River comprehensive management and control data collaborative optimization method and system based on Internet of Things

Abstract

The invention relates to the technical field of data processing, in particular to a river comprehensive management and control data collaborative optimization method and system based on the Internet of Things. The method comprises the following steps: firstly, acquiring multi-source heterogeneous data, an analysis demand report and a river state evaluation report acquired by an Internet of Things sensor, preprocessing to obtain multi-source preprocessed data, an analysis demand vector and a river state evaluation vector, and inputting the multi-source preprocessed data, the analysis demand vector and the river state evaluation vector into a dynamic multi-source heterogeneous data fusion module based on an attention mechanism to obtain dynamic fusion data; then, inputting the dynamic fusion data into a cross-domain associated riverway state collaborative prediction module to obtain a collaborative prediction result of the riverway state; and finally, inputting the dynamic fusion data and the collaborative prediction result into a multi-target collaborative optimization module, dynamically adjusting the target weight and the decision priority according to the module input, solving by using a multi-target optimization algorithm, obtaining an optimization decision parameter, and performing optimization according to the optimization decision parameter. The method can effectively improve the management decision level of the river data.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and specifically provides a method and system for collaborative optimization of river comprehensive control data based on the Internet of Things. Background Art

[0002] With the development of Internet of Things technology, it has become possible to deploy a large number of sensors in river management, which provides a data basis for obtaining comprehensive and real-time river state information. However, how to effectively process and utilize these massive, multi-source heterogeneous data to support scientific and efficient river comprehensive management decisions remains an important challenge faced by current data processing technologies.

[0003] In the prior art, the processing of river data often has the following deficiencies: on the one hand, for the Internet of Things river data with diverse sources, different formats, and uneven quality, it is difficult for the prior art to dynamically adjust the importance of each data source according to the real-time changing river conditions or management priorities, resulting in insufficient relevance and effectiveness of the fusion results and affecting the accuracy of subsequent analysis and decision-making; on the other hand, the prior art uses multi-objective optimization methods based on fixed weights or rules, which are difficult to adaptively adjust optimization objectives, constraints, and priorities according to the dynamic risks and prediction results revealed by real-time data, and lack accurate prediction of cross-domain impacts; therefore, there is an urgent need to propose a new data processing method that can dynamically fuse multi-source heterogeneous Internet of Things river data, achieve cross-domain associated collaborative prediction, and perform adaptive multi-objective collaborative optimization decisions based on this. Summary of the Invention

[0004] The purpose of the present invention is to provide a method and system for collaborative optimization of river comprehensive control data based on the Internet of Things. First, obtain multi-source heterogeneous data, analysis requirement reports, and river state assessment reports collected by Internet of Things sensors. After preprocessing, obtain multi-source preprocessed data, analysis requirement vectors, and river state assessment vectors, and input them into a dynamic multi-source heterogeneous data fusion module based on the attention mechanism to obtain dynamically fused data; then, input the dynamically fused data into a cross-domain associated river state collaborative prediction module to obtain a collaborative prediction result of the river state; finally, input the dynamically fused data and the collaborative prediction result into a multi-objective collaborative optimization module, dynamically adjust the target weights and decision priorities according to the module input, use a multi-objective optimization algorithm to solve, obtain optimization decision parameters, and perform optimization according to the optimization decision parameters; the present invention can effectively improve the management decision-making level of river data.

[0005] To achieve the above object, the present invention provides the following technical solutions: A method for collaborative optimization of river comprehensive control data based on the Internet of Things, including: Obtain multi-source heterogeneous data, analysis requirement reports, and river state assessment reports collected by Internet of Things sensors; Preprocess the multi-source heterogeneous data, the analysis requirement report, and the river channel state evaluation report to obtain multi-source preprocessed data, an analysis requirement vector, and a river channel state evaluation vector; Input the multi-source preprocessed data, the analysis requirement vector, and the river channel state evaluation vector into the dynamic multi-source heterogeneous data fusion module based on the attention mechanism to generate dynamic fusion data; Input the dynamic fusion data into the cross-domain associated river channel state collaborative prediction module to obtain the collaborative prediction result of the river channel state; Construct a multi-objective collaborative optimization module, input the dynamic fusion data and the collaborative prediction result into the multi-objective collaborative optimization module, dynamically adjust the objective weights and decision priorities according to the dynamic fusion data and the collaborative prediction result, solve using a multi-objective optimization algorithm, obtain the optimized decision parameters, and perform optimization according to the optimized decision parameters.

