A remote monitoring system and method for fishery-solar hybrid systems on mobile devices

Abstract

This invention provides a remote monitoring system and method for solar-fishery complementary energy systems based on a mobile terminal, belonging to the field of photovoltaic power generation control or regulation technology. It includes: a data acquisition module: deploying a sensor array in the solar-fishery complementary area to collect environmental data, and simultaneously collecting power generation data based on a pre-set photovoltaic panel monitoring device; an information conversion module: preprocessing the collected environmental and power generation data and performing information encoding conversion; a mode determination module: extracting features from the encoded data based on a pre-set algorithm to determine preliminary fault modes; a rule construction module: constructing a fault database based on the data corresponding to the preliminary fault modes, and then constructing early warning rules by combining historical data; and an information output module: determining early warning information based on the collected environmental data, power generation data, and early warning rules, and visualizing the early warning information, thereby optimizing energy management efficiency and reducing manual intervention and potential losses.

Description

Technical Field

[0001] This invention relates to the field of photovoltaic power generation control or regulation technology, and in particular to a remote monitoring system and method for fishery-solar hybrid power generation for mobile terminals. Background Technology

[0002] With the development of solar-aquaculture complementary technology, more and more solar-aquaculture complementary projects are being applied to the fields of new energy and ecological protection.

[0003] Existing solar-aquaculture hybrid monitoring systems primarily collect environmental and power generation data using independent sensors and equipment. However, these systems often lack intelligent data processing and fault prediction capabilities, leading to difficulties in timely fault detection and a lack of effective early warning mechanisms. Current technical solutions largely rely on manual analysis, resulting in slow response times and low efficiency.

[0004] Therefore, the present invention provides a remote monitoring system and method for fishery-solar hybrid systems on mobile devices. Summary of the Invention

[0005] This invention provides a remote monitoring system and method for solar-fishery complementary systems on mobile devices. By integrating data acquisition, information conversion, pattern recognition, rule construction, and information output modules, it achieves comprehensive monitoring of solar-fishery complementary areas. Compared with existing technologies, it can collect environmental and power generation data in real time and automatically analyze the data through intelligent algorithms to identify potential fault modes. By constructing a fault database and combining historical data with early warning rules, the system can provide timely warnings before faults occur, greatly improving the operational reliability of the photovoltaic system. In addition, the system supports visual output, facilitating rapid response and decision-making by users, optimizing energy management efficiency, and reducing manual intervention and potential losses.

[0006] This invention provides a remote monitoring system for fishery-solar hybrid systems for mobile devices, comprising:

[0007] Data acquisition module: Deploy sensor arrays in the fishery-solar hybrid area to collect environmental data, and at the same time, collect power generation data based on the pre-set photovoltaic panel monitoring equipment;

[0008] Information conversion module: preprocesses the collected environmental and power generation data and performs information encoding conversion;

[0009] Pattern determination module: Based on a preset algorithm, features are extracted from the encoded and converted data to determine the preliminary fault mode;

[0010] Rule building module: Builds a fault database based on the data corresponding to the initial fault modes, and then builds early warning rules by combining historical data;

[0011] Information output module: Based on the collected environmental data, power generation data and early warning rules, determine the early warning information and visualize the early warning information.

[0012] This invention provides a data acquisition module for a mobile-based fishery-solar hybrid remote monitoring system, comprising:

[0013] Regional Analysis Unit: The fishery-solar complementary area includes several pre-defined areas. Regional analysis is performed on each area to determine the regional characteristics of each area.

[0014] Environmental data acquisition unit: Deploy corresponding sensor arrays based on the regional characteristics of each area, and collect environmental data for each area based on the corresponding sensor arrays and the preset acquisition frequency;

[0015] Power generation data acquisition unit: Collects power generation data based on preset photovoltaic panel monitoring equipment and preset acquisition frequency.

[0016] This invention provides a remote monitoring system for fishery-solar hybrid systems for mobile devices, including an information conversion module comprising:

[0017] Knowledge graph construction unit: Based on a preset knowledge graph algorithm, a knowledge graph in the field of fishery-solar complementary industries is constructed, and the collected environmental data and power generation data are regarded as nodes in the knowledge graph;

[0018] Data Analysis Unit: Performs initial analysis on environmental and power generation data to determine the position and weight of each node in the knowledge graph;

[0019] Data processing unit: Normalizes environmental and power generation data based on graph neural networks;

[0020] Data conversion unit: Based on a preset adversarial network algorithm, it performs information encoding conversion on the normalized environmental data and power generation data, and then obtains the encoded data.

