Intelligent environment monitoring and regulation method for marine ranching

Abstract

The invention discloses an intelligent environment monitoring and regulation method for a marine ranch, and belongs to the field of environment monitoring, and the method comprises the steps: collecting water body parameters, employing a data fusion algorithm to generate a comprehensive environment state index, and solving the business problems of data dispersion, response lag and inaccurate adjustment in water body environment monitoring and regulation. A time sequence analysis model and an optimization algorithm are combined, potential anomalies are analyzed, a risk change trend is predicted, environment adjustment requirements are dynamically calculated, a precise regulation and control instruction set is generated and transmitted to equipment to be executed, and meanwhile, through deviation comparison between real-time feedback data and a preset threshold value, comprehensive indexes and regulation and control parameters are iteratively updated, so that environment stability is ensured. Through multi-level data processing and a self-adaptive adjustment mechanism, the timeliness and accuracy of water body environment regulation and control are remarkably improved, closed-loop management from monitoring to optimization is achieved, and efficient support is provided for intelligent decision making in a complex environment.

Description

Technical Field

[0001] This invention belongs to the field of environmental monitoring, and in particular relates to an intelligent environmental monitoring and control method for marine ranches. Background Technology

[0002] Marine ranching, as an important direction for the development of modern fisheries, shoulders the dual mission of ensuring the sustainable use of marine resources and improving economic benefits. Its intelligent management has become key to driving industry upgrading. Especially in the scenario of near-shore aquaculture of high-value fish, the stability of environmental conditions directly affects the survival rate and economic output of farmed organisms, making the application of intelligent monitoring and control technologies particularly urgent. However, the complexity and variability of the marine environment pose serious challenges to existing technologies, necessitating breakthrough methods to address these issues.

[0003] Currently, environmental monitoring and control methods for marine ranches often fall short in the face of complex environments. Many methods suffer from significant deficiencies in data integration and real-time response, especially when dealing with the efficient processing of multi-source information and rapid decision-making under extreme conditions. The lack of systematic coordination often leads to ineffective interventions. This deficiency not only affects the stability of the aquaculture environment but also increases the risk of economic losses, making the need for intelligent management even more urgent. The core technical challenge lies in achieving seamless integration of multi-source data and the corresponding precise control capabilities. In the marine environment, data from seabed monitoring stations, floating sensors, and underwater equipment are often scattered and in different formats. Integrating this information requires overcoming significant obstacles caused by signal attenuation and interference in underwater communication. When data integration faces bottlenecks, misjudgments of environmental change trends follow, especially in sudden situations where the system struggles to quickly formulate effective control strategies. For example, in an aquaculture area, if dissolved oxygen levels suddenly drop, and multiple data sources cannot be integrated in a timely manner and the further impact of ocean current changes on oxygen supply cannot be accurately predicted, control equipment may miss the optimal intervention opportunity, threatening the survival of farmed organisms.

[0004] Therefore, how to achieve real-time integration of multi-source data in complex marine environments and quickly calculate the optimal operating parameters of control equipment to cope with sudden environmental changes has become a critical issue that urgently needs to be addressed. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides a method for intelligent environmental monitoring and control in marine ranching, comprising:

[0006] Water parameters are collected through an environmental sensor network, and the water parameters are processed using a data fusion algorithm to obtain a comprehensive environmental status index.

[0007] Based on the comprehensive environmental status indicators, potential anomalies are analyzed. If anomalies are found, the time series analysis model is activated to process historical data sequences and determine the risk change trend.

[0008] By combining the aforementioned risk change trends with current fluctuation data, the environmental adjustment needs are calculated, and an optimization algorithm is used to adjust the equipment operating parameters to obtain a set of control instructions.

[0009] The control instruction set is transmitted to the control device, and the corresponding control instruction set is executed. The feedback data is monitored in real time to obtain the adjusted environmental indicators.

[0010] The adjusted environmental indicators are compared with preset thresholds. If there is a deviation, the monitoring data is re-integrated to obtain an updated comprehensive environmental status indicator.

[0011] The updated comprehensive environmental status index is input into the time series analysis model to redetermine the risk change trend and adjust the control weights to obtain the optimized control parameters.

[0012] The optimized control parameters are sent to the control device, and after iterative execution, the final environmental parameters are collected to determine whether the stable conditions have been reached, and the monitoring and control results are obtained.

[0013] Preferably, the process of obtaining the comprehensive environmental status index includes:

[0014] Water-related data are acquired through an environmental sensor network. By using a preset acquisition frequency and range, multiple monitoring points in the target water area are covered to obtain a set of original water parameters.

[0015] Based on the original set of water body parameters, a data cleaning method is used to remove outliers and noise data. If a parameter value is detected to exceed a preset threshold range, it is marked as invalid data and removed, resulting in a cleaned set of water body parameters.

[0016] The data fusion technique is applied to integrate multi-source sensor data on the cleaned water body parameter set. The parameters from different sources are uniformly processed by a weighted average method to obtain the fused water body parameter dataset.

[0017] Based on the fused water parameter dataset, the correlation between each parameter is analyzed. If the correlation between a parameter and other parameters is lower than a preset threshold, the priority is reduced, and the key parameter combination is determined.

[0018] Based on the combination of key parameters, the support vector machine algorithm is used to classify the water state and obtain the set of state features after classification.

[0019] Based on the classified set of state features, a comprehensive environmental state index is calculated, and the environmental state level is obtained by weighted summation of each state feature using a preset scoring rule.