[0006] Further, the multi-source heterogeneous data includes: river channel hydrological state data, river channel water quality state data, and river channel engineering state data.

[0007] Further, the process of preprocessing the multi-source heterogeneous data, the analysis requirement report, and the river channel state evaluation report to obtain multi-source preprocessed data, an analysis requirement vector, and a river channel state evaluation vector includes: Perform data cleaning, format standardization, and spatio-temporal alignment on the multi-source heterogeneous data to obtain the multi-source preprocessed data; Perform text cleaning, text analysis, and vectorization on the analysis requirement report and the river channel state evaluation report respectively to obtain the analysis requirement vector and the river channel state evaluation vector.

[0008] Further, the process of inputting the multi-source preprocessed data, the analysis requirement vector, and the river channel state evaluation vector into the dynamic multi-source heterogeneous data fusion module based on the attention mechanism to generate dynamic fusion data includes: Input the multi-source preprocessed data, the analysis requirement vector, and the river channel state evaluation vector into the data embedding layer of the dynamic multi-source heterogeneous data fusion module to obtain multi-source embedding vectors, an analysis requirement embedding vector, and a river channel state evaluation embedding vector; Input the multi-source embedding vectors, the analysis requirement embedding vector, and the river channel state evaluation embedding vector into the attention weighting layer of the dynamic multi-source heterogeneous data fusion module to obtain multi-source weighted feature vectors; Input the multi-source weighted feature vectors, the analysis requirement embedding vector, and the river channel state evaluation embedding vector into the comprehensive feature fusion layer of the dynamic multi-source heterogeneous data fusion module to obtain fusion feature vectors; Input the fused feature vector into the output layer of the dynamic multi-source heterogeneous data fusion module to obtain the dynamic fusion data.

[0009] Further, the process of inputting the dynamic fusion data into the cross-domain associated river channel state collaborative prediction module to obtain the collaborative prediction result of the river channel state includes: Define the nodes and edges in the graph structure, and construct the node feature vectors based on the dynamic fusion data; Use the historical dynamic fusion data to train the river channel state collaborative prediction module, perform graph data processing and river channel state prediction according to the spatio-temporal graph convolutional layer and the prediction output layer in the river channel state collaborative prediction module respectively, and obtain the pre-trained river channel state collaborative prediction module; Input the dynamic fusion data into the pre-trained river channel state collaborative prediction module to obtain the collaborative prediction result of the river channel state.

[0010] Further, the process of inputting the dynamic fusion data and the collaborative prediction result into the multi-objective collaborative optimization module, dynamically adjusting the objective weights and decision priorities according to the dynamic fusion data and the collaborative prediction result, and using the multi-objective optimization algorithm to solve and obtain the optimized decision parameters includes: Define the objective function and constraint conditions of the multi-objective collaborative optimization module, and determine the decision variables; Input the dynamic fusion data and the collaborative prediction result into the multi-objective collaborative optimization module, and dynamically adjust the objective weights and decision priorities according to the dynamic fusion data and the collaborative prediction result; Combine the adjusted objective weights and the decision priorities, and use the multi-objective optimization algorithm to solve the optimization problem to obtain the optimized decision parameters.

[0011] An Internet of Things-based collaborative optimization system for river comprehensive management and control data includes: a data collection unit, a data processing unit, a data dynamic fusion unit, a river channel state collaborative prediction unit, and a multi-objective collaborative optimization unit; The data collection unit is used to obtain multi-source heterogeneous data, analysis requirement reports, and river channel state assessment reports collected by Internet of Things sensors; The data processing unit is used to preprocess the multi-source heterogeneous data, the analysis requirement reports, and the river channel state assessment reports to obtain multi-source preprocessed data, analysis requirement vectors, and river channel state assessment vectors; The data dynamic fusion unit is used to input the multi-source preprocessed data, the analysis requirement vectors, and the river channel state assessment vectors into the dynamic multi-source heterogeneous data fusion module based on the attention mechanism to generate dynamic fusion data; The river channel state collaborative prediction unit is used to input the dynamic fusion data into a river channel state collaborative prediction module associated across domains to obtain a collaborative prediction result of the river channel state; The multi-objective collaborative optimization unit is used to input the dynamic fusion data and the collaborative prediction result into a multi-objective collaborative optimization module, dynamically adjust the target weights and decision priorities according to the dynamic fusion data and the collaborative prediction result, solve using a multi-objective optimization algorithm, obtain optimized decision parameters, and perform optimization according to the optimized decision parameters.

