoring and early warning system based on the Internet of Things and edge computing according to claim 1, characterized in that, The intelligent communication gateway is equipped with a microprocessor and a 4G / 5G module. The 4G / 5G module supports NB-IoT low-power mode. The intelligent communication gateway also provides at least one RS485 interface for connecting to the electromagnetic flowmeter host. The intelligent communication gateway has a built-in edge computing unit for preprocessing the collected water treatment data, including noise reduction, compression, and preliminary screening of outliers.

4. The aquaculture water treatment volume monitoring and early warning system based on the Internet of Things and edge computing according to claim 3, characterized in that, The edge computing layer also includes an external antenna, which is connected to the smart communication gateway to enhance wireless signal strength. The smart communication gateway supports 4G+RS485 dual-mode switching.

5. A monitoring and early warning system for aquaculture water treatment capacity based on the Internet of Things and edge computing as described in claim 1, characterized in that, The cloud platform supports long-term storage of water treatment volume data and has a data synchronization function to ensure data consistency across multiple terminals.

6. The aquaculture water treatment volume monitoring and early warning system based on the Internet of Things and edge computing according to claim 1, characterized in that, The data analysis platform includes the Insight Big Data Platform Model and an AI Early Warning Module; The aforementioned Insight Big Data Platform model is used to integrate water treatment volume data to achieve standardized data processing and multi-dimensional aggregation analysis. The AI ​​early warning module constructs a dynamic water consumption model based on historical water treatment data stored in the cloud platform, using the following formula: Calculate the dynamic early warning threshold when When this occurs, it is determined to be an abnormal state and an alert is triggered, where, for Dynamic warning threshold at any time, for The time corresponds to the average water treatment volume for the same historical period. This is a seasonal correction factor. This is a correction factor for the breeding cycle. , , These are the weighting coefficients. for Real-time water treatment volume at any given moment This is the early warning sensitivity coefficient.

7. A monitoring and early warning system for aquaculture water treatment capacity based on the Internet of Things and edge computing as described in claim 6, characterized in that, The data analysis platform also includes a big data analysis engine, which uses time series analysis algorithms to mine the long-term variation patterns of water treatment volume data stored in the cloud platform, providing iterative optimization for weighting coefficients and early warning sensitivity coefficients.

8. A monitoring and early warning system for aquaculture water treatment capacity based on the Internet of Things and edge computing as described in claim 1, characterized in that, The application layer includes mobile and PC clients.

9. A monitoring and early warning system for aquaculture water treatment capacity based on the Internet of Things and edge computing as described in claim 6, characterized in that, The data analysis platform also includes an equipment health assessment module, which calculates the current deviation rate by collecting the three-phase current parameters of the water treatment equipment in real time. ,when If the percentage is greater than 15%, the equipment is considered to be malfunctioning.

10. A monitoring and early warning system for aquaculture water treatment capacity based on the Internet of Things and edge computing as described in claim 1, characterized in that, The adaptive encryption algorithm used in the transport layer is either AES-256 or the Chinese national standard SM4 algorithm.

Images