[0021] This invention provides a remote monitoring system for fishery-solar hybrid systems for mobile devices, including a mode determination module comprising:

[0022] Feature extraction unit: Extracts features from encoded and converted data based on a preset algorithm;

[0023] Feature analysis unit: Performs feature analysis on the extracted features to determine several feature parameters;

[0024] Mode determination unit: Determines preliminary fault coefficients based on feature parameters, and determines the corresponding preliminary fault mode by combining them with a preset coefficient-mode database.

[0025] This invention provides a remote monitoring system for solar-fishery hybrid systems for mobile devices, which extracts features including environmental features and power generation features.

[0026] This invention provides a remote monitoring system for fishery-solar hybrid systems for mobile devices, including a feature analysis unit comprising:

[0027] Deviation Acquisition Subunit: By analyzing the light intensity characteristics in the environment, the long-term trend line of light intensity is fitted using the least squares method to determine the deviation value between the light intensity and the trend line at each time point;

[0028] Parameter determination sub-unit: Perform statistical analysis on the deviation values ​​to determine the trend deviation parameters;

[0029] Derivative Determination Sub-unit: For the current characteristic in the power generation characteristics, define a time window and determine the first derivative of the current within the window;

[0030] The mutation determination sub-unit is determined by analyzing the changes in the first derivative, identifying the mutation point of the current, and then determining the sum of the absolute values ​​of the derivatives at the mutation point as the mutation degree parameter.

[0031] Feature joint sub-unit: Construct a joint feature space for environmental features and power generation features, use the mutual information algorithm to determine the mutual information value of light intensity features and power generation features in the joint feature space, and then use the mutual information value as the coupling feature parameter.

[0032] This invention provides a remote monitoring system for fishery-solar hybrid systems for mobile devices, including a mode determination unit comprising:

[0033] Coefficient Determination Subunit: Determining Preliminary Fault Coefficients Based on Characteristic Parameters

[0034]

[0035] in, This is the preliminary failure coefficient. For trend deviation parameters, This is a parameter representing the degree of mutation. For coupling characteristic parameters, For symbolic functions, , , These are preset hyperparameters. These are preset adjustment parameters;

[0036] Pattern determination subunit: Based on the preliminary fault coefficients and the preset coefficient-pattern database, the corresponding preliminary fault mode is determined.

[0037] This invention provides a remote monitoring system for fishery-solar hybrid systems on mobile devices, including a rule construction module comprising:

[0038] The initial failure modes and their corresponding environmental and power generation data are stored in the failure database. At the same time, historical failure data is acquired, organized, and classified.

[0039] Based on a preset data processing algorithm, all data in the fault database is deeply mined and correlation analysis is performed.

[0040] Based on the results of data correlation analysis and combined with pre-defined domain knowledge, early warning rules based on data features and fault correlations are constructed.

[0041] This invention provides a remote monitoring method for fishery-solar hybrid systems on a mobile device, comprising:

[0042] Step 1: Deploy a sensor array in the solar-fishery complementary area to collect environmental data, and at the same time, collect power generation data based on the pre-set photovoltaic panel monitoring equipment;

[0043] Step 2: Preprocess the collected environmental and power generation data and perform information encoding conversion;

[0044] Step 3: Extract features from the encoded and converted data based on a preset algorithm to determine the initial fault mode;

[0045] Step 4: Build a fault database based on the data corresponding to the initial fault modes, and then build early warning rules by combining historical data;

[0046] Step 5: Determine early warning information based on the collected environmental data, power generation data, and early warning rules, and visualize the early warning information.

[0047] Compared with the prior art, the beneficial effects of this application are as follows:

[0048] By integrating data acquisition, information conversion, pattern recognition, rule construction, and information output modules, comprehensive monitoring of the solar-fishery complementary area is achieved. Compared with existing technologies, it can collect environmental and power generation data in real time and automatically analyze the data through intelligent algorithms to identify potential fault modes. By constructing a fault database and combining historical data with early warning rules, the system can provide timely warnings before faults occur, greatly improving the operational reliability of the photovoltaic system. In addition, the system supports visual output, which facilitates users to respond quickly and make decisions, optimizes energy management efficiency, and reduces manual intervention and potential losses. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0050] Figure 1 This is a schematic diagram of a remote monitoring system for fishery-solar hybrid systems provided in an embodiment of the present invention.

