Water quality analysis method and system in water intake project combined with multi-source data fusion

Abstract

The application provides a water quality analysis method and system for a receiving lake in a diversion project combined with multi-source data fusion, and relates to the technical field of water resource management and analysis. Firstly, field basic data is collected and a water quality correlation field is constructed. Then, interaction relationship modeling operation is performed on the field units to obtain a field unit interaction model. Based on the field unit interaction model, dynamic coupling fusion processing is performed on the field basic data to generate coupled water quality data. Then, the water quality correlation field change trend data and the coupled water quality data are combined to deduce the water quality trend evolution path of the receiving lake. Finally, the core interaction unit is identified, and a dynamic regulation and control scheme for the water quality of the diversion project is generated in combination with the trend evolution path. The application can comprehensively analyze the water quality of the receiving lake, realize accurate dynamic regulation and control, and ensure the stability of the water quality.

Description

Technical Field

[0001] This invention relates to the field of water resource management and analysis technology, and more specifically, to a method and system for analyzing the water quality of receiving lakes in water diversion projects that combines multi-source data fusion. Background Technology

[0002] In water diversion projects, the water quality of the receiving lakes directly affects the surrounding ecological environment, residents' drinking water safety, and the overall effectiveness of the project. Currently, the analysis of water quality in receiving lakes in water diversion projects mainly relies on single data sources or simple overlay of multi-source data. Single-data source analysis methods, such as water quality assessment based solely on in-situ sensing data of the receiving lakes, fail to comprehensively reflect the complex causes of water quality changes due to a lack of consideration for information from the water source and the water conveyance process. While the method of simply overlaying multi-source data integrates data from different stages, it does not delve into the inherent dynamic relationships between the data and cannot accurately present the field effects of water quality impact. For example, when faced with complex situations such as changes in water diversion flow or sudden pollution of the surrounding environment, existing methods struggle to accurately capture water quality trends and cannot promptly formulate effective water quality control plans, leading to significant fluctuations in the water quality of the receiving lakes and potentially causing serious problems such as water quality deterioration. Summary of the Invention

[0003] In view of the aforementioned problems, and in conjunction with the first aspect of the present invention, the present invention provides a method for water quality analysis of receiving lakes in water diversion projects that combines multi-source data fusion, the method comprising:

[0004] Collect basic field data, define the dynamic interaction relationship between the data in the basic field data, and construct a water quality correlation field between the water diversion project and the receiving lake. The water quality correlation field presents the field effect of water quality influence through the dynamic interaction relationship between the data. The basic field data includes dynamic output data of the water diversion source, real-time interactive data of the water conveyance process, in-situ sensing data of the receiving lake, and dynamic correlation data of the surrounding environment.

[0005] An interaction relationship modeling operation is performed on each field unit in the water quality correlation field to obtain a field unit interaction model. The field unit interaction model is used to quantitatively describe the transmission mode, transmission intensity and dynamic change characteristics of water quality influence between different field units.

[0006] Based on the interaction relationship described by the field unit interaction model, dynamic coupling and fusion processing is performed on the field basic data of the water quality correlation field to generate real-time updated coupled water quality data. When the interaction relationship changes, the dynamic coupling and fusion processing performs an adjustment operation of the fusion strategy so that the fusion strategy follows the changes in the field interaction.

[0007] By combining the field change trend data of the water quality correlation field with the coupled water quality data, the deduction operation of the water quality trend evolution path of the water-receiving lake is performed to generate a trend evolution path containing the continuous change characteristics of water quality status and key turning point information.

[0008] Identify core interaction units from the water quality correlation field, combine the trend evolution path with the core interaction units, and generate a dynamic water quality control scheme for the water diversion project. The dynamic water quality control scheme for the water diversion project includes control instructions that act on the core field interaction nodes that affect the water quality trend.

[0009] Furthermore, this invention also provides a water quality analysis system for receiving lakes in water diversion projects that combines multi-source data fusion, comprising:

[0010] A processor; a machine-readable storage medium for storing machine-executable instructions of the processor; wherein the processor is configured to execute the above-described method for analyzing the water quality of a receiving lake in a water diversion project by means of multi-source data fusion via executing the machine-executable instructions.

[0011] In another aspect, the present invention also provides a computer program product, the computer program product including machine-executable instructions stored in a computer-readable storage medium, wherein a processor of a water quality analysis system for a water-receiving lake in a water diversion project incorporating multi-source data fusion reads the machine-executable instructions from the computer-readable storage medium, and the processor executes the machine-executable instructions, causing the water quality analysis system for a water-receiving lake in a water diversion project incorporating multi-source data fusion to perform the aforementioned water quality analysis method for a water-receiving lake in a water diversion project incorporating multi-source data fusion.

[0012] Based on the above, by collecting basic field data including dynamic output data from the water source, real-time interactive data during the water conveyance process, in-situ sensing data of the receiving lake, and dynamic correlation data of the surrounding environment, and defining the dynamic interaction relationships between the data to construct a water quality correlation field, the field effect of water quality impact is presented. Then, interaction relationship modeling is performed on each field unit in the water quality correlation field. The resulting field unit interaction model can quantitatively describe the transmission mode, intensity, and dynamic change characteristics of water quality impact between different field units. Based on this field unit interaction model, the basic field data is dynamically coupled and fused to generate real-time updated coupled water quality data. The fusion strategy can be adjusted according to the changes in interaction relationships. By combining the water quality correlation field change trend data with the coupled water quality data, the evolution path of water quality trend in the receiving lake can be deduced. A path containing continuous change characteristics of water quality status and key turning point information can be generated. Core interaction units are identified from the water quality correlation field and combined with the trend evolution path to generate a dynamic water quality control scheme for the water diversion project, including control instructions acting on the core field interaction nodes. This enables precise dynamic control of water quality and effectively ensures the stability of water quality in the receiving lake. Attached Figure Description

[0013] Figure 1 This is a schematic diagram of the execution flow of the water quality analysis method for water-receiving lakes in water diversion projects that combines multi-source data fusion, provided in an embodiment of the present invention.

[0014] Figure 2 This is a schematic diagram of exemplary hardware and software components of a water quality analysis system for a water diversion project that combines multi-source data fusion, provided in an embodiment of the present invention. Detailed Implementation

[0015] The present invention will now be described in detail with reference to the accompanying drawings. Figure 1 This is a flowchart illustrating a method for analyzing the water quality of receiving lakes in a water diversion project that combines multi-source data fusion, according to an embodiment of the present invention. The following is a detailed description of this method for analyzing the water quality of receiving lakes in a water diversion project that combines multi-source data fusion.

[0016] Step S110: Collect basic field data, define the dynamic interaction relationship between the data in the basic field data based on the basic field data, and construct the water quality correlation field between the water diversion project and the receiving lake. The water quality correlation field presents the field effect of water quality influence through the dynamic interaction relationship between the data. The basic field data includes dynamic output data of the water diversion source, real-time interactive data of the water conveyance process, in-situ sensing data of the receiving lake, and dynamic correlation data of the surrounding environment.

[0017] In this embodiment, a water diversion project is used as an example to introduce water into a receiving lake. This project aims to improve the water quality of the receiving lake by introducing an external water source. To achieve a comprehensive analysis of the water quality of the receiving lake, it is necessary to first collect various types of basic field data. This basic field data covers relevant information from the entire process from the water source to the receiving lake, as well as the surrounding environment. After collecting sufficient data, the intrinsic relationships between the basic field data are analyzed in depth to determine how they interact and affect water quality, thereby constructing a water quality correlation field that reflects the aforementioned field effects.

[0018] Step S111: Collect dynamic output data from the water source at fixed time intervals and record the collection time information for each collection. The dynamic output data from the water source includes water quality composition data, flow rate change data, and output status data during the water supply process.

[0019] In this scenario, the water source is a river, and a data acquisition station is established at the river's intake point. Dynamic output data from the water source is collected at fixed time intervals, with each acquisition precisely recording the time information, which serves as the time reference for subsequent data processing and analysis. Water quality composition data during the water supply process is collected using various water quality sensors installed at the intake point. These sensors monitor in real-time dissolved oxygen, pH, ammonia nitrogen, total phosphorus, total nitrogen, and other water quality-related indicators. Flow rate change data during the water supply process is obtained using a flow monitoring device installed on the intake pipeline, which continuously tracks the water flow rate per unit time. Output status data during the water supply process includes the operating parameters of the water pump, such as pump speed, inlet and outlet pressure, current, and voltage, as well as information on the valve opening controlling the water flow rate. This data is collected through interfaces connected to the pump and valve control systems.

[0020] Step S112: Collect real-time interactive data of the water conveyance process in segments according to the water conveyance process, and mark the collection location information of each collection. The real-time interactive data of the water conveyance process includes the interaction data between the water conveyance pipeline and the water body, the water quality exchange data along the water conveyance route, and the associated data of the operation status of the water conveyance facilities.

[0021] The water transport process begins at the water intake point, passing through a series of pipelines, booster pump stations, regulating reservoirs, and other facilities, ultimately delivering water to the receiving lake. Based on the characteristics of the water transport process, the entire route is divided into several segments. For example, the first segment runs from the water intake point to the first booster pump station; the second segment runs from the first booster pump station to the regulating reservoir; and the third segment runs from the regulating reservoir to the inlet of the receiving lake. Data collection points are set up at key locations within each segment, such as the pipeline's starting point, intermediate points, and connections with other facilities. Each collection point is assigned a unique location identifier, which is stored along with the collected data to clearly identify the data's source. The interaction data between the pipeline and the water body includes data on the adsorption of certain substances from the water by the pipeline's inner wall, the release of substances from the pipeline material into the water, and the impact of pipeline corrosion on the water body. This data is collected using special sensors installed on the inner wall of the pipeline. Water quality exchange data along the water conveyance route is obtained by analyzing water samples collected at various collection points. This data primarily includes changes in the concentration of various pollutants and dissolved oxygen content at different locations. The operational status data of the water conveyance facilities covers the operating efficiency and energy consumption of pumps in the booster pumping stations, changes in water level in the regulating pools, and the working status of various valves. This data is acquired in real-time from the facility's control system.

[0022] Step S113: Collect in-situ sensing data of the receiving lake synchronously at multiple points through distributed sensing devices. The in-situ sensing data of the receiving lake includes water quality parameter data of different areas of the lake, water flow data of different areas of the lake, and lake ecological correlation data of different areas of the lake.

[0023] The lake has a large area, so distributed sensing devices were deployed in different areas to comprehensively understand its water quality. These devices are located in the nearshore area, the central area, the inlet area, and areas with special ecological significance. All devices use a unified time synchronization mechanism to collect data simultaneously from multiple points. Water quality parameters in different areas of the lake are collected by water quality sensors, including water temperature, transparency, turbidity, and concentrations of various nutrients. These sensors can stably monitor the water quality in their respective areas over a long period. Water flow data in different areas of the lake are obtained using devices such as flow velocity and direction meters, which can measure the speed and direction of water flow to understand the water movement patterns within the lake. Lake ecological correlation data in different areas include the growth status of aquatic plants, the types and quantities of plankton, and the distribution of benthic organisms. This data is collected through periodic sampling combined with image recognition technology. The growth status of aquatic plants can be obtained by analyzing remote sensing images or images taken by underwater cameras, while plankton and benthic organisms are obtained through laboratory analysis of water and sediment samples.

[0024] Step S114: Collect dynamic correlation data of the surrounding environment according to the type of environmental impact. The dynamic correlation data of the surrounding environment includes atmospheric environmental data, surface runoff data, vegetation cover data and human activity correlation data.

[0025] The surrounding environment of a lake has a significant impact on its water quality; therefore, it is necessary to collect dynamic correlation data of the surrounding environment according to the type of environmental impact. Atmospheric environmental data is collected by setting up meteorological stations around the lake, including data on temperature, air pressure, relative humidity, wind direction, wind speed, rainfall, solar radiation intensity, and the concentration of various pollutants in the atmosphere. This data reflects the direct and indirect impacts of the atmospheric environment on lake water quality. Surface runoff data is collected by setting up monitoring points at various surface runoff inlets flowing into the lake, including runoff flow rate, velocity, and the types and concentrations of pollutants carried in the runoff. This data helps to understand the contribution of surface runoff to lake water quality. Vegetation cover data is collected through a combination of satellite remote sensing technology and ground surveys. Satellite remote sensing can obtain information on vegetation cover over a large area, while ground surveys can provide detailed information on vegetation types and growth status in specific areas. Vegetation cover affects soil erosion and nutrient input, thus affecting lake water quality. Human activity-related data includes the discharge and treatment of domestic sewage from surrounding residents, wastewater discharge from surrounding industrial enterprises, the use of chemical fertilizers and pesticides in agricultural production, and the impact of tourism activities on the surrounding environment of the lake. The above data are collected through cooperation with relevant departments, on-site investigations, and monitoring of pollution sources.

[0026] Step S115: Classify and map the dynamic output data of the water source, the real-time interactive data of the water conveyance process, the in-situ sensing data of the receiving lake, and the dynamic correlation data of the surrounding environment according to the data dimensions, establish the correspondence between the data dimensions and the water quality influencing factors, and form a field basic data dimension mapping table.