[0020] Based on the environmental state level, a corresponding state description text is generated. The level information is compared with historical data using a preset template to obtain a record of state change trends.

[0021] Preferably, the process of determining the trend of risk change includes:

[0022] By monitoring the environmental status in real time, dynamic data streams of comprehensive indicators are obtained, and a preliminary screening is performed using a preset threshold range to determine whether there are any abnormal signals that deviate from the normal range.

[0023] If the comprehensive index is detected to exceed the preset threshold, a deep analysis is performed on the potential anomaly, relevant time series are extracted from historical data, and the persistence and fluctuation pattern of the abnormal signal are determined.

[0024] Based on the fluctuation pattern of the abnormal signal, a time series analysis model is invoked to process the extracted historical data and obtain a preliminary assessment result of the risk change.

[0025] Based on the preliminary assessment results of the risk changes, the time series is segmented using the sliding window method to obtain the trend characteristics of changes in each time period.

[0026] The characteristics of the changing trend are compared segment by segment to determine whether there are risk signals of continuous deterioration. If a deterioration trend is found, a corresponding early warning sign is generated.

[0027] Based on the warning indicators and the latest comprehensive indicators of the environmental status, the priority of risk changes is determined. For the dataset after priority ranking, a pre-established decision support module is invoked to obtain the final structured results of risk change trends and response strategies.

[0028] Preferably, the process of obtaining the control instruction set includes:

[0029] By collecting data on risk changes and current fluctuations, and using data integration techniques to process the raw information, preliminary assessment results of environmental adjustment needs are obtained.

[0030] Based on the preliminary assessment results of environmental adjustment needs, the trend and fluctuation data are analyzed, and the adjustment needs are quantified using the support vector machine algorithm to determine specific environmental adjustment target values.

[0031] Based on the environmental adjustment target value, combined with the equipment operating status and operating parameters, the current workload data of the equipment is obtained, and it is determined whether it meets the preset threshold range. If it exceeds the threshold range, a preliminary parameter adjustment plan is generated.

[0032] Based on the preliminary plan for parameter adjustment, a genetic algorithm is used to perform optimization calculations on the equipment operation and operating parameters to obtain an optimized combination of operating parameter configurations.

[0033] By combining the optimized combination of operating parameters with the generation logic of control instructions, the specific instructions for equipment adjustment are obtained, and the final control instruction set is determined.

[0034] Preferably, the process of obtaining the adjusted environmental indicators includes:

[0035] The control instruction set is obtained by extracting key parameters from the initial data of environmental indicators through a pre-established instruction generation module, and generating the corresponding control instruction set.

[0036] The control instruction set is transmitted to the control device through a secure transmission protocol, and the device's reception status is obtained.

[0037] If the device's reception status indicates successful reception, the control device is triggered to execute the instruction set, initiate the corresponding control operation, and obtain the device's response information.

[0038] Based on the device response information, the working status of the control device is monitored and adjusted in real time, and feedback data during the execution process is collected to obtain a real-time feedback dataset.

[0039] Based on the real-time feedback dataset, the support vector machine algorithm is used to classify the data, extract key information related to environmental indicators, and determine the adjusted environmental indicator values.

[0040] If the adjusted environmental index value exceeds the preset threshold range, a new set of control instructions is generated and transmitted to the control device for secondary adjustment to obtain new feedback data.

[0041] Preferably, the process of obtaining the updated comprehensive environmental status index includes:

[0042] Raw data of environmental indicators are acquired from multiple monitoring points through a sensor network. The collected data are preprocessed using standardized processing methods to obtain a preliminary environmental monitoring dataset.

[0043] Based on the preliminary environmental monitoring dataset, each environmental indicator is compared with the preset threshold to determine if there is any deviation. If an indicator exceeds the preset threshold range, it is marked as an abnormal data point.

[0044] For the marked abnormal data points, historical monitoring data within the corresponding time period is obtained, and a weighted average method is used to smooth the abnormal data points to obtain the corrected environmental indicator values.

[0045] By integrating multi-source monitoring data and using the corrected environmental indicator values, the comprehensive environmental status is classified and assessed using a support vector machine model to determine the category of the current environmental status.

[0046] Based on the current environmental status category, the data is re-compared with the preset status assessment standards. If the status category does not match the expectations, the data update mechanism is triggered to obtain the latest monitoring data.

[0047] Based on the latest monitoring data and historical data trends, time series analysis is used to predict short-term changes in environmental indicators and update the comprehensive status indicators.

[0048] Based on the predicted short-term trends, the data collection frequency is adjusted, and the density of monitoring points is increased in high-risk areas to continuously acquire environmental monitoring data.

[0049] Preferably, the process of obtaining the optimized control parameters includes:

[0050] Based on the latest environmental status data and combined with a pre-established comprehensive indicator system, the data is standardized to obtain a unified comprehensive indicator dataset.

[0051] Based on the comprehensive indicator dataset, input the preset time series analysis model, perform dynamic calculation of risk changes, and determine the current risk change trend;

[0052] If the current risk trend exceeds the preset threshold range, the control weights will be adjusted to generate an initial weight correction value.

[0053] Based on the preliminary weight correction values ​​and the trend analysis results, the control parameters are recalculated to obtain a preliminary optimized parameter set.

[0054] For the initially optimized parameter set, a secondary verification is performed based on real-time feedback data of the environmental status to determine whether the parameter set meets the requirements of dynamic adaptability.

[0055] If the verification results show that the parameter set does not meet the requirements, the matching relationship between the input data and the weight values ​​is adjusted through an iterative calculation process to obtain the final optimized parameters.