[0051] Figure 2 This is a flowchart illustrating a remote monitoring method for fishery-solar hybrid systems on a mobile device, provided by an embodiment of the present invention. Detailed Implementation

[0052] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0053] Example 1:

[0054] This invention provides a remote monitoring system for fishery-solar hybrid systems for mobile devices, such as... Figure 1 As shown, it includes:

[0055] A remote monitoring system for fishery-solar hybrid systems for mobile devices, characterized in that it comprises:

[0056] Data acquisition module: Deploy sensor arrays in the fishery-solar hybrid area to collect environmental data, and at the same time, collect power generation data based on the pre-set photovoltaic panel monitoring equipment;

[0057] Information conversion module: preprocesses the collected environmental and power generation data and performs information encoding conversion;

[0058] Pattern determination module: Based on a preset algorithm, features are extracted from the encoded and converted data to determine the preliminary fault mode;

[0059] Rule building module: Builds a fault database based on the data corresponding to the initial fault modes, and then builds early warning rules by combining historical data;

[0060] Information output module: Based on the collected environmental data, power generation data and early warning rules, determine the early warning information and visualize the early warning information.

[0061] In this embodiment, the sensor array integrates a high-precision light sensor based on MEMS technology, a water quality sensor based on nanomaterials, a photovoltaic module performance sensor with integrated AI chip, and a sensor for monitoring the physical parameters of the device (such as vibration, stress, etc.).

[0062] In this embodiment, the preliminary fault mode refers to the initial signs or trends of potential faults identified by the system after feature extraction using a preset algorithm based on collected and processed data (such as environmental data and power generation data). This is the first step in fault detection, helping the system to initially determine the likelihood of potential anomalies or faults. For example, in a solar-aquaculture hybrid system, if the power generation data of the photovoltaic panel suddenly drops, or if there are abnormal changes in temperature and humidity in the environmental data, the system may use a preliminary algorithm to analyze these changes and identify certain possible fault modes, such as photovoltaic panel surface contamination, equipment aging, or disconnected connection wires.

[0063] In this embodiment, the early warning rules are a set of rules built based on historical data and preliminary failure modes collected by the system. These rules are used to identify and predict potential system failures, and an early warning is triggered when the data collected by the system in real time matches the rules. Early warning rules are typically implemented through machine learning or manually formulated rules. For example, assuming the photovoltaic panel's power generation efficiency is below a preset threshold, if the ambient temperature is high and the humidity is low, the early warning rule could be set as "If the power generation efficiency drops by more than 10% and the ambient temperature exceeds 40°C, then trigger an overheating risk warning." This type of early warning rule can help detect potential overheating problems in the system.

[0064] In this embodiment, the early warning information refers to the warning message issued by the system when an anomaly is detected after comparing the system's real-time data with the early warning rules. This information typically includes the possible fault type, severity level, time of occurrence, and location, helping operators quickly locate the problem and take necessary measures. For example, when the system detects abnormal photovoltaic panel power generation data that meets the early warning rules, the system will generate an early warning message, such as: "Photovoltaic panel fault warning: The power generation efficiency of photovoltaic panel group 1 has decreased by 15%, possibly due to equipment aging or loose connections. Please check promptly." This information can be pushed to relevant maintenance personnel via mobile devices, reminding them to check the equipment and repair it.

[0065] The beneficial effects of the above technical solution are as follows: By integrating data acquisition, information conversion, pattern recognition, rule construction, and information output modules, comprehensive monitoring of the solar-fishery complementary area is achieved. Compared with existing technologies, it can collect environmental and power generation data in real time and automatically analyze the data through intelligent algorithms to identify potential fault modes. By constructing a fault database and combining historical data with early warning rules, the system can provide timely warnings before faults occur, greatly improving the operational reliability of the photovoltaic system. In addition, the system supports visual output, which facilitates rapid response and decision-making by users, optimizes energy management efficiency, and reduces manual intervention and potential losses.

[0066] Example 2:

[0067] This invention provides a remote monitoring system for fishery-solar hybrid systems for mobile devices, including a data acquisition module comprising:

[0068] Regional Analysis Unit: The fishery-solar complementary area includes several pre-defined areas. Regional analysis is performed on each area to determine the regional characteristics of each area.

[0069] Environmental data acquisition unit: Deploy corresponding sensor arrays based on the regional characteristics of each area, and collect environmental data for each area based on the corresponding sensor arrays and the preset acquisition frequency;

[0070] Power generation data acquisition unit: Collects power generation data based on preset photovoltaic panel monitoring equipment and preset acquisition frequency.