[0027] First, a detailed dimensional analysis was conducted on the collected dynamic output data from the water source, real-time interactive data during the water conveyance process, in-situ sensing data from the receiving lake, and dynamic correlation data of the surrounding environment. The dimensions of the dynamic output data from the water source include water quality composition (specific water quality indicators such as dissolved oxygen and pH), flow rate (water volume per unit time), and output status (such as pump operating parameters and valve opening). The dimensions of the real-time interactive data during the water conveyance process include pipeline interaction (interaction indicators between the pipeline and the water body), water quality exchange (water quality change indicators along the water conveyance route), and facility operating status (operating parameters of the water conveyance facilities). The dimensions of the in-situ sensing data from the receiving lake include water quality parameters (water quality indicators in different areas of the lake), water flow (flow velocity, direction, etc.), and ecological correlation (indicators related to aquatic organisms). The dimensions of the dynamic correlation data of the surrounding environment include atmospheric environment (meteorological indicators, air pollutant concentrations, etc.), surface runoff (runoff flow rate, pollutant concentrations, etc.), vegetation cover (vegetation type, coverage, etc.), and human activity (indicators related to various human activities). Water quality influencing factors refer to various factors that can affect water quality, such as nutrient content, dissolved oxygen level, water flow conditions, and pollutant input. Then, based on the practical significance of each data dimension and their intrinsic relationship with water quality, a correspondence is established between different data dimensions and their corresponding water quality influencing factors. For example, in the dynamic output data from the water source, indicators such as ammonia nitrogen, total phosphorus, and total nitrogen in the water composition dimension correspond to nutrient content in the water quality influencing factor; in the real-time interactive data of the water transfer process, dissolved oxygen changes in the water exchange dimension correspond to dissolved oxygen level in the water quality influencing factor; and in the dynamic correlation data of the surrounding environment, pollutant concentration in the surface runoff dimension corresponds to pollutant input in the water quality influencing factor. After organizing these correspondences, a field-based data dimension mapping table is formed, which shows the relationship between each data dimension and the water quality influencing factors.

[0028] Step S116: Based on the field basic data dimension mapping table, perform standardized preprocessing on data from different sources mapped to the same water quality influencing factor to generate standardized field basic data; based on the standardized field basic data, calculate the water quality influence correlation strength between data from different sources to form a data correlation strength comparison table, wherein the correlation strength is calculated based on the degree of contribution of standardized data to the same water quality influencing factor.

[0029] Step S1161: Extract all water quality influencing factors from the field basic data dimension mapping table to form a water quality influencing factor set, where each water quality influencing factor corresponds to at least two data dimensions from different sources.

[0030] From the established field-based data dimension mapping table, all water quality influencing factors are extracted one by one, ensuring no duplication or omission. These extracted factors are then combined to form a water quality influencing factor set. During the extraction process, it is necessary to check whether each water quality influencing factor corresponds to at least two data dimensions from different sources. This is to ensure that the impact of the factor on water quality can be considered from multiple perspectives in subsequent analysis, avoiding potential biases from a single data source. For example, the "nutrient content" in the water quality influencing factor should correspond to at least two different data dimensions: the water quality composition dimension in the dynamic output data of the water source and the surface runoff dimension in the dynamic correlation data of the surrounding environment.

[0031] Step S1162: For each water quality influencing factor in the set of water quality influencing factors, extract the original data associated with the corresponding data dimension, and perform standardized preprocessing on the original data to form a standardized dedicated data set for that water quality influencing factor. The standardized dedicated data set contains relevant standardized data records from different sources.

[0032] For each water quality influencing factor in the set of water quality influencing factors, the corresponding data dimension is found according to the correspondence between the factor and the data dimension in the field basic data dimension mapping table. Then, the original data associated with the above data dimension is extracted from various data sources. For example, for the water quality influencing factor "dissolved oxygen level", the original data such as dissolved oxygen index data in the water quality composition dimension of the dynamic output data of the water source and dissolved oxygen change data in the water quality exchange dimension of the real-time interactive data of the water transfer process are found according to the mapping table. After extraction, the original data from different data dimensions are preprocessed for standardization. An appropriate standardization method is selected according to the type and characteristics of the original data, such as the Z-score standardization or min-max standardization mentioned above. After standardization, the obtained standardized data is classified according to the water quality influencing factors, forming a standardized data set exclusive to each water quality influencing factor. This standardized data set contains standardized data records related to the water quality influencing factor from different data sources, and each record has corresponding time and location information.

[0033] Step S1163: Analyze the sensitivity of each standardized data record in the standardized dedicated dataset to the characterization of water quality influencing factors. The characterization sensitivity is determined based on the response relationship between the amount of change in the standardized data record and the amount of change in the standardized value of the water quality influencing factor.

[0034] For each standardized dataset of water quality influencing factors, each standardized data record is analyzed to determine its sensitivity in representing the water quality influencing factor. Sensitivity reflects the extent to which changes in the standardized data record reflect changes in the water quality influencing factor. Specifically, sensitivity is determined by analyzing the response relationship between the amount of change in the standardized data record and the amount of change in the standardized value of the water quality influencing factor. If a small change in the standardized data record causes a large change in the standardized value of the water quality influencing factor, the data record has high sensitivity; conversely, if a large change in the standardized data record results in a small change in the standardized value of the water quality influencing factor, the sensitivity is low. During the analysis, sensitivity can be assessed by comparing the trends of change over different time periods and calculating the ratio of changes.

[0035] Step S1164: Based on the representation sensitivity of each standardized data record, prioritize all standardized data records in the standardized dedicated data set to form a data priority sequence.

[0036] After determining the characterization sensitivity of each standardized data record in the standardized dedicated dataset, all standardized data records are sorted according to their characterization sensitivity. Standardized data records with higher characterization sensitivity are more important in reflecting changes in water quality influencing factors and should therefore be given higher priority in subsequent analysis and calculations. After sorting, a data priority sequence is formed, in which the order of data records reflects their priority.

[0037] Step S1165: Based on the data priority sequence, assign a basic contribution weight to each data source. The basic contribution weight of the data source with higher priority is set to be higher than that of the data source with lower priority. At the same time, dynamic adjustment space is reserved for the basic contribution weight to adapt to data changes.

[0038] A data source refers to the original source of standardized data records, such as data from the water source or data from the water transfer process. The priority of each data source is determined based on the priority ranking of standardized data records in the data priority sequence. Generally, data sources with higher priority rankings belong to data sources with higher priority. Then, a basic contribution weight is assigned to each data source. The value of the basic contribution weight corresponds to the priority of the data source; the higher the priority, the larger the value of the basic contribution weight. Simultaneously, considering that data may change over time and under varying environmental conditions, dynamic adjustment space is reserved for the basic contribution weight. This allows for appropriate adjustments to the basic contribution weight based on actual data changes during subsequent processing, ensuring that it accurately reflects the contribution of the data source to water quality influencing factors.

[0039] Step S1166: Calculate the synergistic contribution coefficient between different data sources under the same water quality influencing factor. The synergistic contribution coefficient is determined based on the synchronicity and complementarity of standardized data changes between data sources. The synergistic contribution coefficient of data sources with prominent synchronicity and complementarity is set to be higher than that of data sources with less prominent synchronicity and complementarity.

[0040] For the same water quality influencing factor, the synchronicity and complementarity of standardized data changes across different data sources are analyzed. Synchronicity refers to the consistency in the changing trends of standardized data from different data sources. If the standardized data from two data sources rise or fall simultaneously, and the magnitudes of change are relatively similar, then their synchronicity is high. Complementarity means that standardized data from different data sources can reflect the characteristics of the water quality influencing factor from different perspectives; data from one data source can supplement the deficiencies of data from another data source in certain aspects. Based on the degree of synchronicity and complementarity of standardized data changes between data sources, a synergistic contribution coefficient is calculated. The more prominent the synchronicity and complementarity of data sources, the greater their combined contribution to the water quality influencing factor, and therefore, the higher the value of the synergistic contribution coefficient should be set; conversely, for data sources with less prominent synchronicity and complementarity, the synergistic contribution coefficient should be set lower.

[0041] Step S1167: Combine the basic contribution weight with the collaborative contribution coefficient to calculate the comprehensive contribution of each data source to the corresponding water quality impact factor.

[0042] To comprehensively measure the contribution of each data source to water quality impact factors, the previously calculated basic contribution weights and synergistic contribution coefficients need to be combined. The specific combination method can be determined based on the actual situation. For example, a multiplication operation can be used, multiplying the basic contribution weight by the synergistic contribution coefficient, and the result can be used as the comprehensive contribution of that data source to the corresponding water quality impact factor. This method considers both the basic contribution determined by the priority of the data source itself and the synergistic contribution generated by the interaction between data sources, enabling the comprehensive contribution level to more fully and accurately reflect the actual impact of the data source on water quality impact factors.

[0043] Step S1168: Construct an initial comparison table framework with water quality influencing factors as rows and data sources as columns, and fill in the comprehensive contribution of each data source to the corresponding water quality influencing factor in the corresponding position of the initial comparison table framework.

[0044] First, construct an initial lookup table framework. The rows of this framework represent the individual water quality influencing factors in the set of water quality influencing factors, and the columns represent the various data sources. Then, calculate the overall contribution of each data source to the corresponding water quality influencing factor, and fill it into the appropriate cell in the initial lookup table framework according to the correspondence between the water quality influencing factor and the data source. For example, if a data source's overall contribution to the water quality influencing factor "nutrient content" is a certain value, then that value will be filled into the cell with "nutrient content" as the row and that data source as the column.

[0045] Step S1169: Perform normalization processing on all comprehensive contribution levels in the initial comparison table framework, bind the normalized comparison table with the source identifier and data dimension identifier of the field basic data, supplement the metadata information of the normalized comparison table, and form a data association strength comparison table.

[0046] Since the overall contribution of data sources under different water quality influencing factors may have different orders of magnitude, it is necessary to normalize all overall contribution levels in the initial comparison table framework to facilitate comparison and comprehensive consideration in subsequent analyses. The normalization method can be to divide each overall contribution level by the sum of all overall contribution levels in its row (i.e., under the same water quality influencing factor), ensuring that the sum of the overall contribution levels of all data sources under the same water quality influencing factor is 1. After normalization, the resulting comparison table is bound to the source identifiers of the field basic data (used to clarify the specific information of the data source) and the data dimension identifiers of the field basic data (used to clarify the specific content of the data dimension). Simultaneously, metadata information for the comparison table is supplemented, such as the data collection time range, standardization method, and calculation method version. After the above processing, the final data association strength comparison table is formed.

[0047] Step S117: Based on the data correlation strength comparison table, perform the operation of establishing the field unit division criteria, classify data with correlation strength higher than the preset threshold into the same field unit, so that each field unit corresponds to a core water quality influencing factor and a set of related data.

[0048] A data correlation strength comparison table reflects the degree of correlation between different data sources and water quality influencing factors. Based on this table, the criteria for dividing the field units are established. First, a preset threshold for correlation strength is set. This threshold can be determined comprehensively based on factors such as actual engineering experience, data characteristics, and the accuracy requirements of water quality analysis. Then, data with correlation strengths higher than the preset threshold are grouped into the same field unit. Data within the same field unit have strong correlations, revolving around a core water quality influencing factor, which is the primary water quality factor affected by all data within that field unit. Simultaneously, each field unit corresponds to a correlation data set, which contains all relevant data belonging to that field unit. These data come from different data sources but all have a strong correlation with the core water quality influencing factor.

[0049] Step S118: Determine the spatial boundary of each field unit based on the physical location range of the data acquisition, and determine the temporal boundary of each field unit based on the time synchronization range of the data acquisition, thus forming the field unit boundary features.

[0050] For each field unit, its spatial boundary is defined based on the physical location of data collection in its associated dataset. The range of the physical location of data collection refers to the actual geographical location of the sensors or monitoring equipment collecting the data. All data collection locations belonging to this field unit are integrated to determine a minimum spatial range encompassing these locations as the spatial boundary of the field unit. Simultaneously, the temporal boundary of each field unit is defined based on the time synchronization range of data collection. The time synchronization range refers to the range within which data collection time is synchronized to ensure data consistency and comparability. The collection time ranges of all data belonging to this field unit are unified to determine a common time interval as the temporal boundary of the field unit. The spatial and temporal boundaries together constitute the boundary characteristics of the field unit, clearly defining its spatial and temporal extent.

[0051] Step S119: Spatially arrange and time-synchronize each field unit according to the correlation relationship in the data correlation strength comparison table to construct an initial water quality correlation field framework. The initial water quality correlation field framework includes the distribution of field units, the correlation channels between units, and the data flow path.

[0052] In this embodiment, the data association strength comparison table clearly defines the association strength values ​​between each field unit. These values ​​are calculated based on standardized field base data and can reflect the degree of mutual influence between field units. To construct the initial water quality association field framework, the spatial layout and temporal synchronization of field units must first be carried out according to these association relationships to ensure that the framework can accurately reflect the field effects of water quality influence.

[0053] Step S1191: Extract the correlation strength value corresponding to each field unit in the data correlation strength comparison table. The correlation strength value is calculated based on the standardized field basic data. Sort the correlation relationships between field units according to the magnitude of the correlation strength value to form a correlation relationship priority sequence.

[0054] From the data association strength comparison table, the association strength value between each field unit and other field units is extracted one by one. Since these values ​​are calculated based on standardized field data, the influence of dimensions has been eliminated, and they can be directly used for comparison. For example, if the association strength value between field unit A and field unit B is 0.8, and the association strength value between field unit A and field unit C is 0.5, then the association relationship between field units A and B takes precedence over that between A and C. The association relationships between all field units are sorted in descending order of association strength value to form an association relationship priority sequence.