[0012] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. The present invention proposes a multi-source data dynamic fusion method for obtaining dynamic fusion data; this method inputs multi-source preprocessed data, an analysis requirement vector, and a river channel state evaluation vector into a dynamic multi-source heterogeneous data fusion module, dynamically assigns feature weights using an attention weighting layer to highlight key features related to analysis requirements and state evaluation, and uses a comprehensive feature fusion layer to comprehensively fuse the weighted multi-source data, analysis requirements, and river channel state evaluation information; this method can adaptively adjust the data fusion method according to different analysis requirements and the changing river channel state to meet actual needs, providing data support for subsequent river channel state prediction and decision optimization.

[0013] 2. The present invention proposes a river channel state collaborative prediction method for obtaining a river channel state prediction result; this method uses the spatio-temporal graph convolutional layer in the river channel state collaborative prediction module to simultaneously capture the spatial dependence relationship in the graph structure and the time dependence relationship in the time series, which enables the module to learn a richer and more comprehensive river channel state feature representation; this method obtains a more accurate river channel state prediction result by capturing and utilizing the internal correlation between hydrological, water quality, and engineering field data, which is beneficial for subsequent comprehensive management data collaborative optimization.

[0014] 3. The present invention proposes a multi-objective collaborative optimization method for effectively adjusting river channel data decisions; this method combines dynamic fusion data, collaborative prediction results, and a multi-objective collaborative optimization module, dynamically adjusts target weights and decision priorities using the dynamic fusion data and the collaborative prediction result to achieve a more flexible optimization strategy adjustment; solving using a multi-objective optimization algorithm can achieve the balance and coordination between various objectives; this method realizes the intelligence and refinement of river channel comprehensive data decisions, effectively improving the river channel comprehensive management level. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a flowchart of a method for collaborative optimization of river channel comprehensive management data based on the Internet of Things according to the present invention; Figure 2Schematic diagram of the dynamic multi-source heterogeneous data fusion module based on the attention mechanism of the present invention; Figure 3 Schematic diagram of a river comprehensive management and control data collaborative optimization system based on the Internet of Things according to the present invention. Specific implementation manners

[0016] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0017] Please refer to Figures 1 to 3 The present invention provides a river comprehensive management and control data collaborative optimization method and system based on the Internet of Things, and the technical solutions are as follows:

[0018] Embodiment 1: In order to improve the level of river comprehensive management, an enterprise uses a river comprehensive management and control data collaborative optimization method proposed by the present invention. The process schematic of this method can be referred to Figure 1 and specifically includes: Obtain multi-source heterogeneous data, analysis requirement reports, and river state assessment reports collected by Internet of Things sensors; Furthermore, the multi-source heterogeneous data includes: river hydrological state data, river water quality state data, and river engineering state data; Furthermore, the river hydrological state data includes: water level, flow rate, velocity, rainfall, water temperature, etc.; the river water quality state data includes: dissolved oxygen, biochemical oxygen demand, ammonia nitrogen, pH value, turbidity, etc.; the river engineering state data mainly refers to the state data of engineering facilities such as dams, sluices, and reservoirs; Furthermore, the analysis requirement report includes: analysis theme, analysis objective, analysis scope, analysis indicators, etc.; the river state assessment report includes: report title, assessment scope, assessment indicators, etc.

[0019] By using multi-source data in three different fields of hydrology, water quality, and engineering, it provides solid data support for subsequent data fusion, state prediction, and multi-objective collaborative optimization; at the same time, there is a correlation between data in different fields, and they affect each other, which provides an objective factual basis for the interactive processing and collaborative analysis of data.

[0020] Preprocess the multi-source heterogeneous data, analysis requirement reports, and river state assessment reports to obtain multi-source preprocessed data, analysis requirement vectors, and river state assessment vectors; Further, the process of preprocessing multi-source heterogeneous data, analysis requirement reports, and river channel state evaluation reports to obtain multi-source preprocessed data, analysis requirement vectors, and river channel state evaluation vectors includes: Perform data cleaning, format standardization, and spatio-temporal alignment on the multi-source heterogeneous data to obtain multi-source preprocessed data; Respectively perform text cleaning, text analysis, and vectorization on the analysis requirement report and the river channel state evaluation report to obtain analysis requirement vectors and river channel state evaluation vectors; Further, the data cleaning process includes: missing value processing, outlier processing, and noise processing; format standardization includes unifying data types, units, date formats, etc.; spatio-temporal alignment operations use timestamp alignment to ensure that the timestamps of multi-source data are at the same moment, and use geographic coordinate transformation and spatial matching operations for spatial alignment; Further, the text cleaning process includes: removing useless characters, case conversion, and spelling correction; the text analysis process includes: word segmentation, part-of-speech tagging, named entity recognition, and keyword extraction; the vectorization process uses the bag-of-words model to convert text data into vector form.