[0071] In this embodiment, the preset area includes: a photovoltaic panel area and a breeding area;

[0072] In this embodiment, deploying corresponding sensor arrays based on the regional characteristics of each area includes: In the photovoltaic panel area, installing light sensors in locations that can receive sufficient and unobstructed sunlight, such as the frame or support of the photovoltaic panel, to ensure accurate reflection of the sunlight received by the photovoltaic panel. Photovoltaic module performance sensors are directly installed on the photovoltaic panel, close to the cells or key circuit components, to collect power generation data in real time. Vibration sensors are installed at key nodes of the photovoltaic module frame and support structure, while stress sensors are installed in areas prone to stress concentration, such as the joints of the support and key load-bearing components. In aquaculture waters, water quality sensors are installed in water at different depths and locations to comprehensively monitor the vertical and horizontal distribution of water quality. For example, sensors are deployed in the center of the water area, near the shore, and in areas with concentrated aquaculture organisms to ensure accurate reflection of the water quality of the entire water area. Simultaneously, attention must be paid to the installation depth and fixing method of the sensors to avoid interference from water flow and aquaculture organisms.

[0073] In this embodiment, the preset photovoltaic panel monitoring equipment and preset acquisition frequency are used to collect power generation data. The photovoltaic module performance sensor with integrated AI chip is installed on the photovoltaic panel to collect power generation data such as voltage, current, power, and temperature of the photovoltaic panel in real time. At the same time, the installation parameters such as the orientation and tilt angle of the photovoltaic panel are recorded to provide basic data for subsequent analysis of the power generation efficiency and performance of the photovoltaic panel.

[0074] The beneficial effects of the above technical solution are as follows: By accurately dividing the solar-fishery complementary areas and analyzing the characteristics of each area, more customized and efficient data collection can be achieved. Compared with existing technologies, deploying sensor arrays and photovoltaic panel monitoring equipment according to regional characteristics ensures the relevance and accuracy of data collection. Simultaneously, based on a preset acquisition frequency, real-time monitoring of environmental and power generation data can be achieved, providing comprehensive support for subsequent fault detection and optimized management. This significantly improves monitoring efficiency, reduces invalid data collection, thereby optimizing resource allocation and fault early warning, and enhancing overall operational stability and economic benefits.

[0075] Example 3:

[0076] This invention provides a remote monitoring system for fishery-solar hybrid systems for mobile terminals, including an information conversion module comprising:

[0077] Knowledge graph construction unit: Based on a preset knowledge graph algorithm, a knowledge graph in the field of fishery-solar complementary industries is constructed, and the collected environmental data and power generation data are regarded as nodes in the knowledge graph;

[0078] Data Analysis Unit: Performs initial analysis on environmental and power generation data to determine the position and weight of each node in the knowledge graph;

[0079] Data processing unit: Normalizes environmental and power generation data based on graph neural networks;

[0080] Data conversion unit: Based on a preset adversarial network algorithm, it performs information encoding conversion on the normalized environmental data and power generation data, and then obtains the encoded data.

[0081] In this embodiment, the preset graph algorithm refers to an algorithm predefined during the system design phase for constructing and processing a knowledge graph in the field of fishery-solar hybridization. This algorithm can convert information such as environmental data and power generation data into nodes in the graph, and determine the relationships and structure between these nodes. The graph algorithm helps transform data into a more systematic and structured form, facilitating subsequent data analysis and processing. For example, suppose sensors collect temperature, humidity, and power generation data in a fishery-solar hybridization area. The preset graph algorithm can convert this data into nodes in the knowledge graph, such as "temperature nodes," "humidity nodes," and "power generation data nodes," and represent the relationships between them through edges in the graph (e.g., the relationship between temperature changes and power generation efficiency).

[0082] In this embodiment, the initial analysis refers to performing preliminary statistical analysis on the collected environmental and power generation data to determine the position and weight of each data point (node) in the knowledge graph. This step helps the system understand the importance of each node and provides a foundation for subsequent data processing and graph construction. For example, the system first analyzes the collected environmental data and finds that temperature data has a greater impact on power generation efficiency, while humidity data has a smaller impact. Through the initial analysis, the system may assign a higher weight to the "temperature node" and a lower weight to the "humidity node." In this way, the temperature node in the knowledge graph will occupy a more important position in subsequent processing.

[0083] In this embodiment, the encoded data refers to the result obtained by encoding the data after it has been normalized by a graph neural network and then encoded using an adversarial network algorithm. This transformed data is represented in a standardized and simplified form, facilitating further analysis, prediction, or decision-making by the system. Encoding transformation typically aims to improve data processing efficiency and ensure that the data can be effectively utilized by subsequent modules of the system. For example, in a system, after normalization, the numerical ranges of the original environmental data (such as temperature and humidity) and power generation data may become more consistent. Using an adversarial network algorithm, the system converts this normalized data into an encoded form, possibly representing the state of each data point as a numerical sequence. This helps improve the accuracy of subsequent pattern recognition and fault prediction.