[0056] The final optimized parameters are applied to the time series analysis model to update the prediction results of risk change trends.

[0057] Preferably, the process of obtaining monitoring and control results includes:

[0058] By using a pre-established database of control parameters, optimized control parameter data for the current environment can be obtained.

[0059] Based on the control parameter data, the data is sent to the corresponding control device using an automated transmission protocol;

[0060] After receiving the control parameter data, the control device performs corresponding control operations, monitors the device's operating status in real time, and records the operation log.

[0061] After the control operation is completed, environmental parameter data is acquired, and real-time environmental information is collected from multiple monitoring points using a sensor network.

[0062] If the collected environmental parameter data does not match the preset stable condition range, the parameter deviation is calculated through a regression analysis model, and corresponding adjustment suggestions are generated.

[0063] Based on the aforementioned adjustment recommendations, new control parameters are regenerated and transmitted to the control equipment to execute a new round of control operations, determining whether stable conditions have been achieved.

[0064] The system iterates through multiple control operations until the stability condition is met, and records the control results for each operation in real time.

[0065] Compared with the prior art, the present invention has the following advantages and technical effects:

[0066] This invention discloses an intelligent water environment control method based on an environmental sensor network. It collects water parameters and uses a data fusion algorithm to generate comprehensive environmental status indicators. Based on a time series analysis model for potential anomaly triggers, it determines risk change trends and calculates environmental adjustment needs based on current fluctuation data. An optimization algorithm is then used to adjust equipment operating parameters to form a control command set, which is transmitted to the control equipment for execution and monitored in real time. If the adjusted environmental indicators deviate from preset thresholds, the monitoring data is re-fused to update the comprehensive indicators, and the control parameters are optimized until the environmental parameters reach a stable state. This invention, through multi-level data processing and dynamic optimization control, solves the problems of data complexity and insufficient real-time performance in water environment monitoring and control, significantly improving the accuracy and stability of environmental control, and realizing intelligent management and risk early warning of the water environment. Attached Figure Description

[0067] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0068] Figure 1 This is a schematic diagram of the method flow according to an embodiment of the present invention. Detailed Implementation

[0069] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0070] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0071] like Figure 1 As shown, this embodiment provides a method for intelligent environmental monitoring and control of marine ranches, including:

[0072] Water parameters are collected by an environmental sensor network, and the water parameters are processed by a data fusion algorithm to obtain a comprehensive environmental status index.

[0073] Based on the comprehensive environmental status indicators, potential anomalies are analyzed. If anomalies are found, the time series analysis model is activated to process historical data sequences and determine the trend of risk changes.

[0074] By combining risk change trends with current volatility data to calculate environmental adjustment needs, and using optimization algorithms to adjust equipment operating parameters, a set of control instructions is obtained.

[0075] The control command set is transmitted to the control equipment. After the corresponding control command set is executed, the feedback data is monitored in real time to obtain the adjusted environmental indicators.

[0076] The adjusted environmental indicators are compared with the preset thresholds. If there is a deviation, the monitoring data is re-integrated to obtain the updated comprehensive environmental status indicators.

[0077] The updated comprehensive environmental status index is input into the time series analysis model to redetermine the risk change trend and adjust the control weights to obtain the optimized control parameters.

[0078] The optimized control parameters are sent to the control equipment. After iterative execution, the final environmental parameters are collected to determine whether the stable conditions have been reached, and the monitoring and control results are obtained.

[0079] Furthermore, the process of obtaining comprehensive environmental status indicators includes:

[0080] Water-related data are acquired through an environmental sensor network. By using a preset acquisition frequency and range, multiple monitoring points in the target water area are covered to obtain a set of original water parameters.

[0081] Based on the original set of water body parameters, a data cleaning method is used to remove outliers and noise data. If a parameter value is detected to exceed the preset threshold range, it is marked as invalid data and removed, resulting in a cleaned set of water body parameters.

[0082] Data fusion technology is applied to integrate multi-source sensor data on the cleaned water body parameter set. The parameters from different sources are uniformly processed by a weighted averaging method to obtain the fused water body parameter dataset.

[0083] Based on the fused water parameter dataset, the correlation between each parameter is analyzed. If the correlation between a parameter and other parameters is lower than a preset threshold, the priority is reduced, and the key parameter combination is determined.

[0084] Based on the combination of key parameters, the support vector machine algorithm is used to classify the state of water bodies and obtain the set of state features after classification.

[0085] Based on the classified set of state features, a comprehensive environmental state index is calculated. The environmental state level is obtained by weighting and summing each state feature according to a preset scoring rule.

[0086] Based on the environmental status level, a corresponding status description text is generated. The status information is compared with historical data using a preset template to obtain a record of status change trends.

[0087] For example, in acquiring water-related data through an environmental sensor network, taking a monitoring system covering a certain sea area as an example, this embodiment deploys 10 monitoring points, each equipped with multiple types of sensors such as temperature, pH, and dissolved oxygen. Assume the preset data collection frequency is once per hour, and the coverage area is a core 5 square kilometer region of the sea. In this way, the raw data set includes temperature values ​​collected 24 times from a single monitoring point within a day, such as 20.5, 21.0, and 19.8.

[0088] During the data cleaning phase, a temperature threshold of 15 to 30 degrees Celsius was set to handle outliers and noise. If a collected temperature value was 35 degrees Celsius, significantly exceeding the range, it was marked as invalid data and discarded. The cleaned dataset will be more accurate, for example, retaining temperature values ​​between 19.5 and 22.0 degrees Celsius, ensuring the reliability of subsequent analyses.