[0055] Step S1192: Based on the priority sequence of association relationships, establish the spatial layout basis of field units. Field units with high association strength values ​​are preferentially arranged in an adjacent manner in space.

[0056] In the priority sequence of correlation relationships, field units ranked higher have stronger mutual influences. Therefore, they should be prioritized as adjacent units in spatial layout. For example, if the top three in the priority sequence are field units XY, YZ, and XZ, then X, Y, and Z should be placed as close as possible in the spatial layout to form a closely related unit group. This layout principle ensures the spatial proximity of field units with more direct and intense water quality impact transmission, thus more accurately reflecting the field effect.

[0057] Step S1193: Determine the actual spatial coordinates of each field unit based on the physical layout of the water diversion project, the geographical zoning of the receiving lakes, and the distribution characteristics of the surrounding environment.

[0058] The physical layout of a water diversion project includes the actual geographical locations of facilities such as the water intake, water pipelines, booster pumping stations, and lake inlets; the geographical divisions of the receiving lake may include the central lake area, nearshore area, inlet area, and outlet area; and the distribution characteristics of the surrounding environment, such as the location of residential areas, industrial areas, and vegetated areas. Based on this actual geographical information, specific spatial coordinates are assigned to each field unit. For example, the coordinates of field units related to the water source correspond to the latitude and longitude of the water intake, the coordinates of field units related to the water transmission process are distributed along the water pipeline, and the coordinates of field units related to the receiving lake cover different zones of the lake.

[0059] Step S1194: According to the spatial layout and actual spatial coordinates, place each field unit in the virtual field space. Based on the correlation relationship in the correlation strength comparison table, build a correlation channel between adjacent field units. The width of the correlation channel is proportional to the correlation strength value.

[0060] A virtual field space corresponding to the actual geographic space is constructed within a computer virtual environment. Based on the determined actual spatial coordinates, each field unit is placed at its corresponding position in the virtual field space. Then, referring to the priority sequence of association relationships and the association strength lookup table, association channels are established between field units arranged in an adjacent manner. The width of the association channel visually reflects the association strength; the higher the association strength value, the larger the channel width. For example, the channel width between field unit pairs with an association strength value of 0.9 is set to three times the channel width between field unit pairs with an association strength value of 0.3.

[0061] Step S1195: Configure data transmission rules for each associated channel. The data transmission rules are determined based on the type and strength of the association relationship, and determine the transmission direction, transmission rate and transmission priority of data in the associated channel.

[0062] Correlation relationships can be unidirectional or bidirectional, and the correlation strength also varies. These factors collectively determine the data transmission rules. For unidirectional correlations, data transmission proceeds from the influencing party to the affected field unit; for bidirectional correlations, bidirectional data transmission is allowed. The transmission rate is positively correlated with the correlation strength; the higher the correlation strength, the faster the data transmission rate is set in the channel to reflect more frequent interactions. Transmission priority is also determined based on the correlation strength. Data in channels with high correlation strength has higher transmission priority, ensuring that important water quality impact information is transmitted first. For example, if field units M and N have a strong bidirectional correlation (correlation strength 0.8), data can be transmitted bidirectionally at high speed in the MN channel, with a higher priority than the field unit OP channel with a correlation strength of 0.4.

[0063] Step S1196: Extract timestamp information from the field basic data, analyze the time synchronization of data from different sources, determine the time synchronization benchmark, and adjust the timestamp of the basic data in each field unit based on the time synchronization benchmark so that the data of different field units in the same period are aligned in the time dimension.

[0064] The basic data for the field comes from multiple sources, including the water source, the water conveyance process, the receiving lake, and the surrounding environment. The collection times of data from these sources may differ. Timestamp information is extracted from all data, and their collection frequency and time deviation are analyzed. A high-precision, high-stability time source, such as a GPS clock or a network time server, is selected as the time synchronization benchmark. Based on this benchmark, the timestamps of the basic data in each field unit are calibrated, adjusting data from different collection times for the same period to a unified timeline. For example, if the timestamp of data collection in a certain field unit has a deviation of ±2 minutes, the timestamps of the above data are uniformly corrected to whole minutes using the time synchronization benchmark as a reference, ensuring the comparability of data from different field units at the same time point.

[0065] Step S1197: Based on the data flow requirements of the field units after time synchronization and in conjunction with the transmission rules of the associated channels, plan the data flow path between field units.

[0066] After time synchronization, data from the field units needs to flow through the associated channels according to their relationships. Based on the function and data generation characteristics of each field unit, the data source and receiving target are determined. Combining the transmission direction, transmission rate, and transmission priority rules of the associated channels, specific data flow paths are planned. For example, water quality and flow data generated by the water source field unit need to be transmitted through the associated channel to the first field unit in the water conveyance process, then sequentially to downstream field units, and finally to the receiving lake field unit. Simultaneously, data from surrounding environmental field units also needs to flow into the receiving lake field unit or the water conveyance process field unit through corresponding associated channels. The planning of data flow paths should ensure that data can be transmitted in an orderly and efficient manner according to their relationships, meeting the requirements of dynamic coupling and fusion processing.

[0067] Step S1110: Divide the field basic data into the corresponding field units and association channels according to the field unit boundary characteristics and the initial water quality association field framework, and add dynamic rules for data flow between units to the association channels to form the water quality association field.

[0068] Based on the spatial and temporal boundaries defined in the field unit boundary characteristics, and the unit division in the initial water quality correlation field framework, the basic field data is allocated to the corresponding field units. Data belonging to a particular field unit is populated within that unit according to its physical location and time range of collection. Simultaneously, corresponding data needs to be populated for the correlation channels between field units; this data primarily reflects the correlation relationships between field units. Furthermore, dynamic rules for data flow between units are added to the correlation channels. These rules specify the conditions, methods, speeds, and data processing methods for data flow within the correlation channels. For example, when the data in a certain field unit reaches a certain threshold, the data begins to flow to another correlated field unit; data may require format conversion or standardization during the flow process. By populating the field units and correlation channels with data and supplementing the dynamic rules, a complete water quality correlation field is ultimately formed.

[0069] Step S120: Perform interaction relationship modeling operation on each field unit in the water quality correlation field to obtain the field unit interaction model. The field unit interaction model is used to quantitatively describe the transmission mode, transmission intensity and dynamic change characteristics of water quality influence between different field units.

[0070] After constructing the water quality correlation field, it is necessary to analyze the interaction relationships between the field units. For each field unit in the water quality correlation field, an interaction relationship modeling operation must be performed to establish a model that can accurately describe the interactions between field units, i.e., the field unit interaction model. The main function of this field unit interaction model is to quantify the transmission mode of water quality influence between different field units, such as whether it is direct or indirect transmission, unidirectional or bidirectional transmission, etc.; the transmission intensity, i.e., the degree of influence of water quality changes in one field unit on the water quality of another field unit; and the dynamic change characteristics, i.e., how the above influence transmission relationship changes with time and other factors. By establishing the field unit interaction model, we can gain a deeper understanding of the intrinsic connections between the various parts of the water quality correlation field.

[0071] Step S121: Extract the core data features of each field unit from the water quality correlation field. The core data features are key data representations that can represent the water quality impact attributes of the field unit. One field unit corresponds to a set of core data features.

[0072] Core data features are the key data representations of a field unit that centrally reflect its water quality impact attributes. From the data contained in each field unit within the water quality correlation field, data analysis and feature extraction methods are used to select the data features that are most critical to water quality and best represent the characteristics of the field unit. These data features may include statistical characteristics such as the average, maximum, minimum, and rate of change of certain key water quality indicators in the field unit, or they may include spectral characteristics, trend characteristics, etc. Each field unit corresponds to a set of core data features, and the core data features in this set collectively reflect the water quality impact attributes of that field unit.

[0073] Step S122: After the core data features are extracted, analyze the overlapping features between the core data feature sets of adjacent field units to determine the water quality influencing factors that coexist between adjacent units.

[0074] Adjacent field units refer to field units that are geographically adjacent in the spatial layout of the water quality correlation field. After extracting the core data features of all field units, a comparative analysis of the core data feature sets of adjacent field units is performed to identify overlapping features. Overlapping features are data features that exist in the core data feature sets of two adjacent field units and have similar meanings and properties. By analyzing these overlapping features, common water quality influencing factors between adjacent units can be identified. These common water quality influencing factors are the basis for the interaction between adjacent field units, enabling them to mutually influence water quality conditions.

[0075] Step S123: Based on the common water quality influencing factors, identify the interaction types between adjacent field units. The interaction types are enhancement interaction, inhibition interaction, and neutral interaction. The identification of the interaction type is based on the change response relationship of core data features.

[0076] Based on the common water quality influencing factors among adjacent field units, the interaction types between them are further identified. These interaction types mainly include reinforcing interactions, inhibiting interactions, and neutral interactions. Reinforcing interactions refer to a change in a common water quality influencing factor in one field unit leading to a similar change in the same direction in another adjacent field unit, with the degree of change being amplified. Inhibiting interactions refer to a change in a common water quality influencing factor in one field unit leading to a reverse change in the same factor in another adjacent field unit, or a weakening of the change. Neutral interactions refer to a change in a common water quality influencing factor in one field unit having a very small, almost negligible, impact on the same factor in another adjacent field unit. The identification of interaction types is based on the response relationship to changes in core data features. By observing the direction and degree of change in the corresponding core data features in adjacent field units when the core data features of one field unit change, the type of interaction is determined.

[0077] Step S124: Calculate the interaction strength parameter corresponding to each interaction type. The interaction strength parameter is determined based on the contribution degree of the standardized common influence factor and the corresponding value in the data association strength comparison table to form an initial interaction strength comparison table.

[0078] For each identified interaction type, its corresponding interaction strength parameter needs to be calculated. The interaction strength parameter quantifies the strength of the interaction. The calculation is based on the contribution of standardized common influencing factors, combined with the corresponding values ​​in the data association strength lookup table related to these common influencing factors. The contribution of standardized common influencing factors reflects the importance of the factor in adjacent field units, while the values ​​in the data association strength lookup table reflect the closeness of the association between the data source and the factor. By comprehensively considering these two factors, the interaction strength parameter is calculated. After organizing the interaction types and their corresponding interaction strength parameters between different adjacent field units, an initial interaction strength lookup table is formed.

[0079] Step S125: Monitor the changes in core data characteristics of each field unit in the water quality correlation field, capture the real-time fluctuations of core data characteristics, and establish a data characteristic fluctuation sequence.

[0080] Continuous monitoring of the core data characteristics of each field unit in the water quality correlation field is conducted to track their changes in real time. By setting the monitoring frequency and data collection points, the latest values ​​of the core data characteristics are continuously acquired, capturing their real-time fluctuations. The fluctuation values ​​of the core data characteristics arranged in chronological order are recorded to form a data characteristic fluctuation sequence. The data characteristic fluctuation sequence reflects the changing trends and fluctuation patterns of the core data characteristics over time.

[0081] Step S126: Based on the data feature fluctuation sequence, perform dynamic adjustment of the interaction intensity parameters in the initial interaction intensity lookup table to match the interaction intensity parameters with the amplitude and frequency of the data feature fluctuations, forming a dynamically updated interaction intensity lookup table.

[0082] Step S1261: Extract fluctuation features from the standardized data feature fluctuation sequence, extracting key fluctuation parameters such as fluctuation amplitude, fluctuation frequency, fluctuation period, and fluctuation trend. The core data features of each field unit correspond to a set of fluctuation parameters based on the standardized data.

[0083] First, the data feature fluctuation sequence is standardized to eliminate differences in units and orders of magnitude between different data features. Standardization can be achieved using methods such as Z-score standardization or min-max standardization. Then, fluctuation features are extracted from the standardized data feature fluctuation sequence. Key extracted fluctuation parameters include fluctuation amplitude, fluctuation frequency, fluctuation period, and fluctuation trend. Fluctuation amplitude is determined by calculating the difference between the maximum and minimum values ​​in the data feature fluctuation sequence; fluctuation frequency is calculated by counting the number of fluctuation periods per unit time; fluctuation period refers to the time required for a data feature to complete one full fluctuation; and fluctuation trend is determined by linear fitting or other trend analysis methods to determine whether the data exhibits an upward, downward, or stable trend. Each core data feature of a field unit corresponds to a set of the above-mentioned fluctuation parameters based on standardized data, which describe the fluctuation characteristics of the core data feature from different perspectives.

[0084] Step S1262: Integrate the physical laws of water quality impact transmission and historical data verification results to form a mapping relationship model between fluctuation parameters and interaction intensity parameters. The mapping relationship model records the adjustment direction and adjustment magnitude of the interaction intensity parameters corresponding to different combinations of fluctuation parameters.

[0085] The physical laws governing the transmission of water quality effects refer to the natural physical laws governing the transmission of water quality between different field units, such as the law of conservation of mass and the law of diffusion. Historical data verification results refer to empirical verification of the relationship between fluctuation parameters and interaction intensity parameters obtained through the analysis of water quality data and interaction intensity parameters over a past period. Integrating these two aspects of information, a mapping model between fluctuation parameters and interaction intensity parameters is constructed. This mapping model records the direction (increase or decrease) and magnitude (adjustment amount) of the interaction intensity parameters under different combinations of fluctuation parameters. For example, when the fluctuation amplitude and frequency are large, according to physical laws and historical data verification, the interaction intensity parameter may need to be increased by a certain amount; when the fluctuation trend shows a downward trend, the interaction intensity parameter may need to be decreased, and so on.