[0084] The beneficial effects of the above technical solution are as follows: The information conversion module enables efficient processing and intelligent analysis of environmental and power generation data. Compared with existing technologies, the graph construction unit transforms data into knowledge graph nodes, and the data analysis unit precisely determines the weight and position of each node, enhancing the structured processing capability of the data. Combined with the normalization processing of graph neural networks and the encoding conversion of adversarial network algorithms, the accuracy and efficiency of data analysis can be improved, fault detection and early warning functions can be optimized, the system's intelligence level can be increased, manual intervention can be reduced, potential faults can be identified more quickly, and the stable operation of the fishery-solar hybrid system can be ensured.

[0085] Example 4:

[0086] This invention provides a remote monitoring system for solar-aquaculture hybrid systems on mobile devices, including a mode determination module comprising:

[0087] Feature extraction unit: Extracts features from encoded and converted data based on a preset algorithm;

[0088] Feature analysis unit: Performs feature analysis on the extracted features to determine several feature parameters;

[0089] Mode determination unit: Determines preliminary fault coefficients based on feature parameters, and determines the corresponding preliminary fault mode by combining them with a preset coefficient-mode database.

[0090] In this embodiment, the preset algorithm is the Attention-BiLSTM algorithm.

[0091] In this embodiment, feature extraction of the encoded and converted data based on a preset algorithm includes: environmental data and power generation data are mostly time series data, which Attention-BiLSTM can effectively handle. For environmental data, such as light intensity, temperature, and water quality parameters, they are organized into a time series data input into the model. The forward and backward networks of BiLSTM learn the time series from the forward and backward directions, respectively, capturing the feature dependencies at different time steps. For example, when processing light intensity data, it can not only learn the relationship between the current light intensity and previous times, but also consider the impact of subsequent times. The attention mechanism further focuses on the features of key time steps, assigning different weights to the features of different time steps, highlighting features that are more important for fault judgment, such as the sharp change in light intensity during a certain key period. For power generation data, such as the voltage, current, and power of photovoltaic panels, it is also input into Attention-BiLSTM in time series form. Since power generation data is closely related to the performance of photovoltaic panels, the model can learn features related to potential faults, such as abnormal power fluctuations and sudden changes in voltage and current. For example, when a photovoltaic panel is partially shaded, its power output will drop abnormally. Attention-BiLSTM can capture this feature and assign it a high weight.

[0092] The beneficial effects of the above technical solution are as follows: The pattern determination module improves the accuracy and efficiency of fault detection. Compared with existing technologies, the feature extraction unit extracts key features from the encoded data, and the feature analysis unit performs in-depth analysis of these features, ensuring accurate identification of potential fault signs. By combining feature parameters and a preset coefficient-pattern database, preliminary fault modes can be quickly determined, providing accurate basis for fault early warning and handling, improving the intelligence level of fault detection, reducing false alarm rates, optimizing the maintenance and management efficiency of the solar-aquaculture hybrid system, and ensuring the long-term stable operation of the system.

[0093] Example 5:

[0094] This invention provides a remote monitoring system for solar-aquaculture hybrid systems on mobile devices, which extracts features including environmental features and power generation features.

[0095] In this embodiment, environmental characteristics refer to data features related to the environment of the solar-fishery complementary area, such as temperature, humidity, light intensity, and wind speed. These characteristics reflect the impact of environmental factors on the performance of the photovoltaic power generation system.

[0096] In this embodiment, power generation characteristics refer to key data in the photovoltaic panel's power generation process, such as power generation, power output, and power generation efficiency. These characteristics reflect the photovoltaic panel's operating status and output performance.

[0097] The beneficial effects of the above technical solution are as follows: By extracting environmental and power generation characteristics, the system's data analysis capabilities and fault detection accuracy are significantly improved. Compared with existing technologies, this system comprehensively evaluates the operating status of the fishery-solar hybrid system by extracting environmental data (such as temperature, humidity, and light intensity) and power generation data (such as power generation, power, and efficiency) separately. This feature extraction method can more accurately reflect the impact of the environment on power generation performance, promptly detect potential system faults or efficiency losses, and, with the help of these characteristics, enable the system to perform intelligent analysis and early warning, improve the stability and efficiency of the fishery-solar hybrid system, and reduce the cost of manual monitoring.