[0089] Regarding data fusion technology, assuming that different sensors may measure the same parameter differently (e.g., the temperature at a certain point is measured as 20.8 degrees and 21.2 degrees by two sensors respectively), a weighted average method is used, giving greater weight to the higher-precision sensor, resulting in a final fused value of 21.0 degrees. This method improves data stability. When analyzing parameter correlations, assuming the correlation coefficient between dissolved oxygen and temperature is 0.85, higher than the preset threshold of 0.6, while the correlation coefficient between turbidity and temperature is only 0.3, turbidity is given lower priority, and temperature and dissolved oxygen are identified as the key parameter combination. This selection helps focus on core influencing factors. Using a support vector machine algorithm to classify water conditions, based on key parameters, water conditions are categorized into four classes: excellent, good, moderate, and poor.

[0090] For example, a temperature of 21.0 degrees Celsius and dissolved oxygen of 6.5 mg / L might be classified as "good." This classification method intuitively reflects the health status of the water body. When calculating the comprehensive environmental status index, assuming the scoring rules are that temperature accounts for 40% and dissolved oxygen accounts for 60%, the weighted sum yields a comprehensive score of 82 points, corresponding to a "good" level. This method quantifies the environmental status, facilitating management decisions.

[0091] Finally, regarding the generation of status description text, assuming the current level is "Good" and historical data is "Excellent," the template can generate the description: "The current water condition is 'Good,' a slight decrease from last month's 'Excellent,' requiring attention to temperature trends." This comparative record helps in the timely identification of problems and the implementation of appropriate measures.

[0092] The above-mentioned links support each other, ensuring the integrity and scientific nature of data collection and status assessment, and significantly improving the efficiency of water body monitoring and early warning capabilities.

[0093] Furthermore, the process of determining the trend of risk changes includes:

[0094] By monitoring the environmental status in real time, dynamic data streams of comprehensive indicators are obtained, and a preliminary screening is performed using a preset threshold range to determine whether there are any abnormal signals that deviate from the normal range.

[0095] If the comprehensive index is detected to exceed the preset threshold, a deep analysis is performed on the potential anomaly, relevant time series are extracted from historical data, and the persistence and fluctuation pattern of the abnormal signal are determined.

[0096] Based on the fluctuation patterns of abnormal signals, a time series analysis model is invoked to process the extracted historical data and obtain preliminary assessment results of risk changes.

[0097] Based on the preliminary assessment results of risk changes, the time series is segmented using the sliding window method to obtain the trend characteristics of change within each time period.

[0098] The characteristics of the changing trend are compared segment by segment to determine whether there are risk signals of continuous deterioration. If a deterioration trend is found, a corresponding early warning sign is generated.

[0099] Based on the early warning indicators and the latest comprehensive indicators of environmental conditions, the priority of risk changes is determined. For the dataset with the priority ranking, a pre-established decision support module is invoked to obtain the final structured results of risk change trends and response strategies.

[0100] For example, in this embodiment, for the dynamic data stream acquisition of aquatic environmental parameters in real-time environmental monitoring, a multi-point sensor network deployed in the target water area continuously collects key indicators such as dissolved oxygen, turbidity, and pH value. Suppose that the dissolved oxygen data at a certain monitoring point drops sharply from 8.5 mg / L to 3.2 mg / L within 24 hours, significantly lower than the normal threshold of 5.0 mg / L, the system will immediately mark it as an abnormal signal. This initial screening threshold setting can quickly identify potential problem areas, providing a basis for subsequent analysis.

[0101] For example, this embodiment focuses on in-depth analysis of abnormal signals. It extracts the dissolved oxygen time series from historical data over the past 7 days and finds that the fluctuation pattern shows a continuous downward trend, accompanied by an increase in turbidity data from 10 NTU to 25 NTU. This extraction and pattern analysis of historical data helps determine whether the anomaly is an isolated event or a long-term problem, providing data support for risk assessment.

[0102] For example, in this embodiment, when calling the time series analysis model, a trend decomposition-based method is used to split the data into two parts: long-term trend and short-term fluctuations. Assuming the analysis results show that the dissolved oxygen decline trend accelerated over the past three days, and the short-term fluctuation amplitude increased, a preliminary assessment suggests a medium-to-high risk. This analytical approach can clearly reveal the dynamic characteristics of risk changes, providing a basis for subsequent processing.

[0103] For example, this embodiment uses a segmented processing approach for the sliding window method, setting each window to 6 hours to analyze the changing trends of dissolved oxygen and turbidity within each window. If the dissolved oxygen decrease rate is 0.5 mg / L and 0.8 mg / L in the first two windows, respectively, and the turbidity continues to rise, then it can be determined that there is a risk of continued deterioration. This segmented comparison method can refine the evolution process of risk signals and improve the accuracy of early warning.

[0104] For example, after generating an early warning indicator, the system combines the latest comprehensive indicators, such as dissolved oxygen at 3.2 mg / L and turbidity at 25 NTU, to determine the risk priority as high and output an early warning indicator. This indicator generation mechanism ensures that high-risk issues are given priority attention, avoiding waste of resources.

[0105] For example, in this embodiment, the decision support module generates structured results for the prioritized dataset. It is assumed that the module's analysis recommends immediately increasing the monitoring frequency to once per hour and notifying relevant departments to inspect nearby pollution sources. This method of outputting recommendations provides clear guidance for practical operations and improves response efficiency.