[0086] Step S1263: Input the fluctuation parameters of each field unit into the mapping relationship model to obtain the interaction intensity parameter adjustment suggestion for each field unit. The adjustment suggestion includes the parameter adjustment amount and the adjusted target interaction intensity parameter.

[0087] The fluctuation parameters of each field unit extracted earlier are input into the mapping model. Based on the input fluctuation parameters, the mapping model finds the corresponding combination of fluctuation parameters and, according to the adjustment direction and magnitude recorded in the model, generates an adjustment suggestion for the interaction intensity parameter of each field unit. The adjustment suggestion explicitly includes the parameter adjustment amount, i.e., the value that needs to be increased or decreased for the current interaction intensity parameter, and the target interaction intensity parameter after adjustment, i.e., the value that the interaction intensity parameter should reach after adjustment.

[0088] Step S1264: Extract the current interaction intensity parameters between the corresponding field units in the initial interaction intensity comparison table, compare the current parameters with the target interaction intensity parameters in the adjustment suggestions, and determine the actual adjustment difference.

[0089] From the initial interaction intensity lookup table, locate the current interaction intensity parameter between the corresponding field units. Compare this current interaction intensity parameter with the target interaction intensity parameter in the adjustment suggestion, and calculate the difference between the two. This difference is the actual adjustment difference. The actual adjustment difference reflects the specific amount of adjustment required to bring the interaction intensity parameter to the target value.

[0090] Step S1265: Adjust the corresponding parameters in the initial interaction intensity comparison table point by point according to the actual adjustment difference. First, adjust the direct interaction intensity parameters between adjacent field units, and then derive the adjustment amount of the indirect interaction intensity parameters based on the adjustment results of the direct parameters.

[0091] Based on the actual adjustment difference, the interaction intensity parameters in the initial interaction intensity comparison table are adjusted point by point. The adjustment order is to first adjust the direct interaction intensity parameters between adjacent field units, because direct interactions are usually more direct and significant. After completing the adjustment of the direct interaction intensity parameters, the adjustment amount of the indirect interaction intensity parameters is derived based on the adjustment results of these direct parameters. The indirect interaction intensity parameters refer to the interaction intensity parameters generated through intermediate field units, and their adjustment amount can be derived based on the changes in the direct interaction intensity parameters and the correlation between field units. For example, if the direct interaction intensity parameter between field unit A and field unit B increases, and the direct interaction intensity parameter between field unit B and field unit C also changes, then the indirect interaction intensity parameter between field unit A and field unit C needs to be adjusted accordingly based on the adjustment results of AB and BC.

[0092] Step S1266: Perform internal consistency verification on the adjusted interaction intensity comparison table to ensure that the interaction intensity parameters between the same field unit and different adjacent units are logically unified, and that the interaction intensity parameters at different levels are mutually adapted.

[0093] After adjusting the interaction intensity parameters, an internal consistency check is performed on the adjusted interaction intensity comparison table. The purpose of this check is to ensure that the interaction intensity parameters between the same field unit and its adjacent units are logically consistent and do not contradict each other. For example, the interaction type and intensity parameters of the same field unit to its adjacent units should be consistent with its own core data characteristics and the characteristics of the adjacent units. Simultaneously, it is necessary to ensure that interaction intensity parameters at different levels are compatible; that is, parameters at different levels, such as direct interaction intensity parameters and indirect interaction intensity parameters, can coordinate with each other to accurately reflect the interaction between field units. If inconsistencies or mismatches are found during the check, the corresponding interaction intensity parameters need to be readjusted and corrected.

[0094] Step S1267: Based on the overall stability requirements of the water quality correlation field, set an adjustment threshold for the interaction intensity parameter. When the single adjustment amount exceeds the adjustment threshold, perform the adjustment operation in stages.

[0095] To ensure the overall stability of the water quality correlation field and avoid instability caused by sudden and significant adjustments to the interaction intensity parameter, an adjustment threshold for the interaction intensity parameter needs to be set. The adjustment threshold refers to the maximum allowable change in the interaction intensity parameter during a single adjustment operation. When the calculated actual adjustment difference exceeds the adjustment threshold, a large-scale adjustment is not performed all at once; instead, the adjustment operation is carried out gradually in stages. The adjustment range of each stage does not exceed the adjustment threshold. After multiple stages of adjustment, the interaction intensity parameter eventually reaches the target value. This allows the water quality correlation field to maintain a relatively stable state during the adjustment process.

[0096] Step S1268: Record the time point, reason for adjustment and adjustment range of each parameter adjustment to form a parameter adjustment log.

[0097] Every adjustment to the interaction strength parameter was meticulously recorded, including the time of the adjustment, the reason for the adjustment (such as changes in the data characteristic fluctuation sequence, model analysis results, etc.), and the magnitude of the adjustment. These records were then compiled into a parameter adjustment log.

[0098] Step S1269: Align the adjusted interaction intensity comparison table with the timestamps of the data characteristic fluctuation sequence, integrate the adjusted interaction intensity comparison tables of all time nodes, and form a dynamically updated interaction intensity comparison table sequence.

[0099] The adjusted interaction intensity comparison table is precisely aligned with the timestamps of the data characteristic fluctuation sequence to ensure that the adjustment time of the interaction intensity parameter corresponds to the time of data characteristic fluctuation. Then, all adjusted interaction intensity comparison tables at different time points are integrated and arranged in chronological order to form a dynamically updated interaction intensity comparison table sequence. This interaction intensity comparison table sequence reflects the dynamic change process of the interaction intensity parameter over time. By viewing this sequence, the historical trend and pattern of the interaction intensity between field units can be obtained.

[0100] Step S127: Analyze the indirect interaction relationship between non-directly adjacent field units through intermediate field units, and derive the indirect interaction strength parameter based on the direct interaction strength parameter.

[0101] Besides the direct interaction between adjacent field units, indirect interaction can also occur between non-directly adjacent field units through intermediate field units. For example, field units A and C are not directly adjacent, but they are both adjacent to field unit B. Therefore, field unit A can indirectly interact with field unit C through field unit B. Analyzing these indirect interaction relationships, the indirect interaction strength parameter is derived based on the direct interaction strength parameter. The derivation method can be determined based on factors such as the number of intermediate field units, the magnitude of the direct interaction strength parameter, and the transmission path. For example, for an indirect interaction relationship transmitted through one intermediate field unit, the indirect interaction strength parameter can be obtained by multiplying or combining the two direct interaction strength parameters in a certain way; for indirect interaction relationships transmitted through multiple intermediate field units, a similar chain-like derivation method can be used.

[0102] Step S128: After supplementing the indirect interaction relationship, determine the transmission delay characteristics of the interaction of each field unit. The transmission delay characteristics are determined based on the spatial distance of the field units, the length of the data flow path and the characteristics of the transmission medium, forming a set of transmission delay parameters.

[0103] After considering indirect interactions, it is also necessary to determine the propagation delay characteristics of each field unit's interaction. The propagation delay characteristic refers to the time required for an interaction from one field unit to propagate to another. It is primarily determined based on the spatial distance between field units, the length of the data flow path, and the characteristics of the transmission medium. The greater the spatial distance between field units and the longer the data flow path, the greater the propagation delay typically is. The characteristics of the transmission medium, such as its density and fluidity, also affect the transmission speed, thus influencing the propagation delay. By analyzing these factors, the propagation delay time of each field unit's interaction is calculated, forming a propagation delay parameter set. The parameters in the propagation delay parameter set correspond to the interactions between field units and are used to describe the temporal characteristics of the interaction propagation.

[0104] Step S129: Integrate the interaction type, the dynamically updated interaction intensity lookup table, and the transmission delay parameter set to construct a field unit interaction model.

[0105] The construction of the field unit interaction model aims to organically integrate key elements such as interaction type, dynamically changing interaction intensity, and transmission delay to form a model that can quantitatively describe the transmission law of water quality influence between field units. This model will serve as the core tool for dynamic coupling and fusion processing and water quality trend prediction; therefore, it is necessary to clarify its inputs and outputs, module composition, parameter configuration, and calculation logic.

[0106] For example, step S1291: Define the input and output variables of the field unit interaction model. The input variables are the core data characteristics of each field unit, the dynamic interaction intensity comparison table parameters, and the transmission delay parameters. The output variables are the water quality influence transmission results.

[0107] Input variables are the foundational data for model calculations. The core data features of each field unit represent its own water quality impact attributes and are the source of interactions. The dynamic interaction intensity table parameters reflect the real-time intensity of interactions between field units; the propagation delay parameter characterizes the time required for the impact to propagate. The output variable, the water quality impact propagation result, is the final product of the model calculations. It quantitatively represents the specific degree and direction of the impact of a field unit's core data feature, after being propagated to another field unit through interaction, on the water quality of that field unit. For example, if the core data feature of field unit A (such as high ammonia nitrogen concentration) interacts with field unit B (enhanced type, interaction intensity parameter 0.7, propagation delay 1 hour), it will cause the ammonia nitrogen concentration in field unit B to increase by a certain amount after 1 hour. This increase is the water quality impact propagation result.

[0108] Step S1292: Configure the interaction calculation module of the model according to the interaction type. Enhanced interaction corresponds to the positive gain calculation submodule, suppression interaction corresponds to the reverse attenuation calculation submodule, and neutral interaction corresponds to the no gain and no attenuation calculation submodule.

[0109] Different interaction types have drastically different ways of transmitting water quality impacts, therefore, a dedicated interaction calculation submodule is needed for each type. Enhanced interactions amplify water quality impacts; the positive gain calculation submodule receives the core data features of the field unit and performs positive amplification processing based on the interaction intensity parameter, for example, multiplying the input value by a gain coefficient greater than 1 (derived from the interaction intensity parameter). Inhibitory interactions weaken or cancel out water quality impacts; the reverse attenuation calculation submodule reverses or attenuates the input core data features, for example, multiplying the input value by an attenuation coefficient between 0 and 1, or inverting the sign before multiplying by the attenuation coefficient. Neutral interactions have virtually no effect on water quality impact transmission; the no-gain, no-attenuation calculation submodule directly outputs the input core data features as is or with minor adjustments.

[0110] Step S1293: Use the parameters in the dynamic interaction intensity comparison table as the weight coefficients of the interaction calculation module. Set the weight coefficient of enhanced interaction to a positive value, the weight coefficient of inhibited interaction to a negative value, and the weight coefficient of neutral interaction to zero.

[0111] The parameter values ​​in the dynamic interaction strength lookup table reflect the strength of the interaction, which are then converted into weight coefficients in the interaction calculation module. For enhanced interactions, the weight coefficient is positive and equal to the corresponding parameter value in the dynamic interaction strength lookup table; the larger the value, the stronger the gain effect. For example, an enhanced interaction with an interaction strength parameter of 0.6 has a weight coefficient of 0.6. The weight coefficient for inhibiting interactions is negative, and its value is the opposite of the parameter value in the dynamic interaction strength lookup table. For example, an inhibiting interaction with an interaction strength parameter of 0.5 has a weight coefficient of -0.5; the larger the absolute value, the stronger the attenuation effect. The weight coefficient for neutral interactions is zero, meaning that the core input data features have no impact on the output result after passing through this submodule.

[0112] Step S1294: Embed a propagation delay calculation module in the field unit interaction model to convert the delay information in the propagation delay parameter set into a time decay factor.

[0113] The delay times (e.g., 1 hour, 2 hours) in the propagation delay parameter set need to be converted into time decay factors that the model can handle. The time decay factor simulates the decay effect of the influence over time during propagation; generally, the longer the delay time, the smaller the decay factor. For example, an exponential decay function can be used to convert the propagation delay parameter into a time decay factor. Assuming the propagation delay parameter is t, the time decay factor might be exp(-k*t), where k is the decay coefficient, which can be calibrated based on historical data or physical laws. The propagation delay calculation module receives the preliminary influence results output by the interactive calculation module and multiplies them by the time decay factor to obtain the water quality influence propagation result after time delay decay.

[0114] Step S1295: Set the coupling calculation logic of the field unit interaction model. The coupling calculation logic includes first calculating the basic interaction results between field units through the interaction calculation module, and then adjusting the time decay of the basic interaction results through the transmission delay calculation module to obtain the final interaction results.

[0115] The coupled computation logic defines the computation order and data transfer method for each module within the model. First, the core data features of the source field unit are input into the interaction computation module. Based on the interaction type, the corresponding sub-module (positive gain, negative attenuation, or no gain / no attenuation) is selected, and the basic interaction result is calculated using the weight coefficients from the dynamic interaction intensity lookup table. For example, if the core data feature value of the source field unit is C, and the enhanced interaction weight coefficient is W, then the basic interaction result is C*W. Next, the basic interaction result is input into the propagation delay calculation module. This module calculates the time decay factor based on the corresponding propagation delay parameter and multiplies the basic interaction result by the time decay factor to obtain the final water quality impact propagation result. For example, if the basic interaction result is R and the time decay factor is D, then the final interaction result is R*D.

[0116] Step S1296: Collect historical water quality data and corresponding multi-source raw data of the water diversion project and the receiving lake, and perform standardized preprocessing on the multi-source raw data to form a training dataset for the field unit interaction model. The training dataset includes standardized field unit data, interaction parameters and standardized actual water quality impact results under different working conditions.