[0021] By processing the multi-source heterogeneous data through data cleaning, format standardization, and spatio-temporal alignment to ensure the consistency and comparability of the data, the preprocessing vector results of the analysis requirement report and the river channel state evaluation report can provide a basis for subsequent dynamic data fusion, thus ensuring the effectiveness of river channel state prediction and data decision optimization.

[0022] Input the multi-source preprocessed data, analysis requirement vectors, and river channel state evaluation vectors into the dynamic multi-source heterogeneous data fusion module based on the attention mechanism to generate dynamic fusion data; Further, the structure of the dynamic multi-source heterogeneous data fusion module based on the attention mechanism is as Figure 2 shown, including: a data embedding layer, an attention weighting layer, a comprehensive feature fusion layer, and an output layer; Further, the process of inputting the multi-source preprocessed data, analysis requirement vectors, and river channel state evaluation vectors into the dynamic multi-source heterogeneous data fusion module based on the attention mechanism to generate dynamic fusion data includes: Input the multi-source preprocessed data, analysis requirement vectors, and river channel state evaluation vectors into the data embedding layer of the dynamic multi-source heterogeneous data fusion module to obtain multi-source embedding vectors, analysis requirement embedding vectors, and river channel state evaluation embedding vectors; Input the multi-source embedding vectors, analysis requirement embedding vectors, and river channel state evaluation embedding vectors into the attention weighting layer of the dynamic multi-source heterogeneous data fusion module to obtain multi-source weighted feature vectors; Input the multi-source weighted feature vector, analysis requirement embedding vector, and river channel state evaluation embedding vector into the comprehensive feature fusion layer of the dynamic multi-source heterogeneous data fusion module to obtain a fused feature vector; Input the fused feature vector into the output layer of the dynamic multi-source heterogeneous data fusion module to obtain dynamic fused data; Furthermore, the data embedding layer processing procedure is as follows: Use a recurrent neural network to process the multi-source preprocessed data to obtain multi-source embedding vectors; Use a linear layer to process the analysis requirement vector and the river channel state evaluation vector respectively to obtain the analysis requirement embedding vector and the river channel state evaluation embedding vector; Furthermore, the process for the attention weighting layer to obtain the multi-source weighted feature vector is as follows: Use the channel merged embedding vector of the analysis requirement embedding vector and the river channel state evaluation embedding vector as the query vector, and use the multi-source embedding vectors as the key vector and value vector; Use the dot product operation between the query vector and the key vector and the normalization process of the softmax activation function to obtain the attention weights; Apply the attention weights to the value vector to obtain the multi-source weighted feature vector; Furthermore, the comprehensive feature fusion layer performs feature fusion using weighted summation; Among them, the weight of the multi-source weighted feature vector comes from the attention weights in the attention weighting layer; The weights of the analysis requirement embedding vector and the river channel state evaluation embedding vector are gated weights obtained through processing by a linear layer and a Sigmoid activation function.

[0023] By using the dynamic multi-source heterogeneous data fusion module based on the attention mechanism, it can adaptively adjust data fusion according to different analysis requirements and the changing river channel state, combining the attention and gating mechanisms to meet the actual needs, providing data support for subsequent river channel state prediction and decision-making optimization.