[0106] For example, the structured results of the final risk change trends and response strategies are presented in the form of reports, including risk levels, trend charts, and specific recommendations. This presentation not only intuitively reflects the state of the aquatic environment but also provides a reliable basis for management decisions, ensuring that environmental issues are responded to and addressed in a timely manner.

[0107] Furthermore, the process of obtaining the control instruction set includes:

[0108] By collecting data on risk changes and current fluctuations, and using data integration techniques to process the raw information, preliminary assessment results of environmental adjustment needs are obtained.

[0109] Based on the preliminary assessment results of environmental adjustment needs, the trend and fluctuation data are analyzed, and the adjustment needs are quantified using the support vector machine algorithm to determine specific environmental adjustment target values.

[0110] Based on the environmental adjustment target value, combined with the equipment operating status and operating parameters, the current workload data of the equipment is obtained, and it is determined whether it meets the preset threshold range. If it exceeds the threshold range, a preliminary parameter adjustment plan is generated.

[0111] Based on the preliminary parameter adjustment plan, a genetic algorithm is used to optimize the equipment operation and operating parameters to obtain an optimized combination of operating parameter configurations.

[0112] By optimizing the combination of operating parameters and combining them with the generation logic of control commands, the specific command content for equipment adjustment is obtained, and the final control command set is determined.

[0113] Furthermore, the process of obtaining the adjusted environmental indicators includes:

[0114] The control instruction set is obtained by extracting key parameters from the initial data of environmental indicators through a pre-established instruction generation module, and generating the corresponding control instruction set.

[0115] The control command set is transmitted to the control device through a secure transmission protocol, and the device's reception status is obtained.

[0116] If the device's reception status shows successful reception, the control device is triggered to execute the instruction set, initiate the corresponding control operation, and obtain the device's response information.

[0117] Based on the equipment response information, monitor and control the working status of the equipment in real time, collect feedback data during the execution process, and obtain a real-time feedback dataset;

[0118] Based on the real-time feedback dataset, the support vector machine algorithm is used to classify the data, extract key information related to environmental indicators, and determine the adjusted environmental indicator values.

[0119] If the adjusted environmental indicator value exceeds the preset threshold range, a new set of control instructions is generated and transmitted to the control equipment for secondary adjustment to obtain new feedback data.

[0120] For example, employing a secure transmission protocol is particularly important during the transmission of instruction sets to the control equipment. In this embodiment, the instruction set is transmitted to the device via an encrypted channel to prevent data interception or tampering, ensuring the integrity and security of the instructions. This approach effectively protects the reliability of the control process.

[0121] It is understood that the feedback data in this embodiment includes information such as device operating power and current environmental values. Assuming the device operating power remains stable at 80% and temperature data is updated every 5 minutes, the system uses this data to form a real-time feedback dataset, providing a basis for subsequent analysis. This continuous monitoring method helps to promptly identify potential problems. The process involves using a support vector machine algorithm to classify the real-time feedback dataset.

[0122] Specifically, the system will divide the data into two categories: normal and abnormal.

[0123] For example, if the feedback data shows a temperature consistently above 26 degrees Celsius, the system classifies it as abnormal, extracts relevant information, and determines the adjusted environmental indicator value. This classification process accurately pinpoints the key points requiring adjustment. Regarding the step of generating a new instruction set when the indicator value exceeds a preset threshold, if the adjusted temperature remains at 27 degrees Celsius, higher than the target value, the system will generate a new cooling instruction, transmit it to the device for secondary adjustment, and obtain new feedback data, such as the temperature dropping to 25.5 degrees Celsius. This iterative adjustment method gradually approaches the target state. Finally, there is a step of continuously monitoring changes in environmental indicators and recording trends.

[0124] For example, the system records the temperature drop from 28 degrees to 25 degrees Celsius, analyzes the trend, and determines the final control result. This long-term recording and analysis can provide data support for subsequent optimization and improve the overall efficiency of environmental control.

[0125] Furthermore, the process of obtaining the updated integrated environmental status indicators includes:

[0126] Raw data of environmental indicators are acquired from multiple monitoring points through a sensor network. The collected data are preprocessed using standardized processing methods to obtain a preliminary environmental monitoring dataset.

[0127] Based on the preliminary environmental monitoring dataset, each environmental indicator is compared with the preset threshold to determine if there is any deviation. If an indicator exceeds the preset threshold range, it is marked as an abnormal data point.

[0128] For the marked abnormal data points, historical monitoring data within the corresponding time period is obtained, and a weighted average method is used to smooth the abnormal data points to obtain the corrected environmental indicator values.

[0129] By integrating multi-source monitoring data and using the corrected environmental indicator values, the comprehensive environmental status is classified and assessed using a support vector machine model to determine the category of the current environmental status.

[0130] Based on the current environmental status category, the data is re-compared with the preset status assessment standards. If the status category does not match the expectations, the data update mechanism is triggered to obtain the latest monitoring data.

[0131] By using the latest monitoring data and combining it with historical data trends, time series analysis methods are employed to predict short-term changes in environmental indicators and update the comprehensive status indicators.

[0132] Based on the predicted short-term trends, the data collection frequency is adjusted, and the density of monitoring points is increased in high-risk areas to continuously acquire environmental monitoring data.

[0133] It should be noted that the design of the sensor network in this embodiment needs to consider coverage and data transmission stability. For example, relay devices may be added in areas with weak signals to ensure data integrity.