[0117] To ensure the model accurately reflects actual interaction patterns, historical data is required for training. Historical water quality data of the water diversion project under different operating conditions (e.g., different seasons, different flow rates, different pollution source inputs) are collected, along with corresponding multi-source raw data on the water source, water conveyance process, receiving lakes, and surrounding environment. This multi-source raw data is processed using the same standardization preprocessing method as the basic field data to obtain standardized field unit data. Simultaneously, based on water quality changes in historical data, the actual water quality impact transmission results between field units are analyzed and standardized. The standardized field unit data, corresponding interaction parameters (interaction type, historical interaction intensity parameters, historical transmission delay parameters), and standardized actual water quality impact results are combined to form a training dataset.

[0118] Step S1297: Divide the training dataset into a training subset and a validation subset according to the proportion. Use the training subset to calibrate the parameters of the interaction calculation module, the propagation delay calculation module and the coupled calculation logic of the field unit interaction model. Use the validation subset to verify the performance of the calibrated field unit interaction model. Analyze the deviation rate index between the interaction results output by the field unit interaction model and the actual water quality impact results.

[0119] The training dataset is randomly divided into a training subset and a validation subset according to a certain ratio (e.g., 7:3 or 8:2). The training subset is used for model parameter calibration. By inputting standardized field unit data and interaction parameters from the training subset into the model, the interaction results output by the model are obtained and compared with the actual water quality impact results in the training subset. Based on the comparison results, parameters such as the gain coefficient and attenuation coefficient in the interaction calculation module and the attenuation coefficient k in the propagation delay calculation module are adjusted to minimize the deviation between the model output results and the actual results. Common calibration methods include least squares and gradient descent. After calibration, the performance is verified using the validation subset. Data from the validation subset is input into the model, and the deviation rate index (e.g., root mean square error, mean absolute percentage error, etc.) between the model output results and the actual water quality impact results is calculated. If the deviation rate index is within an acceptable range, the model calibration is successful; otherwise, the parameters need to be readjusted or the model structure needs to be checked.

[0120] Step S1298: Based on the deviation rate index, adjust the calculation parameters and coupling calculation logic of the field unit interaction model, connect the finally calibrated and optimized field unit interaction model with the field management system of the water quality correlation field, and configure the real-time calling interface and parameter update interface of the field unit interaction model.

[0121] Based on the deviation rate obtained from the validation subset, the sources of model error are analyzed. If the deviation mainly comes from the interaction calculation module, the conversion method of the weight coefficients or the calculation function of the sub-module are adjusted; if the deviation mainly comes from the propagation delay calculation module, the attenuation coefficient k is recalibrated or different attenuation function forms are tried; if the deviation comes from the overall calculation process, it may be necessary to adjust the coupled calculation logic, such as adding intermediate calculation steps or adjusting the data transfer method between modules. After multiple iterations of adjustments, the model performance is adjusted until it meets the requirements. The final calibrated and optimized field unit interaction model needs to be connected to the field management system of the water quality associated field to receive the core data characteristics, dynamic interaction intensity parameters, and propagation delay parameters of the field units in real time, and output the water quality impact propagation results. For this purpose, a real-time calling interface (such as an API interface) of the model needs to be configured to allow the field management system to trigger model calculations in real time; at the same time, a parameter update interface needs to be configured so that when the model parameters need to be adjusted according to new data or operating conditions, they can be updated online through this interface.

[0122] Step S1210: After the field unit interaction model is constructed, the interaction process of each field unit is simulated through the field unit interaction model to verify the degree of fit between the interaction results output by the model and the actual data changes in the water quality correlation field. Based on the degree of fit, the model parameters are optimized to form a stable interaction relationship modeling result.

[0123] After the interaction model of the field units is constructed, it needs to be validated and optimized. The model simulates the interaction process of each field unit, i.e., by inputting the core data characteristics of each field unit, the parameters of the dynamic interaction intensity comparison table, and the transmission delay parameter, and then running the model to obtain the interaction results. The interaction results output by the model are compared with the actual observed data changes in the water quality correlation field to analyze the degree of fit between the two. The higher the degree of fit, the more accurately the model reflects the actual interaction process; the lower the degree of fit, the more biased the model has, requiring optimization. Based on the results of the degree of fit analysis, the parameters in the model, such as the coefficients of the interaction calculation module and the transformation relationship of the transmission delay calculation module, are adjusted and optimized. The simulation, validation, and optimization process is repeated until the degree of fit between the interaction results output by the model and the actual data changes reaches the predetermined requirements, forming a stable interaction relationship modeling result.

[0124] Step S130: Based on the interaction relationship described by the field unit interaction model, perform dynamic coupling and fusion processing on the field basic data of the water quality correlation field to generate real-time updated coupled water quality data. When the interaction relationship changes, the dynamic coupling and fusion processing performs an adjustment operation of the fusion strategy so that the fusion strategy follows the changes in the field interaction.

[0125] The field unit interaction model describes the interaction relationships between field units. Based on these relationships, dynamic coupling and fusion processing is performed on the basic field data of the water quality correlation field. Dynamic coupling and fusion processing refers to organically combining and integrating basic field data from different field units according to their interaction relationships to generate coupled water quality data that comprehensively reflects the water quality status of the entire water quality correlation field. Since the interaction relationships between field units may change with time and environmental conditions, dynamic coupling and fusion processing needs to be able to adjust the fusion strategy promptly when these relationships change. The fusion strategy includes data fusion methods, weight allocation, and processing procedures.

[0126] Step S131: Extract core interaction factors from the interaction relationships. The core interaction factors are key data items that determine the intensity and direction of interaction between field units. Each interaction type corresponds to a set of core interaction factors.

[0127] Interaction relationships encompass various factors influencing interactions between field units, from which core interaction factors are extracted. Core interaction factors refer to key data items that play a decisive role in the intensity and direction of interactions between field units. Different interaction types, such as enhancing, inhibiting, and neutral interactions, may have different core interaction factors; therefore, each interaction type corresponds to a set of core interaction factors. For example, for enhancing interactions, core interaction factors might be key water quality indicators or data features that promote and enhance the transmission of water quality impacts; for inhibiting interactions, core interaction factors might be key data items that hinder or weaken the transmission of water quality impacts. Extracting core interaction factors simplifies subsequent data fusion processing, improving fusion efficiency and accuracy.

[0128] Step S132: Integrate the rules related to the core interaction factors to form a coupling fusion operator, which includes interaction intensity weight allocation rules, data complementarity fusion rules, and dynamic adjustment rules.

[0129] The fusion of core interaction factors needs to follow certain rules. These rules, related to core interaction factors, are integrated to form a coupling fusion operator. The coupling fusion operator is the core component for performing data fusion operations, and it includes interaction strength weight allocation rules, data complementarity fusion rules, and dynamic adjustment rules. The interaction strength weight allocation rules specify how to assign fusion weights to core interaction factor data from different field units based on interaction strength parameters; the data complementarity fusion rules specify how to perform complementary fusion of core interaction factor data from different sources and with different characteristics to fully utilize the advantages of each data source; and the dynamic adjustment rules specify how to adjust the fusion weights and fusion methods when the interaction relationship changes to adapt to new interaction situations.

[0130] Step S133: The field basic data of each field unit in the water quality correlation field is classified and extracted according to the core interaction factor to form a factor correlation data subset. One factor correlation data subset corresponds to one core interaction factor and correlation field unit data.

[0131] Based on the extracted core interaction factors, the basic field data of each field unit in the water quality correlation field are classified and extracted. Field data related to the same core interaction factor are extracted and combined to form factor correlation data subsets. Each factor correlation data subset corresponds to one core interaction factor and contains data from all correlation field units related to that core interaction factor.

[0132] Step S134: Input the factor-related data subset into the corresponding coupling fusion operator, and perform standardization preprocessing on the original data in the data subset; assign fusion weights to the standardized data according to the interaction strength weight allocation rule, so that the assigned weight values ​​are consistent with the corresponding interaction strength parameters in the dynamically updated interaction strength lookup table; perform fusion calculation on the standardized data with assigned fusion weights according to the data complementarity fusion rule, integrate the complementary information about the same core interaction factor in the data of different field units, and generate factor-level fusion data.

[0133] A subset of factor-related data is input into a coupling fusion operator corresponding to its core interaction factor for processing. First, the original data in the subset is standardized to eliminate dimensional differences between data points, ensuring comparability. Standardization can employ methods such as Z-score standardization or min-max standardization. Then, fusion weights are assigned to the standardized data according to the interaction strength weight allocation rules in the coupling fusion operator. The allocation of fusion weights ensures that their values ​​match the corresponding interaction strength parameters in the dynamically updated interaction strength lookup table, reflecting the impact of interaction strength on data fusion. Next, fusion calculations are performed on the standardized data with assigned fusion weights according to the data complementarity fusion rules. These rules aim to integrate complementary information about the same core interaction factor from different field units. For example, different field units may reflect the characteristics of the core interaction factor from different perspectives. Fusion calculations integrate this information to generate factor-level fused data that comprehensively reflects the core interaction factor.

[0134] Step S135: Extract the delay information from the set of transmission delay parameters, perform time synchronization calibration on the factor-level fusion data, so that the fusion data of different field units are consistent in the time dimension, and generate time-synchronized fusion data.

[0135] Factor-level fusion data originates from different field units. Due to factors such as data acquisition time and interaction propagation delays between field units, these fusion data may be out of sync in time. To ensure the accuracy of subsequent analysis and application, time synchronization calibration of the factor-level fusion data is necessary. Delay information is extracted from the propagation delay parameter set, reflecting the time delay in the propagation of interactions between different field units. Based on this delay information, the factor-level fusion data is adjusted in time, bringing the fusion data from different field units to a unified time reference, ensuring consistency across the time dimension. After time synchronization calibration, time-synchronized fusion data is generated.

[0136] Step S136: Monitor the dynamic changes of the interaction relationship. When the parameters in the dynamically updated interaction strength lookup table are adjusted, trigger the parameter update of the coupling fusion operator so that the weight allocation rule and fusion rule of the coupling fusion operator follow the changes of the interaction.

[0137] Continuous monitoring of dynamic changes in interaction relationships is crucial, with a focus on parameter changes in the dynamically updated interaction strength lookup table. When parameters in the table are adjusted, it indicates a change in the interaction strength between field units, necessitating a parameter update for the coupling fusion operator. The parameters of the coupling fusion operator include weight coefficients in the weight allocation rules and fusion parameters in the data complementarity fusion rules. Updating these parameters ensures that the weight allocation and fusion rules of the coupling fusion operator adjust to changes in interactions, guaranteeing that the fusion process adapts to new interaction scenarios and generates accurate fused data.

[0138] Step S137: Based on the updated coupling fusion operator, perform real-time fusion processing on the subsequently collected field basic data to generate new factor-level fusion data and new time-synchronized fusion data.

[0139] Step S1371: Establish a real-time interactive link for field basic data, so that the subsequently collected field basic data can be transmitted to the fusion processing module within a preset time limit. Set up a data receiving buffer in the fusion processing module to temporarily store and format the real-time transmitted field basic data. Extract the weight allocation rules and fusion rules from the updated coupled fusion operators and load them into the real-time fusion processing engine.

[0140] To achieve real-time fusion processing of subsequently acquired field baseline data, a real-time interaction link for the field baseline data needs to be established. This real-time interaction link ensures that newly acquired field baseline data can be rapidly transmitted to the fusion processing module within a preset time limit via network or other transmission methods. A data receiving buffer is set up in the fusion processing module to temporarily store the real-time transmitted field baseline data and to standardize the data format, converting different formats into a unified format for subsequent processing. The latest weight allocation rules and fusion rules are extracted from the updated coupling fusion operator and loaded into the real-time fusion processing engine, enabling the engine to perform data fusion processing according to the new rules.

[0141] Step S1372: Extract real-time data from the data receiving buffer according to the core interaction factors, and perform standardized preprocessing on the real-time data to form a standardized real-time factor-related data subset. Input the standardized real-time factor-related data subset into the real-time fusion processing engine, and assign real-time fusion weights to each standardized data item in the standardized real-time factor-related data subset according to the updated weight allocation rules. The assigned weight values ​​are synchronized with the latest dynamic interaction intensity lookup table parameters.

[0142] Real-time data related to each core interaction factor is extracted from the data receiving buffer based on the classification of core interaction factors. The extracted real-time data undergoes standardization preprocessing to eliminate dimensional differences, forming a standardized subset of real-time factor-related data. This standardized subset is then input into the real-time fusion processing engine. The engine assigns real-time fusion weights to each standardized data item in the subset according to the updated weight allocation rules. The assigned real-time fusion weights must be synchronized with the latest dynamic interaction strength lookup table parameters to ensure that the weights accurately reflect the current interaction strength.

[0143] Step S1373: Perform complementary fusion calculations on the data assigned real-time fusion weights according to the updated fusion rules, integrate the effective information in the data, and generate real-time factor-level fusion data.

[0144] After allocating real-time fusion weights, the real-time fusion processing engine performs complementary fusion calculations on the weighted data according to the updated fusion rules. The fusion calculation process aims to integrate effective information from the data, fully utilizing the complementary information about core interaction factors in real-time data from different field units. Through certain calculation methods, such as weighted averaging and logical combination, real-time factor-level fusion data is generated. This real-time factor-level fusion data reflects the overall status of the core interaction factors at the current moment.