[0098] Example 6:

[0099] This invention provides a remote monitoring system for fishery-solar hybrid systems for mobile terminals, including a feature analysis unit comprising:

[0100] Deviation Acquisition Subunit: By analyzing the light intensity characteristics in the environment, the long-term trend line of light intensity is fitted using the least squares method to determine the deviation value between the light intensity and the trend line at each time point;

[0101] Parameter determination sub-unit: Perform statistical analysis on the deviation values ​​to determine the trend deviation parameters;

[0102] Derivative Determination Sub-unit: For the current characteristic in the power generation characteristics, define a time window and determine the first derivative of the current within the window;

[0103] The mutation determination sub-unit is determined by analyzing the changes in the first derivative, identifying the mutation point of the current, and then determining the sum of the absolute values ​​of the derivatives at the mutation point as the mutation degree parameter.

[0104] Feature joint sub-unit: Construct a joint feature space for environmental features and power generation features, use the mutual information algorithm to determine the mutual information value of light intensity features and power generation features in the joint feature space, and then use the mutual information value as the coupling feature parameter.

[0105] In this embodiment, statistical analysis is performed on the deviation values ​​to determine the trend deviation parameter. For example, the median of the deviation values ​​is calculated; the larger the median, the more severe the deviation of the light intensity from the normal trend. The trend deviation parameter here is measured in lux (lx) and is used to measure the degree of deviation of the light intensity from the normal trend.

[0106] In this embodiment, the point of abrupt change in current is determined, and the sum of the absolute values ​​of the derivatives at the point of abrupt change is determined as the degree of change parameter: the larger the sum of the absolute values, the more drastic the current change, which may indicate a potential fault. The unit of the degree of change parameter is amperes per second (A / s), reflecting the drastic degree of current change.

[0107] In this embodiment, the coupling characteristic parameter, mutual information value, reflects the degree of information sharing between the two. An abnormally low mutual information value may indicate a system malfunction, affecting the normal correlation between light intensity and power generation. The coupling characteristic parameter is a dimensionless value, and its magnitude reflects the tightness of the correlation between light intensity and power generation.

[0108] The beneficial effects of the above technical solution are as follows: The accuracy of fault diagnosis and performance evaluation is optimized through feature analysis units. Compared with existing technologies, a multi-level analysis method is adopted, and trend deviation parameters are determined through statistical analysis of deviation values, further revealing the stability of system operation. Combining the analysis of the first derivative of current and abrupt change points, sudden problems in the power generation system can be identified. Finally, a joint feature space of environmental and power generation characteristics is constructed using a mutual information algorithm to comprehensively evaluate the coupling effect of the system, which helps to more accurately predict system performance changes and fault risks, and improves the intelligent monitoring capabilities of the solar-aquaculture hybrid system.

[0109] Example 7:

[0110] This invention provides a remote monitoring system for fishery-solar hybrid systems for mobile terminals, including a mode determination unit comprising:

[0111] Coefficient Determination Subunit: Determining Preliminary Fault Coefficients Based on Characteristic Parameters

[0112]

[0113] in, This is the preliminary failure coefficient. For trend deviation parameters, This is a parameter representing the degree of mutation. For coupling characteristic parameters, For symbolic functions, , , These are preset hyperparameters. These are preset adjustment parameters;

[0114] Pattern determination subunit: Based on the preliminary fault coefficients and the preset coefficient-pattern database, the corresponding preliminary fault mode is determined.

[0115] In this embodiment, the influence of the trend deviation parameter and the abrupt change parameter on the fault coefficient is enhanced by performing exponentiation and square root operations on them. A sign function is used to determine the relative magnitude of the three parameters, further influencing the sign and magnitude of the fault coefficient. The cube root operation on the coupling characteristic parameter and its placement in the denominator highlights its suppressive effect on the fault coefficient; when the coupling characteristic parameter is abnormally small, the fault coefficient increases. The subsequent fractions... The calculation of the fault coefficient is further refined by comprehensively considering the product and sum of the trend deviation parameter and the abrupt change parameter. Through these complex calculations, when the trend of solar intensity deviates significantly, the power generation data changes drastically, and the coupling characteristics of the two are abnormal, the preliminary fault coefficient will increase significantly, indicating an increased probability of system failure.

[0116] In this embodiment, the three hyperparameters are used to balance the influence of different parameters on the failure coefficient;

[0117] In this embodiment, the preset adjustment parameter is a very small positive number to avoid the case where the denominator is zero.