[0134] For example, standardization methods for data preprocessing can be understood as converting data with different dimensions into a unified standard. For instance, temperature and humidity values ​​vary significantly; standardization transforms them into datasets with a mean of 0 and a standard deviation of 1, facilitating subsequent analysis. Assuming the original temperature data is between 35 and 40 degrees Celsius, standardization might map it to a range of 0.2 to 0.8, thus preventing any single indicator from excessively influencing the analysis results due to excessively large values.

[0135] For example, in the process of comparing data against preset thresholds and marking outlier data points, assuming the temperature threshold is set at 30 to 35 degrees Celsius, if a monitoring point's data is 38 degrees Celsius, it is marked as an anomaly. To smooth out these outlier data points, a weighted average method can be used, combining historical data from the previous three hours (e.g., 36 degrees, 37 degrees, and 37.5 degrees Celsius), assigning higher weight to recent data, resulting in a corrected value of 37.2 degrees Celsius. This approach reduces the impact of random fluctuations on the overall assessment.

[0136] For example, when using a support vector machine (SVM) model for classification and assessment, environmental conditions can be categorized into three types: normal, slightly abnormal, and severely abnormal. If, after comprehensive data analysis, the current temperature and air quality indicators are high, the model might classify it as slightly abnormal. This classification result helps to quickly locate the problem area.

[0137] It should be noted that multi-source data, such as historical environmental data and weather forecast data, should be used during model training to improve classification accuracy.

[0138] For example, when the data update mechanism is triggered, if the current state is slightly abnormal, the data collection frequency can be increased from once per hour to once every 30 minutes, while temporary monitoring points are added in high-risk areas. This dynamic adjustment method can capture environmental changes more promptly.

[0139] It should be noted that when this embodiment predicts short-term trends through time series analysis, it can combine data from the past 24 hours to predict that the temperature may rise by 1.5 degrees Celsius in the next 6 hours, thereby adjusting the monitoring strategy in advance.

[0140] For example, in high-risk areas, the data collection frequency is adjusted and the density of monitoring points is increased. For instance, monitoring points near heat sources have their collection frequency increased to once every 15 minutes, and two additional monitoring points are added in the surrounding area to form a grid-based monitoring system. This approach provides more accurate data support, offering a reliable basis for subsequent environmental control and significantly improving the comprehensiveness and response speed of monitoring.

[0141] Furthermore, the process of obtaining the optimized control parameters includes:

[0142] Based on the latest environmental status data and combined with a pre-established comprehensive indicator system, the data is standardized to obtain a unified comprehensive indicator dataset.

[0143] Based on the comprehensive indicator dataset, input the preset time series analysis model, perform dynamic calculation of risk changes, and determine the current risk change trend;

[0144] If the current risk trend exceeds the preset threshold range, the control weights will be adjusted to generate an initial weight correction value.

[0145] Based on the initial weight correction values ​​and the trend analysis results, the control parameters are recalculated to obtain a preliminary optimized parameter set.

[0146] For the initially optimized parameter set, a secondary verification is performed based on real-time feedback data of the environmental status to determine whether the parameter set meets the requirements of dynamic adaptability.

[0147] If the verification results show that the parameter set does not meet the requirements, the matching relationship between the input data and the weight values ​​is adjusted through an iterative calculation process to obtain the final optimized parameters.

[0148] The final optimized parameters are applied to the time series analysis model to update the prediction results of risk change trends.

[0149] For example, regarding the process of recalculating control parameters, this embodiment adjusts the data collection frequency by combining the revised weights and trend analysis results. Assuming the original frequency was once per hour, if the risk trend increases, it is adjusted to once every 30 minutes. This dynamic adjustment helps to obtain data more promptly and enhances the system's responsiveness.

[0150] For example, in the secondary verification of the parameter set in this embodiment, if the initially optimized parameter set shows a temperature prediction deviation of 2 degrees Celsius in the real-time feedback, exceeding the allowable range by 1 degree Celsius, then a secondary adjustment is required. This process ensures the adaptability of the parameters and reduces the impact of errors by comparing actual data with predicted data.

[0151] For example, this embodiment adjusts the matching relationship between input data and weights for iterative calculations. Through multiple simulations, it finds the optimal match when the temperature weight is 0.55 and the humidity weight is 0.25, ultimately generating optimized parameters. This method can continuously approach the optimal solution, improving the stability of the system.

[0152] For example, when applying optimized parameters to model updates for risk prediction, assuming the updated prediction shows that the air quality index risk has decreased from 85 to 78, below the threshold of 80, the system can formulate a more moderate control strategy accordingly. This approach optimizes resource allocation and avoids waste caused by excessive intervention.

[0153] Furthermore, the process of obtaining monitoring and control results includes:

[0154] By using a pre-established database of control parameters, optimized control parameter data for the current environment can be obtained.

[0155] Based on the control parameter data, the data is sent to the corresponding control equipment using an automated transmission protocol;

[0156] After receiving the control parameter data, the control equipment executes the corresponding control operations, monitors the equipment's operating status in real time, and records the operation log.

[0157] After the control operation is completed, environmental parameter data is acquired, and real-time environmental information is collected from multiple monitoring points using a sensor network.

[0158] If the collected environmental parameter data does not match the preset stable condition range, the parameter deviation is calculated through a regression analysis model, and corresponding adjustment suggestions are generated.

[0159] Based on the adjustment recommendations, new control parameters are regenerated and transmitted to the control equipment to execute a new round of control operations to determine whether stable conditions have been met.

[0160] The system iterates through multiple control operations until the stability condition is met, and records the control results for each operation in real time.