[0145] Step S1374: Retrieve the latest delay information from the set of transmission delay parameters, perform time synchronization calibration on the real-time factor-level fusion data, and splice the time-synchronized real-time factor-level fusion data with the historical factor-level fusion data to form a continuous factor-level fusion data sequence.

[0146] To ensure the temporal continuity and consistency of real-time factor-level fusion data, the latest delay information from the transmission delay parameter set is retrieved to perform time synchronization calibration on the real-time factor-level fusion data. The time synchronization calibration method is the same as described above, adjusting the data's timestamp based on the delay information. After calibration, the time-synchronized real-time factor-level fusion data is concatenated with historically stored factor-level fusion data in chronological order to form a continuous factor-level fusion data sequence. This factor-level fusion data sequence completely records the changes of the core interaction factors over time.

[0147] Step S1375: Perform outlier removal processing on the continuous factor-level fusion data sequence to remove abnormal data points caused by data acquisition errors or transmission interference. Store the processed factor-level fusion data sequence and the corresponding time-synchronized fusion data in the fusion data warehouse and update the real-time data identifier in the data warehouse.

[0148] Continuous factor-level fused data sequences may contain outliers, typically caused by errors during data acquisition or interference during transmission. To ensure data quality, outlier removal is necessary for the factor-level fused data sequences. Outlier removal can employ statistical analysis methods, such as standard deviation-based methods or box plot-based methods, to identify and remove outliers from the sequence. The processed factor-level fused data sequences and their corresponding time-synchronized fused data are stored in a fused data warehouse. Simultaneously, the real-time data identifiers in the data warehouse are updated to indicate that the data is recently processed and updated, facilitating subsequent queries and applications.

[0149] Step S138: Integrate the new time-synchronized fusion data corresponding to all core interaction factors across factors, eliminate redundant information and conflicting data between factors, and form initial global coupled fusion data.

[0150] The new time-synchronized fused data is generated separately according to the core interaction factors. To obtain data that reflects the overall water quality status of the entire water quality correlation field, it is necessary to perform cross-factor integration of the time-synchronized fused data corresponding to all core interaction factors. During cross-factor integration, the fused data of different core interaction factors need to be comprehensively analyzed and processed to eliminate potentially redundant information between factors, i.e., duplicate or mutually substitutable information; simultaneously, it is necessary to resolve potentially conflicting data between factors. If the fused data of different core interaction factors provide contradictory descriptions of the same water quality status, further analysis and verification are needed to determine the correct data or to correct the data. After cross-factor integration, the initial globally coupled fused data is formed.

[0151] Step S139: Perform field-adaptive adjustments on the initial global coupled fusion data, combining the field boundary characteristics of the water quality correlation field and data flow rules to optimize the spatial distribution and time series characteristics of the data, and generate real-time updated coupled water quality data.

[0152] While the initial globally coupled and fused data integrates information from various core interaction factors, it may not yet be fully adapted to the field characteristics of the water quality correlation field. Therefore, field-adaptive adjustments are necessary. By considering the field boundary characteristics of the water quality correlation field, such as spatial and temporal boundaries, it is ensured that the initial globally coupled and fused data spatially conforms to the division of field units and temporally matches the temporal boundaries of these units. Simultaneously, based on the data flow rules of the water quality correlation field, the spatial distribution and temporal characteristics of the data are optimized to accurately reflect the data flow between field units and the transmission process of water quality influences. After field-adaptive adjustments, real-time updated coupled water quality data is generated, which accurately reflects the real-time water quality status of the entire water quality correlation field.

[0153] Step S140: Combine the field change trend data of the water quality correlation field with the coupled water quality data, perform the deduction operation of the water quality trend evolution path of the water-receiving lake, and generate a trend evolution path containing the continuous change characteristics of water quality status and key turning point information.

[0154] The field change trend data of the water quality correlation field reflects the direction and pattern of change of the entire water quality correlation field over time, while coupled water quality data provides real-time water quality status. Combining these two sets of data allows for the simulation of the water quality trend evolution path of the receiving lake. Based on the current water quality status and field change trends, the simulation uses established mathematical models or prediction algorithms to model the changes in the water quality of the receiving lake over a future period. The generated trend evolution path includes continuous water quality state change characteristics, i.e., continuous change curves or descriptions of water quality indicators over time; and key inflection point information, which are the points in time when the water quality state undergoes significant changes, marking the transition from one state to another.

[0155] Step S141: Extract the core water quality index sequence from the coupled water quality data and standardize the core water quality index sequence. The core water quality index sequence corresponds one-to-one with the core interaction factors in the water quality correlation field.

[0156] The coupled water quality data contains various water quality indicators, from which core water quality indicator sequences are extracted. Core water quality indicators refer to those that are representative and have a significant impact on the water quality of the affected lakes, such as dissolved oxygen, pH, ammonia nitrogen, total phosphorus, and total nitrogen. The core water quality indicator sequences correspond one-to-one with the core interaction factors in the water quality correlation field, because the core interaction factors are key to determining the interactions between field units, and the core water quality indicators are the direct reflection of the results of these interactions. The extracted core water quality indicator sequences are standardized to eliminate dimensional differences between different indicators, enabling them to be compared and analyzed under the same standard. Standardization can be performed using the aforementioned standardization methods to convert the core water quality indicator sequences into standardized sequence data.

[0157] Step S142: Analyze the temporal variation characteristics of the standardized core water quality index sequence, extract the temporal characteristic parameters of the rate of change, direction of change and magnitude of change in the standardized core water quality index sequence, and form a set of temporal characteristic parameters.

[0158] Temporal variation characteristic analysis of standardized core water quality indicator sequences aims to understand their dynamic changes over time. The analysis includes the rate of change (the amount of change per unit time), the direction of change (whether the indicator is trending upward, downward, or stable), and the magnitude of change (the degree to which the indicator deviates from its initial or average value). By extracting these temporal variation characteristics, a set of time-series characteristic parameters is formed. These parameters quantitatively describe the temporal variation characteristics of the core water quality indicator sequences.

[0159] Step S143: Extract field change trend data from the water quality correlation field. The field change trend data includes the expansion and contraction trends, interaction intensity change trends, and data flow rate change trends of each field unit.

[0160] The field change trend data of the water quality correlation field reflects the dynamic development of the entire field. From the water quality correlation field, we can extract the expansion and contraction trends of each field unit, i.e., whether the spatial range of the field unit is expanding or shrinking; the interaction intensity trend, i.e., whether the interaction intensity between field units is increasing or decreasing; and the data flow rate trend, i.e., whether the speed of data flow between field units is accelerating or slowing down. These trend data can be obtained by analyzing and fitting trends to historical data of the water quality correlation field, and they reflect the overall changes in the water quality correlation field from different perspectives.

[0161] Step S144: Perform correlation analysis between the time series characteristic parameter set and the field change trend data to determine the causal correspondence between the changes in core water quality indicators and the changes in the field.

[0162] By employing correlation analysis methods, such as correlation analysis and regression analysis, we can analyze the time-series characteristic parameter set and the field change trend data to determine the causal relationship between changes in core water quality indicators and changes in the field. For example, we can analyze whether an increase in the interaction intensity of field units leads to a faster rate of change in core water quality indicators, or whether the expansion of field units causes a change in the direction of change in core water quality indicators.

[0163] Step S145: Integrate relevant data from the causal relationship map to form a water quality trend evolution model. The water quality trend evolution model uses field change trend data as input variables and the change characteristics of core water quality indicators as output variables. The model parameters are calibrated based on historical correlation data.

[0164] The causal correlation map data contains information on the causal correspondence between changes in core water quality indicators and changes in the field. Integrating this data, a water quality trend evolution model is constructed. This model uses field change trend data as input variables, reflecting the field factors influencing water quality changes; and uses the change characteristics of core water quality indicators, such as the rate, direction, and magnitude of change, as output variables. The model parameters are calibrated using historical correlation data; that is, by using past field change trend data and corresponding core water quality indicator change characteristic data, the parameters in the model are adjusted so that the model can accurately simulate historical water quality changes. The calibrated water quality trend evolution model can be used to predict future changes in core water quality indicators based on current field change trend data.

[0165] Step S146: Input the current field change trend data into the water quality trend evolution model to calculate the recent prediction results of the core water quality indicators. The recent prediction results include the indicator change trajectory within a preset recent prediction time period.

[0166] After the water quality trend evolution model is constructed and calibrated, the current field change trend data is input into the model. Based on the input data and its built-in computational logic, the model performs calculations and analysis to obtain recent predictions of core water quality indicators. The recent prediction timeframe is pre-set, such as the next week or month. The recent prediction results include the change trajectory of the core water quality indicators within this timeframe, that is, the predicted value of the core water quality indicators at each point in time. These predicted values ​​are arranged chronologically to form a trajectory, intuitively showing the recent change trend of the core water quality indicators.

[0167] Step S147: Based on recent forecast results and combined with the future development trend characteristics of field change trends, deduce the long-term change trends of core water quality indicators. The long-term change trends include the phased characteristics and development direction of indicator changes within a preset long-term trend deduction time period, wherein the long-term trend deduction time period is longer than the recent forecast time period.

[0168] Recent forecasts reflect short-term changes in core water quality indicators. To understand longer-term water quality trends, it is necessary to deduce the long-term trends of core water quality indicators based on recent forecasts and the future development characteristics of field change trends. The future development characteristics of field change trends can be obtained through long-term trend analysis and forecasting of field change trend data. For example, based on the current expansion and contraction trends of field units and the changing trends of interaction intensity, the development direction and possible state over a relatively long period can be predicted. The time frame for long-term trend derivation is preset to be longer than that of recent forecasts, such as the next year or three years. Long-term trends mainly describe the stage-specific characteristics of core water quality indicators within the long-term trend derivation time frame, i.e., the characteristics of changes in different time periods, and the overall development direction, such as continuous improvement, gradual deterioration, or stabilization.

[0169] Step S148: By analyzing the extreme points in the core water quality index sequence and the abrupt change points in the field change trend, the potential key turning points of water quality change are determined. The potential key turning points correspond to the time nodes when the water quality state changes significantly.

[0170] Extreme points in the core water quality indicator series, such as maximum and minimum values, typically reflect that the water quality indicators have reached an extreme state at a certain point in time, potentially indicating a change in water quality status. Abrupt change points in the field change trend refer to the time points when the field change trend suddenly changes, such as a sudden increase or decrease in interaction intensity, or a sudden increase or decrease in data flow rate. By analyzing the extreme points in the core water quality indicator series and the abrupt change points in the field change trend, potential critical turning points in water quality changes can be identified. These potential critical turning points correspond to time points when the water quality status may change significantly. At these time points, water quality may change from one state to another, such as from a good state to a polluted state, or from a deteriorating trend to an improving trend.

[0171] Step S149: Based on recent prediction results, long-term change trends and key turning point information, construct the trend evolution path of the water quality of the receiving lake. The trend evolution path is based on time and records the continuous change characteristics of the water quality state and the specific information of key turning points.

[0172] By integrating recent forecasts, long-term trends, and key inflection points, a trend evolution path for the water quality of the receiving lakes is constructed. This path, structured around time, integrates the recent predicted indicator trajectories, the phased characteristics and development direction of long-term trends, and the specific time points and corresponding water quality changes at key inflection points. The trend evolution path records the continuous changes in water quality over time, such as the range and rate of change of water quality indicators, as well as significant changes in water quality at key inflection points. Through this trend evolution path, the overall development of the water quality of the receiving lakes over a future period can be clearly understood.

[0173] Step S1410: Combine the trend evolution path with the water quality protection target of the receiving lake and the operational constraints of the water diversion project, and mark the sections in the trend evolution path that exceed the protection target range or violate the operational constraints.

[0174] Receiving lakes typically have clearly defined water quality protection targets, such as the required water quality standards and limits for various pollutants. Water diversion projects also have operational constraints, such as maximum and minimum water diversion flows and the treatment capacity of water purification facilities. By combining the constructed trend evolution path with these water quality protection targets and operational constraints, the trend evolution path is evaluated. Sections on the trend evolution path where water quality indicators exceed the protection targets, and sections that may violate the operational constraints of the water diversion project, are marked. These marked sections serve as a reminder for relevant personnel to pay attention and take intervention measures to ensure that the water quality of the receiving lake meets the protection targets and that the water diversion project operates safely and stably under the constraints.

[0175] Step S150: Identify the core interaction unit from the water quality correlation field, combine the trend evolution path with the core interaction unit, and execute the generation operation of the water quality dynamic control scheme of the water diversion project. The water quality dynamic control scheme of the water diversion project includes control instructions acting on the core field interaction nodes that affect the water quality trend.

[0176] Core interaction units are the most critical and actively interacting field units in the water quality correlation field, exerting the greatest influence on water quality. These core interaction units are identified from the water quality correlation field; they are typically those that have strong interactions with multiple other field units and contribute significantly to changes in core water quality indicators. The trend evolution path is then combined with the core interaction units to analyze the impact of their interactions on the water quality trend evolution path. Based on the analysis results, a dynamic water quality control scheme for the water diversion project is generated. The purpose of the dynamic control scheme is to influence the interactions of the core interaction units by adjusting relevant facilities and parameters of the water diversion project, thereby guiding the water quality trend in a favorable direction. The control scheme includes control instructions acting on the core field interaction nodes that influence the water quality trend. These instructions specifically specify which core field interaction nodes need to be adjusted, as well as the method, extent, and timing of the adjustment.