[0024] Input the dynamic fused data into the cross-domain associated river channel state collaborative prediction module to obtain the collaborative prediction result of the river channel state; Furthermore, the process of inputting the dynamic fused data into the cross-domain associated river channel state collaborative prediction module to obtain the collaborative prediction result of the river channel state includes: Define the nodes and edges in the graph structure, and construct node feature vectors based on the dynamic fused data; Use the historical dynamic fused data to train the river channel state collaborative prediction module, and perform graph data processing and river channel state prediction according to the spatio-temporal graph convolutional layer and the prediction output layer in the river channel state collaborative prediction module respectively to obtain the pre-trained river channel state collaborative prediction module; Input the dynamic fused data into the pre-trained river channel state collaborative prediction module to obtain the collaborative prediction result of the river channel state; Further, the nodes include: hydrological monitoring stations, key water quality monitoring points, and engineering facility points; the edges represent the connection relationships between the nodes, including: water flow paths / spatial adjacency, control relationships, and strong correlations; among them, the edge structure representing the water flow path / spatial adjacency is directional, that is, the water flow direction; the edge structure representing the strong correlation is obtained based on domain prior knowledge and data analysis, and is used to connect those nodes that are not adjacent in space; Further, at each time step, for each node in the graph structure, its corresponding dynamically fused data record is used as the node feature vector of the node at that moment; Further, the spatio-temporal graph convolutional layer in the river channel state collaborative prediction module includes a graph convolutional network (GCN) and a long short-term memory network (LSTM); the spatio-temporal graph convolutional layer first uses the GCN to aggregate the spatial information of neighboring nodes and updates the node representation to reflect the spatial dependence; then, the updated node representation sequence is processed by the LSTM to capture the time dynamics; Further, after multi-layer spatio-temporal graph convolutional processing, the module generates a final hidden representation rich in spatio-temporal and cross-domain information for each node; the prediction output layer performs prediction processing on the final hidden representation to obtain the state prediction result; the prediction output layer uses a fully connected layer.

[0025] To verify the effectiveness of the river channel state collaborative prediction method proposed by the present invention, 500 sets of historical data are input into three different prediction schemes for the effectiveness test of river channel state collaborative prediction; Scheme 1 is the prediction method proposed by the present invention, that is, dynamically fused data + river channel state collaborative prediction module; Scheme 2 is to replace the dynamically fused data with the data that can only be collected by each node; Scheme 3 is to replace the spatio-temporal graph convolutional layer in the river channel state collaborative prediction module with a graph convolutional network; the modules of each scheme are pre-trained, and the prediction results obtained by each scheme are manually compared with the actual results. The effectiveness test results of the river channel state collaborative prediction are shown in Table 1.

[0026] Table 1. Effectiveness test results of river channel state collaborative prediction

[0027] As can be seen from Table 1, compared with other schemes, the state prediction scheme (Scheme 1) proposed by the present invention can obtain more accurate and reliable prediction results, which helps to improve the effectiveness and implementation efficiency of subsequent river channel data decision optimization, thereby improving the river channel data management level.

[0028] By using the spatio-temporal graph convolutional layer in the river channel state collaborative prediction module, the spatial dependence in the graph structure and the temporal dependence in the time series can be captured simultaneously, enabling the module to learn a richer and more comprehensive representation of the river channel state characteristics. By capturing and leveraging the internal correlations among hydrological, water quality, and engineering domain data, more accurate river channel state prediction results can be obtained, which is beneficial for the subsequent collaborative optimization of integrated management data.

[0029] Construct a multi-objective collaborative optimization module. Input the dynamically fused data and the collaborative prediction results into the multi-objective collaborative optimization module. Dynamically adjust the objective weights and decision priorities according to the dynamically fused data and the collaborative prediction results. Use a multi-objective optimization algorithm to solve the problem and obtain the optimized decision parameters, and perform optimization according to the optimized decision parameters.

[0030] Furthermore, the process of inputting the dynamically fused data and the collaborative prediction results into the multi-objective collaborative optimization module, dynamically adjusting the objective weights and decision priorities according to the dynamically fused data and the collaborative prediction results, and using a multi-objective optimization algorithm to solve the problem and obtain the optimized decision parameters includes: Define the objective function and constraints of the multi-objective collaborative optimization module, and determine the decision variables; Input the dynamically fused data and the collaborative prediction results into the multi-objective collaborative optimization module, and dynamically adjust the objective weights and decision priorities according to the dynamically fused data and the collaborative prediction results; Combine the adjusted objective weights and decision priorities, and use a multi-objective optimization algorithm to solve the optimization problem to obtain the optimized decision parameters; Furthermore, the objective function is divided into flood control safety objectives, ecological protection objectives, and economic benefit objectives. Among them, the flood control safety objectives include minimizing flood risk and reducing flood inundation area; the ecological protection objectives include maximizing the dissolved oxygen index, minimizing the ammonia nitrogen index, maximizing species richness, etc.; the economic benefit objectives include maximizing water resource utilization efficiency, minimizing operation costs, minimizing engineering facility energy consumption, etc. Furthermore, the constraints include flood control safety thresholds, minimum ecological flow requirements, water quality standards, etc.; the decision variables include flood discharge, water storage, dredging area, levee reinforcement strength, sewage discharge, etc. In this embodiment, the different objective functions in the multi-objective collaborative optimization module affect each other. For example, optimizing the ecological protection objective will increase the operation cost, thus affecting the optimization of the economic benefit objective. Therefore, a dynamic weight adjustment strategy is adopted to balance the influence among objectives in the multi-objective collaborative optimization process.