[0161] For example, this embodiment considers the establishment and application of the control parameter database from two aspects: data classification and storage, and dynamic updating. The control parameter database, as a core resource, stores historical optimization parameters for various environmental scenarios, covering key indicators such as temperature and humidity. Historical data similar to the current environment can be extracted from the database, such as parameter combinations for temperature control within the range of 25 to 28 degrees Celsius. Through an intelligent matching algorithm, the system quickly filters out the parameter data closest to the current requirements. This approach significantly improves the efficiency of parameter acquisition and ensures the targeted nature of the control measures.

[0162] For example, in the application of automated transmission protocols, the real-time nature and security of data are key considerations. Suppose control parameters need to be transmitted to multiple temperature control devices. The system uses an encrypted transmission protocol to send the parameter data to the devices in packet form, ensuring that the data is not tampered with during transmission. Simultaneously, the devices return an acknowledgment signal upon receiving the data; if no acknowledgment is received, the system automatically retransmits. This mechanism guarantees the reliability of parameter transmission, laying the foundation for subsequent control operations.

[0163] For example, this embodiment monitors the operational status of control equipment by recording the equipment's working status in real time. Suppose that when a device is performing temperature control, the operation log shows abnormal fluctuations in its power output, the system can immediately mark the device as needing inspection and notify maintenance personnel. This real-time monitoring method helps to quickly identify potential problems and ensures the stability of the control process.

[0164] For example, in regression analysis of parameter deviations, the system generates adjustment suggestions based on historical data and the current deviation value. Assuming the current temperature is 30 degrees Celsius, exceeding the target value by 2 degrees, the system, by analyzing historical control effects, suggests increasing the equipment power by 10%. This analytical approach provides a scientific reference for parameter adjustments, reducing the possibility of blind operation.

[0165] For example, in scenarios where stable conditions are not achieved after multiple adjustments, detailed data recording is particularly important. Suppose that after three rounds of adjustments, the temperature still hasn't dropped to the target range, the system will record the parameter values, environmental feedback data, and equipment operating status for each adjustment, forming a complete data chain. This recording method provides crucial clues for subsequent in-depth analysis and helps optimize control strategies.

[0166] For example, in generating and executing a new round of control parameters, the system will optimize based on feedback data from the previous adjustment. Suppose the temperature dropped by 1 degree Celsius after the previous adjustment, but still did not meet the target, the system will further fine-tune the parameters, such as increasing the equipment running time to 30 minutes. This iterative adjustment method can gradually approach the target state, improving the accuracy of control.

[0167] The above are merely preferred embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for intelligent environmental monitoring and control in marine ranching, characterized in that, include: Water parameters are collected through an environmental sensor network, and the water parameters are processed using a data fusion algorithm to obtain a comprehensive environmental status index. Based on the comprehensive environmental status indicators, potential anomalies are analyzed. If anomalies are found, the time series analysis model is activated to process historical data sequences and determine the risk change trend. By combining the aforementioned risk change trends with current fluctuation data, the environmental adjustment needs are calculated, and an optimization algorithm is used to adjust the equipment operating parameters to obtain a set of control instructions. The control instruction set is transmitted to the control device, and the corresponding control instruction set is executed. The feedback data is monitored in real time to obtain the adjusted environmental indicators. The adjusted environmental indicators are compared with preset thresholds. If there is a deviation, the monitoring data is re-integrated to obtain an updated comprehensive environmental status indicator. The updated comprehensive environmental status index is input into the time series analysis model to redetermine the risk change trend and adjust the control weights to obtain the optimized control parameters. The optimized control parameters are sent to the control device, and after iterative execution, the final environmental parameters are collected to determine whether the stable conditions have been reached, and the monitoring and control results are obtained.

2. The method according to claim 1, characterized in that, The process of obtaining comprehensive environmental status indicators includes: Water-related data are acquired through an environmental sensor network. By using a preset acquisition frequency and range, multiple monitoring points in the target water area are covered to obtain a set of original water parameters. Based on the original set of water body parameters, a data cleaning method is used to remove outliers and noise data. If a parameter value is detected to exceed a preset threshold range, it is marked as invalid data and removed, resulting in a cleaned set of water body parameters. The data fusion technique is applied to integrate multi-source sensor data on the cleaned water body parameter set. The parameters from different sources are uniformly processed by a weighted average method to obtain the fused water body parameter dataset. Based on the fused water parameter dataset, the correlation between each parameter is analyzed. If the correlation between a parameter and other parameters is lower than a preset threshold, the priority is reduced, and the key parameter combination is determined. Based on the combination of key parameters, the support vector machine algorithm is used to classify the water state and obtain the set of state features after classification. Based on the classified set of state features, a comprehensive environmental state index is calculated, and the environmental state level is obtained by weighted summation of each state feature using a preset scoring rule. Based on the environmental state level, a corresponding state description text is generated. The level information is compared with historical data using a preset template to obtain a record of state change trends.

3. The method according to claim 1, characterized in that, The process of determining risk trends includes: By monitoring the environmental status in real time, dynamic data streams of comprehensive indicators are obtained, and a preliminary screening is performed using a preset threshold range to determine whether there are any abnormal signals that deviate from the normal range. If the comprehensive index is detected to exceed the preset threshold, a deep analysis is performed on the potential anomaly, relevant time series are extracted from historical data, and the persistence and fluctuation pattern of the abnormal signal are determined. Based on the fluctuation pattern of the abnormal signal, a time series analysis model is invoked to process the extracted historical data and obtain a preliminary assessment result of the risk change. Based on the preliminary assessment results of the risk changes, the time series is segmented using the sliding window method to obtain the trend characteristics of changes in each time period. The characteristics of the changing trend are compared segment by segment to determine whether there are risk signals of continuous deterioration. If a deterioration trend is found, a corresponding early warning sign is generated. Based on the warning indicators and the latest comprehensive indicators of the environmental status, the priority of risk changes is determined. For the dataset after priority ranking, a pre-established decision support module is invoked to obtain the final structured results of risk change trends and response strategies.