[0118] The beneficial effects of the above technical solution are as follows: The pattern determination unit improves the accuracy and response speed of fault identification. The coefficient determination subunit transforms multiple characteristic parameters (such as trend deviation parameters, abrupt change parameters, and coupling characteristic parameters) into preliminary fault coefficients, and, combined with preset adjustment parameters and hyperparameters, flexibly adjusts the system's fault identification sensitivity. Based on these preliminary fault coefficients, the pattern determination subunit can accurately search and match the preset coefficient-pattern database, quickly determining the corresponding fault mode. This not only enhances the accuracy of fault early warning but also enables real-time, intelligent fault detection, improving the stability and automated management capabilities of the solar-aquaculture hybrid system.

[0119] Example 8:

[0120] This invention provides a remote monitoring system for fishery-solar hybrid systems on mobile devices, including a rule construction module comprising:

[0121] The initial failure modes and their corresponding environmental and power generation data are stored in the failure database. At the same time, historical failure data is acquired, organized, and classified.

[0122] Based on a preset data processing algorithm, all data in the fault database is deeply mined and correlation analysis is performed.

[0123] Based on the results of data correlation analysis and combined with pre-defined domain knowledge, early warning rules based on data features and fault correlations are constructed.

[0124] In this embodiment, sorting and classification are indexed according to dimensions such as fault type, fault cause, and fault occurrence frequency, which facilitates subsequent querying and analysis.

[0125] In this embodiment, correlation analysis is used to analyze the correlation between different failure modes and environmental factors (such as light intensity, temperature, and humidity) and equipment operating parameters (such as power generation, current, and voltage) to identify potential patterns and influencing factors of failure occurrence.

[0126] In this embodiment, early warning rules, for example, trigger a photovoltaic module fault warning when the light intensity drops sharply in a short period of time and the photovoltaic module temperature rises abnormally; and trigger an aquaculture risk warning when the dissolved oxygen content in the water is below a certain threshold and the ammonia nitrogen content is above another threshold. These early warning rules are stored in an early warning rule base in the form of logical expressions or mathematical models.

[0127] The beneficial effects of the above technical solution are as follows: The rule-building module improves the accuracy of fault prediction and the intelligence of system response. Compared with existing technologies, this system ensures the comprehensiveness and timeliness of data by storing preliminary fault modes, corresponding environmental data, and power generation data in a fault database, and then organizing and classifying them in conjunction with historical fault data. Through pre-set data processing algorithms, the system performs in-depth mining and correlation analysis on the data in the fault database, enabling it to discover potential fault correlations and generate accurate early warning rules. These early warning rules, based on data characteristics and fault correlations, can help the system identify potential risks in advance, improve fault prevention capabilities, reduce maintenance costs, and enhance the intelligent monitoring level and operational efficiency of the solar-aquaculture hybrid system.

[0128] Example 9:

[0129] This invention provides a remote monitoring method for solar-fishery complementary systems on a mobile device, such as... Figure 2 As shown, it includes:

[0130] Step 1: Deploy a sensor array in the solar-fishery complementary area to collect environmental data, and at the same time, collect power generation data based on the pre-set photovoltaic panel monitoring equipment;

[0131] Step 2: Preprocess the collected environmental and power generation data and perform information encoding conversion;

[0132] Step 3: Extract features from the encoded and converted data based on a preset algorithm to determine the initial fault mode;

[0133] Step 4: Build a fault database based on the data corresponding to the initial fault modes, and then build early warning rules by combining historical data;

[0134] Step 5: Determine early warning information based on the collected environmental data, power generation data, and early warning rules, and visualize the early warning information.

[0135] The beneficial effects of the above technical solution are as follows: it realizes comprehensive monitoring of the solar-fishery complementary area; compared with the existing technology, it can collect environmental and power generation data in real time, and automatically analyze the data through intelligent algorithms to identify potential fault modes. By constructing a fault database and combining historical data with early warning rules, the system can provide timely warnings before faults occur, which greatly improves the operational reliability of the photovoltaic system. In addition, the system supports visual output, which facilitates users to respond quickly and make decisions, optimizes energy management efficiency, and reduces manual intervention and potential losses.