[0177] For example, step S151: Based on the data association strength comparison table and the dynamically updated interaction strength comparison table, calculate the interaction influence degree and network topology importance index of each field unit. The interaction influence degree is the weighted sum of the interaction strength parameters of the field unit to all other units. The network topology importance index is calculated based on the connectivity and betweenness centrality of the field unit in the water quality association field network.

[0178] To identify core interaction units, the importance of each field unit needs to be assessed. Based on a data association strength comparison table and a dynamically updated interaction strength comparison table, the interaction influence degree of each field unit is calculated. Interaction influence degree refers to the weighted sum of the interaction strength parameters of a field unit on all other field units in the water quality association field. The weights can be determined based on the importance of the field unit or other factors; a higher interaction influence degree indicates a greater overall influence of the field unit on other units. Simultaneously, a network topology importance index for the field units is calculated. This index is based on the connectivity and betweenness centrality of the field unit in the water quality association field network. Connectivity refers to the number of connections between the field unit and other units; more connections indicate higher connectivity. Betweenness centrality refers to the frequency with which the field unit acts as an intermediary in the interaction between two other units; a higher betweenness centrality indicates a more important bridging role for the unit in the network. By combining the interaction influence degree and the network topology importance index, the importance of the field units can be comprehensively assessed.

[0179] Step S152: Set the interaction influence threshold and the network topology importance index threshold, and identify the field units that simultaneously satisfy the conditions that the interaction influence is greater than the interaction influence threshold and the network topology importance index is greater than the network topology importance index threshold as core interaction units.

[0180] Based on the actual situation of the water quality correlation field and the needs of water quality regulation, an interaction influence threshold and a network topology importance index threshold are set. The interaction influence threshold is a critical value for judging the magnitude of the interaction influence of field units, and the network topology importance index threshold is a critical value for judging the network topology importance of field units. The interaction influence degree of each field unit is compared with the interaction influence threshold, and the network topology importance index is compared with the network topology importance index threshold. Field units that simultaneously satisfy the condition that the interaction influence degree is greater than the interaction influence threshold and the network topology importance index is greater than the network topology importance index threshold are identified as core interaction units.

[0181] Step S153: Analyze the sensitive points of the core interaction unit in the trend evolution path. The sensitive points refer to the key nodes where a small change in the interaction intensity parameter of the core interaction unit can cause a significant shift in the trend evolution path.

[0182] Core interaction units exhibit several sensitive points in the trend evolution path, which are highly sensitive to changes in interaction intensity parameters. By analyzing the relationship between the interaction intensity parameters of core interaction units and the trend evolution path, we can identify key nodes—the sensitive points—where even minor changes in the interaction intensity parameters lead to significant deviations in the trend evolution path. For example, a slight increase in the interaction intensity parameter of a core interaction unit at a certain point in time could cause a water quality indicator in the trend evolution path to change from compliant to non-compliant. This point in time and the corresponding interaction intensity parameter value constitute the sensitive point.

[0183] Step S154: Based on the baseline value of the interaction intensity parameter of the sensitive action point and the target offset direction of the trend evolution path, calculate the adjustment amount of the control parameter of the core interaction unit. The adjustment amount of the control parameter is the amount of change in the interaction intensity parameter required to move the trend evolution path towards the target offset direction.

[0184] The baseline value of the interaction intensity parameter at the sensitive point refers to the numerical value of the interaction intensity parameter at the sensitive point under the current conditions. The target offset direction of the trend evolution path is the direction in which the trend evolution path is desired to change, such as shifting towards water quality improvement. Based on the baseline value of the interaction intensity parameter and the target offset direction, the adjustment amount of the control parameter of the core interaction unit is calculated. The adjustment amount of the control parameter refers to the amount of change in the interaction intensity parameter required to move the trend evolution path towards the target offset direction. During the calculation, it is necessary to analyze the relationship between the change of the interaction intensity parameter and the offset of the trend evolution path. Through the above relationship, the specific value of the interaction intensity parameter that needs to be adjusted in order to achieve the desired target offset direction and offset amount is determined.

[0185] Step S155: Based on the adjustment amount of the control parameters and the physical relationship of the core interaction unit, generate specific control instructions that act on the water diversion project facilities. The control instructions include facility identification, parameter adjustment type, adjustment range and timeliness requirements.

[0186] The core interaction units have a physical relationship with specific facilities in the water diversion project. For example, a core interaction unit may correspond to a specific pump, valve, or water purification device in the water diversion project. Based on the calculated adjustment amount of the control parameters and the physical relationship between the core interaction units and the water diversion project facilities, specific control instructions are generated. These control instructions include facility identifiers to identify the water diversion project facilities requiring adjustment; parameter adjustment types, such as increasing or decreasing pump flow, increasing or decreasing valve opening, or adjusting the operating parameters of the purification device; adjustment range, i.e., the specific value or proportion of the parameter to be adjusted; and time requirements, i.e., the timeframe within which the adjustment operation must be completed and the duration for which the adjusted parameters must be maintained. These control instructions are specific and clear, and can be directly used to guide the operation and adjustment of the water diversion project.

[0187] Step S156: Integrate the control commands of all core interactive units according to time sequence and spatial correlation to form a dynamic water quality control scheme for water diversion projects that includes multi-level control priorities and collaborative control strategies.

[0188] The control commands from different core interaction units may have a temporal sequence and spatial relationships. This paper sorts all the control commands from the core interaction units according to their time sequence to determine the execution order. Simultaneously, based on the spatial relationships between the core interaction units, such as adjacent units or those with data flow relationships, related control commands are integrated. During the integration process, multi-level control priorities are set, assigning higher priority to control commands that have a more critical impact on water quality trends and require urgent execution, and lower priority to relatively less important commands. Furthermore, a coordinated control strategy is formulated to ensure that the control commands from different core interaction units can cooperate and work synergistically, avoiding contradictions or cancellations. Through integration, a dynamic water quality control scheme for the water diversion project is formed. This scheme can effectively guide the water quality control work of the water diversion project and achieve dynamic management of the water quality of the receiving lakes.

[0189] In one exemplary embodiment, a water quality analysis system for receiving lakes in a water diversion project, incorporating multi-source data fusion, is provided. This system can be a terminal, server, etc., and its internal structure diagram can be as follows: Figure 2As shown, this water quality analysis system for the receiving lake in a water diversion project, which integrates multi-source data fusion, includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computational and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides the environment for the operation of the operating system and computer programs in the non-volatile storage medium. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, near-field communication, or other technologies. When the computer program is executed by the processor, it implements a water quality analysis method for the receiving lake in a water diversion project that integrates multi-source data fusion. The display unit is used to generate a visually visible image and can be a display screen, projection device, or virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device can be a touch layer covering the display screen, or a button, trackball, or touchpad set on the shell of the water quality analysis system of the receiving lake in a water diversion project that combines multi-source data fusion, or an external keyboard, touchpad, or mouse, etc.

[0190] It should be noted that, in order to simplify the description of the present invention and thus help to understand one or more embodiments of the invention, multiple features may sometimes be grouped into one embodiment, drawing or description thereof in the foregoing description of the embodiments of the present invention.

Claims

1. A method for water quality analysis of receiving lakes in water diversion projects that combines multi-source data fusion, characterized in that, The method includes: Collect basic field data, define the dynamic interaction relationship between the data in the basic field data, and construct a water quality correlation field between the water diversion project and the receiving lake. The water quality correlation field presents the field effect of water quality impact through the dynamic interaction relationship between the data. The basic field data includes dynamic output data of the water diversion source, real-time interactive data of the water conveyance process, in-situ sensing data of the receiving lake, and dynamic correlation data of the surrounding environment. The water quality correlation field divides the field into field units by grouping data with correlation strength higher than a preset threshold into the same field unit. Each field unit corresponds to a core water quality influencing factor and a set of correlation data. An interaction relationship modeling operation is performed on each field unit in the water quality correlation field to obtain a field unit interaction model. The field unit interaction model is used to quantitatively describe the transmission mode, transmission intensity and dynamic change characteristics of water quality influence between different field units. Based on the interaction relationship described by the field unit interaction model, dynamic coupling and fusion processing is performed on the field basic data of the water quality correlation field to generate real-time updated coupled water quality data. The field change trend data of the water quality correlation field is combined with the coupled water quality data to perform the deduction operation of the water quality trend evolution path of the water-receiving lake, generating a trend evolution path containing the continuous change characteristics of water quality status and key turning point information. The field change trend data is extracted from the water quality correlation field and includes the expansion and contraction trend, interaction intensity change trend and data flow rate change trend of each field unit. Identify core interaction units from the water quality correlation field, combine the trend evolution path with the core interaction units, and generate a dynamic water quality control scheme for the water diversion project. The dynamic water quality control scheme for the water diversion project includes control instructions that act on the core field interaction nodes that affect the water quality trend.

2. The method for water quality analysis of receiving lakes in water diversion projects based on multi-source data fusion as described in claim 1, characterized in that, The process of collecting basic field data, defining the dynamic interaction relationships between data within the basic field data, and constructing a water quality correlation field between the water diversion project and the receiving lake includes: Dynamic output data from the water source is collected at fixed time intervals, and the collection time information for each collection is recorded. The dynamic output data from the water source includes water quality composition data, flow rate change data, and output status data during the water supply process. Real-time interactive data of the water conveyance process is collected in segments according to the water conveyance process, and the collection location information of each collection is marked. The real-time interactive data of the water conveyance process includes the interaction data between the water conveyance pipeline and the water body, the water quality exchange data along the water conveyance route, and the associated data of the operation status of the water conveyance facilities. In-situ sensing data of the receiving lake is collected synchronously at multiple points using distributed sensing devices. The in-situ sensing data of the receiving lake includes water quality parameter data of different areas of the lake, water flow data of different areas of the lake, and lake ecological correlation data of different areas of the lake. Dynamic correlation data of the surrounding environment are collected according to the type of environmental impact. The dynamic correlation data of the surrounding environment includes atmospheric environmental data, surface runoff data, vegetation cover data and human activity correlation data. The dynamic output data of the water source, the real-time interactive data of the water conveyance process, the in-situ sensing data of the receiving lake, and the dynamic correlation data of the surrounding environment are classified and mapped according to data dimensions to establish the correspondence between data dimensions and water quality influencing factors, forming a field basic data dimension mapping table. Based on the field basic data dimension mapping table, data from different sources mapped to the same water quality influencing factor are standardized and preprocessed to generate standardized field basic data. Based on the standardized field basic data, the water quality influence correlation strength between data from different sources is calculated to form a data correlation strength comparison table. The correlation strength is calculated based on the contribution of standardized data to the same water quality influencing factor. Based on the data correlation strength comparison table, the operation of establishing the field unit division criteria is performed, and data with correlation strength higher than the preset threshold are classified into the same field unit, so that each field unit corresponds to a core water quality influencing factor and a set of related data. The spatial boundary of each field unit is defined based on the physical location range of data acquisition, and the temporal boundary of each field unit is defined based on the temporal synchronization range of data acquisition, thus forming the boundary characteristics of the field unit. Each field unit is spatially arranged and time-synchronized according to the correlation relationship in the data correlation strength comparison table to construct an initial water quality correlation field framework. The initial water quality correlation field framework includes the distribution of field units, the correlation channels between units, and the data flow path. The basic field data is divided into units according to the boundary features of the field units and the initial water quality correlation field framework, and then filled into the corresponding field units and correlation channels. Dynamic rules for data flow between units are added to the correlation channels to form the water quality correlation field.

3. The method for water quality analysis of receiving lakes in water diversion projects based on multi-source data fusion as described in claim 2, characterized in that, The step of performing interaction relationship modeling operations on each field unit in the water quality correlation field to obtain a field unit interaction model includes: Extract the core data features of each field unit from the water quality correlation field. The core data features are key data representations that can represent the water quality impact attributes of the field unit. One field unit corresponds to a set of core data features. After the core data features are extracted, the overlapping features between the core data feature sets of adjacent field units are analyzed to determine the water quality influencing factors that coexist between adjacent units. Based on the common water quality influencing factors, the interaction types between adjacent field units are identified. The interaction types are enhancement interaction, inhibition interaction, and neutral interaction. The identification of the interaction type is based on the change response relationship of core data features. Calculate the interaction strength parameter corresponding to each type of interaction. The interaction strength parameter is determined based on the difference in contribution of the standardized common influence factors and the corresponding value in the data association strength comparison table to form an initial interaction strength comparison table. Monitor the changes in core data characteristics of each field unit in the water quality correlation field, capture the real-time fluctuations of core data characteristics, and establish a data characteristic fluctuation sequence; Based on the data feature fluctuation sequence, the interaction intensity parameters in the initial interaction intensity lookup table are dynamically adjusted to match the interaction intensity parameters with the amplitude and frequency of the data feature fluctuations, thus forming a dynamically updated interaction intensity lookup table. Analyze the indirect interaction relationships between non-directly adjacent field units through intermediate field units, derive the indirect interaction strength parameters based on the direct interaction strength parameters, and supplement the indirect interaction strength parameters into the dynamically updated interaction strength lookup table. After supplementing the indirect interaction relationships, the transmission delay characteristics of the interaction of each field unit are determined. The transmission delay characteristics are determined based on the spatial distance of the field units, the length of the data flow path, and the characteristics of the transmission medium, forming a set of transmission delay parameters. By integrating the interaction types, the dynamically updated interaction intensity lookup table, and the transmission delay parameter set, a field unit interaction model is constructed. After the field unit interaction model is constructed, the interaction process of each field unit is simulated through the field unit interaction model to verify the degree of fit between the interaction results output by the field unit interaction model and the actual data changes in the water quality correlation field. Based on the degree of fit, the model parameters are optimized to form a stable interaction relationship modeling result.