4. The method according to claim 1, characterized in that, The process of obtaining the control instruction set includes: By collecting data on risk changes and current fluctuations, and using data integration techniques to process the raw information, preliminary assessment results of environmental adjustment needs are obtained. Based on the preliminary assessment results of environmental adjustment needs, the trend and fluctuation data are analyzed, and the adjustment needs are quantified using the support vector machine algorithm to determine specific environmental adjustment target values. Based on the environmental adjustment target value, combined with the equipment operating status and operating parameters, the current workload data of the equipment is obtained, and it is determined whether it meets the preset threshold range. If it exceeds the threshold range, a preliminary parameter adjustment plan is generated. Based on the preliminary plan for parameter adjustment, a genetic algorithm is used to perform optimization calculations on the equipment operation and operating parameters to obtain an optimized combination of operating parameter configurations. By combining the optimized combination of operating parameters with the generation logic of control instructions, the specific instructions for equipment adjustment are obtained, and the final control instruction set is determined.

5. The method according to claim 1, characterized in that, The process of obtaining the adjusted environmental indicators includes: The control instruction set is obtained by extracting key parameters from the initial data of environmental indicators through a pre-established instruction generation module, and generating the corresponding control instruction set. The control instruction set is transmitted to the control device through a secure transmission protocol, and the device's reception status is obtained. If the device's reception status indicates successful reception, the control device is triggered to execute the instruction set, initiate the corresponding control operation, and obtain the device's response information. Based on the device response information, the working status of the control device is monitored and adjusted in real time, and feedback data during the execution process is collected to obtain a real-time feedback dataset. Based on the real-time feedback dataset, the support vector machine algorithm is used to classify the data, extract key information related to environmental indicators, and determine the adjusted environmental indicator values. If the adjusted environmental index value exceeds the preset threshold range, a new set of control instructions is generated and transmitted to the control device for secondary adjustment to obtain new feedback data.

6. The method according to claim 1, characterized in that, The process of obtaining updated integrated environmental status indicators includes: Raw data of environmental indicators are acquired from multiple monitoring points through a sensor network. The collected data are preprocessed using standardized processing methods to obtain a preliminary environmental monitoring dataset. Based on the preliminary environmental monitoring dataset, each environmental indicator is compared with the preset threshold to determine if there is any deviation. If an indicator exceeds the preset threshold range, it is marked as an abnormal data point. For the marked abnormal data points, historical monitoring data within the corresponding time period is obtained, and a weighted average method is used to smooth the abnormal data points to obtain the corrected environmental indicator values. By integrating multi-source monitoring data and using the corrected environmental indicator values, the comprehensive environmental status is classified and assessed using a support vector machine model to determine the category of the current environmental status. Based on the current environmental status category, the data is re-compared with the preset status assessment standards. If the status category does not match the expectations, the data update mechanism is triggered to obtain the latest monitoring data. Based on the latest monitoring data and historical data trends, time series analysis is used to predict short-term changes in environmental indicators and update the comprehensive status indicators. Based on the predicted short-term trends, the data collection frequency is adjusted, and the density of monitoring points is increased in high-risk areas to continuously acquire environmental monitoring data.

7. The method according to claim 1, characterized in that, The process of obtaining optimized control parameters includes: Based on the latest environmental status data and combined with a pre-established comprehensive indicator system, the data is standardized to obtain a unified comprehensive indicator dataset. Based on the comprehensive indicator dataset, input the preset time series analysis model, perform dynamic calculation of risk changes, and determine the current risk change trend; If the current risk trend exceeds the preset threshold range, the control weights will be adjusted to generate an initial weight correction value. Based on the preliminary weight correction values ​​and the trend analysis results, the control parameters are recalculated to obtain a preliminary optimized parameter set. For the initially optimized parameter set, a secondary verification is performed based on real-time feedback data of the environmental status to determine whether the parameter set meets the requirements of dynamic adaptability. If the verification results show that the parameter set does not meet the requirements, the matching relationship between the input data and the weight values ​​is adjusted through an iterative calculation process to obtain the final optimized parameters. The final optimized parameters are applied to the time series analysis model to update the prediction results of risk change trends.

8. The method according to claim 1, characterized in that, The process of obtaining monitoring and control results includes: By using a pre-established database of control parameters, optimized control parameter data for the current environment can be obtained. Based on the control parameter data, the data is sent to the corresponding control device using an automated transmission protocol; After receiving the control parameter data, the control device performs corresponding control operations, monitors the device's operating status in real time, and records the operation log. After the control operation is completed, environmental parameter data is acquired, and real-time environmental information is collected from multiple monitoring points using a sensor network. If the collected environmental parameter data does not match the preset stable condition range, the parameter deviation is calculated through a regression analysis model, and corresponding adjustment suggestions are generated. Based on the aforementioned adjustment recommendations, new control parameters are regenerated and transmitted to the control equipment to execute a new round of control operations, determining whether stable conditions have been achieved. The system iterates through multiple control operations until the stability condition is met, and records the control results for each operation in real time.

Images