[0136] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A remote monitoring system for fishery-solar hybrid systems for mobile terminals, characterized in that, include: Data acquisition module: Deploy sensor arrays in the fishery-solar hybrid area to collect environmental data, and at the same time, collect power generation data based on the pre-set photovoltaic panel monitoring equipment; Information conversion module: preprocesses the collected environmental and power generation data and performs information encoding conversion; Pattern determination module: Based on a preset algorithm, features are extracted from the encoded and converted data to determine the preliminary fault mode; The pattern determination module includes: a feature extraction unit: performing feature extraction on the encoded conversion data based on a preset algorithm; Feature analysis unit: Performs feature analysis on the extracted features to determine several feature parameters; The feature analysis unit includes: Deviation Acquisition Subunit: By analyzing the light intensity characteristics in the environment, the long-term trend line of light intensity is fitted using the least squares method to determine the deviation value between the light intensity and the trend line at each time point; Parameter determination sub-unit: Perform statistical analysis on the deviation values ​​to determine the trend deviation parameters; Derivative Determination Sub-unit: For the current characteristic in the power generation characteristics, define a time window and determine the first derivative of the current within the window; The mutation determination sub-unit is determined by analyzing the changes in the first derivative, identifying the mutation point of the current, and then determining the sum of the absolute values ​​of the derivatives at the mutation point as the mutation degree parameter. Feature joint sub-unit: Construct a joint feature space of environmental features and power generation features, use the mutual information algorithm to determine the mutual information value of light intensity features and power generation features in the joint feature space, and then use the mutual information value as the coupling feature parameter; Mode determination unit: Determines preliminary fault coefficients based on feature parameters, and determines the corresponding preliminary fault mode by combining them with a preset coefficient-mode database; The mode determination unit includes: Coefficient Determination Subunit: Determining Preliminary Fault Coefficients Based on Characteristic Parameters in, This is the preliminary failure coefficient. For trend deviation parameters, For mutation degree parameters, For coupling characteristic parameters, For symbolic functions, , , These are preset hyperparameters. These are preset adjustment parameters; Pattern determination subunit: Determines the corresponding preliminary fault mode based on the preliminary fault coefficients and the preset coefficient-pattern database; Rule building module: Builds a fault database based on the data corresponding to the initial fault modes, and then builds early warning rules by combining historical data; Information output module: Based on the collected environmental data, power generation data and early warning rules, determine the early warning information and visualize the early warning information.

2. The remote monitoring system for fishery-solar hybrid systems for mobile terminals according to claim 1, characterized in that, The data acquisition module includes: Regional Analysis Unit: The fishery-solar complementary area includes several pre-defined areas. Regional analysis is performed on each area to determine the regional characteristics of each area. Environmental data acquisition unit: Deploy corresponding sensor arrays based on the regional characteristics of each area, and collect environmental data for each area based on the corresponding sensor arrays and the preset acquisition frequency; Power generation data acquisition unit: Collects power generation data based on preset photovoltaic panel monitoring equipment and preset acquisition frequency.

3. The remote monitoring system for fishery-solar hybrid systems for mobile terminals according to claim 1, characterized in that, The information conversion module includes: Knowledge graph construction unit: Based on a preset knowledge graph algorithm, a knowledge graph in the field of fishery-solar complementary industries is constructed, and the collected environmental data and power generation data are regarded as nodes in the knowledge graph; Data Analysis Unit: Performs initial analysis on environmental and power generation data to determine the position and weight of each node in the knowledge graph; Data processing unit: Normalizes environmental and power generation data based on graph neural networks; Data conversion unit: Based on a preset adversarial network algorithm, it performs information encoding conversion on the normalized environmental data and power generation data, and then obtains the encoded data.

4. A remote monitoring system for fishery-solar hybrid systems for mobile terminals according to claim 1, characterized in that, The extracted features include environmental features and power generation features.

5. A remote monitoring system for fishery-solar hybrid systems for mobile terminals according to claim 1, characterized in that, The rule building module includes: The initial failure modes and their corresponding environmental and power generation data are stored in the failure database. At the same time, historical failure data is acquired, organized, and classified. Based on a preset data processing algorithm, all data in the fault database is deeply mined and correlation analysis is performed. Based on the results of data correlation analysis and combined with pre-defined domain knowledge, early warning rules based on data features and fault correlations are constructed.

6. A method for remote monitoring of fishery-solar hybrid systems on a mobile device, applied to the remote monitoring system for fishery-solar hybrid systems on a mobile device as described in claim 1, characterized in that, include: Step 1: Deploy a sensor array in the solar-fishery complementary area to collect environmental data, and at the same time, collect power generation data based on the pre-set photovoltaic panel monitoring equipment; Step 2: Preprocess the collected environmental and power generation data and perform information encoding conversion; Step 3: Extract features from the encoded and converted data based on a preset algorithm to determine the initial fault mode; Step 4: Build a fault database based on the data corresponding to the initial fault modes, and then build early warning rules by combining historical data; Step 5: Determine early warning information based on the collected environmental data, power generation data, and early warning rules, and visualize the early warning information.

Images