4. The method for water quality analysis of receiving lakes in water diversion projects based on multi-source data fusion as described in claim 1, characterized in that, The process of performing dynamic coupling and fusion processing on the field-based data of the water quality correlation field based on the interaction relationship described by the field unit interaction model to generate real-time updated coupled water quality data includes: Extract core interaction factors from the interaction relationships. These core interaction factors are key data items that determine the intensity and direction of interaction between field units. Each interaction type corresponds to a set of core interaction factors. Integrate the rules related to the core interaction factors to form a coupling fusion operator, which includes interaction intensity weight allocation rules, data complementarity fusion rules and dynamic adjustment rules; The field basic data of each field unit in the water quality correlation field are classified and extracted according to the core interaction factor to form a factor correlation data subset. One factor correlation data subset corresponds to one core interaction factor and correlation field unit data. The factor-related data subset is input into the corresponding coupling fusion operator, and the original data in the data subset is standardized and preprocessed. Fusion weights are assigned to the standardized data according to the interaction strength weight allocation rule, so that the assigned weight values ​​are consistent with the corresponding interaction strength parameters in the dynamically updated interaction strength lookup table. Fusion calculation is performed on the standardized data with assigned fusion weights according to the data complementarity fusion rule, integrating the complementary information about the same core interaction factor in the data of different field units to generate factor-level fused data. According to the data complementarity and fusion rules, the data with assigned fusion weights are fused and calculated to integrate the complementary information about the same core interaction factor in the data of different field units and generate factor-level fused data. Delay information is extracted from the set of transmission delay parameters, and time synchronization calibration is performed on the factor-level fusion data to ensure that the fusion data of different field units are consistent in the time dimension, thereby generating time-synchronized fusion data. The dynamic changes of the interaction relationship are monitored. When the parameters in the dynamically updated interaction strength lookup table are adjusted, the parameter update of the coupling fusion operator is triggered, so that the weight allocation rule and fusion rule of the coupling fusion operator follow the changes of the interaction. Based on the updated coupling fusion operator, the subsequently collected field basic data are fused in real time to generate new factor-level fusion data and new time-synchronized fusion data. The new time-synchronized fusion data corresponding to all core interaction factors are integrated across factors to eliminate redundant information and conflicting data between factors and form initial global coupled fusion data. The initial global coupled fusion data is adjusted for field adaptability. By combining the field boundary characteristics of the water quality correlation field and the data flow rules, the spatial distribution and time series characteristics of the data are optimized to generate real-time updated coupled water quality data.

5. The method for water quality analysis of receiving lakes in water diversion projects based on multi-source data fusion as described in claim 2, characterized in that, Based on the field basic data dimension mapping table, standardized preprocessing is performed on data from different sources mapped to the same water quality influencing factor to generate standardized field basic data. Based on standardized field data, the correlation strength of water quality impacts among data from different sources is calculated, forming a data correlation strength comparison table, including: All water quality influencing factors are extracted from the field basic data dimension mapping table to form a water quality influencing factor set, and each water quality influencing factor corresponds to at least two data dimensions from different sources. For each water quality influencing factor in the set of water quality influencing factors, the original data associated with the corresponding data dimension is extracted, and the original data is standardized and preprocessed to form a standardized dedicated data set for that water quality influencing factor. The standardized dedicated data set contains relevant standardized data records from different sources. The sensitivity of each standardized data record in the standardized dedicated dataset to water quality influencing factors is analyzed. The sensitivity is determined based on the response relationship between the change in the standardized data record and the change in the standardized value of the water quality influencing factor. Based on the representation sensitivity of each standardized data record, all standardized data records in the standardized dedicated data set are prioritized to form a data priority sequence; Based on the data priority sequence, a basic contribution weight is assigned to each data source. The basic contribution weight of the data source with higher priority is set to be higher than that of the data source with lower priority. At the same time, dynamic adjustment space is reserved for the basic contribution weight to adapt to data changes. The collaborative contribution coefficient between different data sources under the same water quality influencing factor is calculated. The collaborative contribution coefficient is determined based on the synchronicity and complementarity of the standardized data changes between the data sources. The collaborative contribution coefficient of the data source with prominent synchronicity and complementarity is set to be higher than that of the data source with insignificant synchronicity and complementarity. By combining the basic contribution weight with the collaborative contribution coefficient, the overall contribution of each data source to the corresponding water quality impact factor is calculated. An initial comparison table framework is constructed with water quality influencing factors as rows and data sources as columns. The overall contribution of each data source to the corresponding water quality influencing factor is filled into the corresponding position of the initial comparison table framework. Normalize all comprehensive contribution levels in the initial comparison table framework, bind the normalized comparison table with the source identifier and data dimension identifier of the field basic data, supplement the metadata information of the normalized comparison table, and form a data association strength comparison table.

6. The method for water quality analysis of receiving lakes in water diversion projects based on multi-source data fusion as described in claim 3, characterized in that, The step of dynamically adjusting the interaction intensity parameters in the initial interaction intensity lookup table based on the data feature fluctuation sequence, so that the interaction intensity parameters match the amplitude and frequency of the data feature fluctuations, and forming a dynamically updated interaction intensity lookup table, includes: Fluctuation features are extracted from the standardized data feature fluctuation sequence, including key fluctuation parameters such as fluctuation amplitude, fluctuation frequency, fluctuation period, and fluctuation trend. The core data features of each field unit correspond to a set of fluctuation parameters based on the standardized data. By integrating the physical laws of water quality impact transmission and historical data verification results, a mapping relationship model between fluctuation parameters and interaction intensity parameters is formed. The mapping relationship model records the adjustment direction and adjustment magnitude of the interaction intensity parameters corresponding to different combinations of fluctuation parameters. The fluctuation parameters of each field unit are input into the mapping relationship model to obtain the interaction intensity parameter adjustment suggestions for each field unit. The adjustment suggestions include the parameter adjustment amount and the adjusted target interaction intensity parameters. Extract the current interaction intensity parameters between the corresponding field units from the initial interaction intensity comparison table, compare the current parameters with the target interaction intensity parameters in the adjustment suggestions, and determine the actual adjustment difference; According to the actual adjustment difference, the corresponding parameters in the initial interaction intensity comparison table are adjusted point by point. First, the direct interaction intensity parameters between adjacent field units are adjusted, and then the adjustment amount of the indirect interaction intensity parameters is derived based on the adjustment results of the direct parameters. The adjusted interaction intensity comparison table is internally consistent to ensure that the interaction intensity parameters between the same field unit and different adjacent units are logically unified, and that the interaction intensity parameters at different levels are mutually compatible. Based on the overall stability requirements of the water quality correlation field, an adjustment threshold for the interaction intensity parameter is set. When the single adjustment amount exceeds the adjustment threshold, the adjustment operation is performed in stages. Record the time point, reason for adjustment, and adjustment range of each parameter adjustment to form a parameter adjustment log; Align the adjusted interaction intensity comparison table with the timestamps of the data characteristic fluctuation sequence, integrate the adjusted interaction intensity comparison tables of all time nodes, and form a dynamically updated interaction intensity comparison table sequence.

7. The method for water quality analysis of receiving lakes in water diversion projects based on multi-source data fusion as described in claim 4, characterized in that, The updated coupling fusion operator performs real-time fusion processing on subsequently acquired field-based data to generate new factor-level fusion data and new time-synchronized fusion data, including: Establish a real-time interactive link for basic field data, so that the subsequently collected basic field data can be transmitted to the fusion processing module within a preset time limit. Set up a data receiving buffer in the fusion processing module to temporarily store and format the real-time transmitted basic field data. Extract weight allocation rules and fusion rules from the updated coupled fusion operators and load them into the real-time fusion processing engine. Real-time data is extracted from the data receiving buffer according to the core interaction factors, and the real-time data is standardized and preprocessed to form a standardized real-time factor-related data subset. The standardized real-time factor-related data subset is input into the real-time fusion processing engine, and real-time fusion weights are assigned to each standardized data item in the standardized real-time factor-related data subset according to the updated weight allocation rules. The assigned weight values ​​are synchronized with the latest dynamic interaction intensity lookup table parameters. According to the updated fusion rules, complementary fusion calculations are performed on the data assigned real-time fusion weights to integrate the effective information in the data and generate real-time factor-level fusion data. The latest delay information in the set of transmission delay parameters is retrieved, and the real-time factor-level fusion data is time-synchronized and calibrated. The time-synchronized and calibrated real-time factor-level fusion data is then time-series spliced ​​with the historical factor-level fusion data to form a continuous factor-level fusion data sequence. Outlier removal is performed on continuous factor-level fused data sequences to remove abnormal data points caused by data acquisition errors or transmission interference. The processed factor-level fused data sequences and their corresponding time-synchronized fused data are then stored in the fused data warehouse, and the real-time data identifiers in the data warehouse are updated.

8. The method for water quality analysis of receiving lakes in water diversion projects based on multi-source data fusion as described in claim 1, characterized in that, The step of combining the field change trend data of the water quality correlation field with the coupled water quality data to perform a deduction operation on the water quality trend evolution path of the receiving lake, generating a trend evolution path containing continuous change characteristics of water quality status and key turning point information, including: The core water quality index sequence is extracted from the coupled water quality data and standardized. The core water quality index sequence corresponds one-to-one with the core interaction factor in the water quality correlation field. Analyze the temporal variation characteristics of the standardized core water quality index sequence, and extract the temporal characteristic parameters of the rate of change, direction of change and magnitude of change in the standardized core water quality index sequence to form a set of temporal characteristic parameters; Extract field change trend data from the water quality correlation field. The field change trend data includes the expansion and contraction trend, interaction intensity change trend, and data flow rate change trend of each field unit. By performing correlation analysis between the set of time-series characteristic parameters and the field change trend data, the causal correspondence between the changes in core water quality indicators and field changes can be determined. By integrating relevant data from the causal relationship map, a water quality trend evolution model is formed. The water quality trend evolution model takes the field change trend data as input variables and the change characteristics of core water quality indicators as output variables. The model parameters are calibrated based on historical correlation data. The current field change trend data is input into the water quality trend evolution model to calculate the recent prediction results of the core water quality indicators. The recent prediction results include the indicator change trajectory within a preset recent prediction time period. Based on recent forecast results and combined with the future development trend characteristics of field change trends, the long-term change trends of core water quality indicators are deduced. The long-term change trends include the phased characteristics and development direction of indicator changes within a preset long-term trend deduction time period, wherein the long-term trend deduction time period is longer than the recent forecast time period. By analyzing the extreme points in the core water quality index sequence and the abrupt change points in the field change trend, potential key turning points of water quality change are identified. These potential key turning points correspond to the time nodes when the water quality status changes significantly. Based on recent forecast results, long-term trends and key turning points, a trend evolution path for the water quality of the receiving lake is constructed. The trend evolution path is based on time and records the continuous change characteristics of the water quality status and the specific information of key turning points. By combining the trend evolution path with the water quality protection targets of the receiving lake and the operational constraints of the water diversion project, sections that exceed the protection target range or violate the operational constraints are marked in the trend evolution path.

9. The method for water quality analysis of receiving lakes in water diversion projects based on multi-source data fusion as described in claim 2, characterized in that, The step of spatially arranging and temporally synchronizing each field unit according to the correlation relationship in the data correlation strength lookup table to construct an initial water quality correlation field framework includes: Extract the correlation strength value corresponding to each field unit in the data correlation strength comparison table. The correlation strength value is calculated based on the standardized field basic data. Sort the correlation relationships between field units according to the magnitude of the correlation strength value to form a correlation relationship priority sequence. Based on the priority sequence of association relationships, the spatial layout of field units is established, and field units with high association strength values ​​are preferentially arranged in an adjacent manner in space. Based on the physical layout of the water diversion project, the geographical zoning of the receiving lakes, and the distribution characteristics of the surrounding environment, the actual spatial coordinates of each field unit are determined. Based on the spatial layout and actual spatial coordinates, each field unit is placed in the virtual field space. Based on the correlation relationship in the correlation strength comparison table, a correlation channel is built between adjacent field units. The width of the correlation channel is proportional to the correlation strength value. Configure data transmission rules for each associated channel. The data transmission rules are determined based on the type and strength of the association relationship, and determine the transmission direction, transmission rate and transmission priority of data in the associated channel. Extract timestamp information from the basic field data, analyze the time synchronization of data from different sources, determine the time synchronization benchmark, and adjust the timestamp of the basic data in each field unit based on the time synchronization benchmark so that the data of different field units in the same period are aligned in the time dimension. According to the data flow requirements of the field units after time synchronization, and combined with the transmission rules of the associated channels, plan the data flow path between field units. By integrating the spatial distribution of field units, the correlation channels between field units, and the planned data flow paths, an initial water quality correlation field framework is formed.

10. A water quality analysis system for receiving lakes in a water diversion project that combines multi-source data fusion, characterized in that, include: processor; A machine-readable storage medium for storing machine-executable instructions of the processor; The processor is configured to execute the water quality analysis method for receiving lakes in a water diversion project that combines multi-source data fusion, as described in any one of claims 1 to 9, by executing the machine-executable instructions.

Images