[0031] Further, the process of dynamically adjusting the target weights and decision priorities according to the dynamically fused data and collaborative prediction results is as follows: Input the dynamically fused data and collaborative prediction results into a regression model based on weight prediction to obtain the dynamic weights of flood control safety objectives, ecological protection objectives, and economic benefit objectives; Input the dynamic weights into a regression model based on priority prediction to obtain the priorities of different decision variables; Further, the regression model can adopt support vector machine, decision tree, random forest, gradient boosting decision tree, neural network, etc.; Further, the multi-objective optimization algorithm adopted is a multi-objective evolutionary algorithm.

[0032] Given an analysis requirement report related to flood control and a river channel status assessment report, Table 2 below gives a comparison table of decision parameter changes before and after using the multi-objective collaborative optimization module; among them, the decision parameters include: reservoir flood discharge, downstream gate water storage, sewage treatment plant sewage discharge limit, and dike reinforcement frequency; Table 2. Comparison Table of Decision Parameter Changes

[0033] By combining the dynamically fused data, collaborative prediction results, and multi-objective collaborative optimization module, using the dynamically fused data and collaborative prediction results to dynamically adjust the target weights and decision priorities to achieve a more flexible adjustment of the optimization strategy; using the multi-objective optimization algorithm for solution can achieve the balance and coordination among various objectives; realizing the intelligence and refinement of river channel comprehensive data decision-making, effectively improving the level of river channel comprehensive management.

[0034] This embodiment proposes a method for collaborative optimization of river channel comprehensive management data based on the Internet of Things. First, obtain multi-source heterogeneous data, analysis requirement reports, and river channel status assessment reports collected by Internet of Things sensors, and after preprocessing, obtain multi-source preprocessed data, analysis requirement vectors, and river channel status assessment vectors, and input them into a dynamic multi-source heterogeneous data fusion module based on the attention mechanism to obtain dynamically fused data; Then, input the dynamically fused data into a cross-domain associated river channel status collaborative prediction module to obtain the collaborative prediction results of the river channel status; Finally, input the dynamically fused data and collaborative prediction results into a multi-objective collaborative optimization module, dynamically adjust the target weights and decision priorities according to the module input, use the multi-objective optimization algorithm for solution to obtain optimized decision parameters and optimize according to the optimized decision parameters; The present invention can effectively improve the management decision-making level of river channel data.

[0035] Embodiment 2: The present invention also proposes a system for collaborative optimization of river channel comprehensive management data based on the Internet of Things. The structure of the system is as Figure 3As shown in the figure, it includes: a data acquisition unit, a data processing unit, a data dynamic fusion unit, a river channel state collaborative prediction unit, and a multi-objective collaborative optimization unit; The data acquisition unit is used to obtain multi-source heterogeneous data collected by Internet of Things sensors, an analysis requirement report, and a river channel state assessment report; The data processing unit is used to preprocess the multi-source heterogeneous data, the analysis requirement report, and the river channel state assessment report to obtain multi-source preprocessed data, an analysis requirement vector, and a river channel state assessment vector; The data dynamic fusion unit is used to input the multi-source preprocessed data, the analysis requirement vector, and the river channel state assessment vector into a dynamic multi-source heterogeneous data fusion module based on an attention mechanism to generate dynamic fusion data; Further, the process of the data dynamic fusion unit generating dynamic fusion data includes: Inputting the multi-source preprocessed data, the analysis requirement vector, and the river channel state assessment vector into the data embedding layer of the dynamic multi-source heterogeneous data fusion module to obtain a multi-source embedding vector, an analysis requirement embedding vector, and a river channel state assessment embedding vector; Inputting the multi-source embedding vector, the analysis requirement embedding vector, and the river channel state assessment embedding vector into the attention weighting layer of the dynamic multi-source heterogeneous data fusion module to obtain a multi-source weighted feature vector; Inputting the multi-source weighted feature vector, the analysis requirement embedding vector, and the river channel state assessment embedding vector into the comprehensive feature fusion layer of the dynamic multi-source heterogeneous data fusion module to obtain a fusion feature vector; Inputting the fusion feature vector into the output layer of the dynamic multi-source heterogeneous data fusion module to obtain dynamic fusion data.

[0036] In order to verify the effectiveness of the dynamic multi-source heterogeneous data fusion module, 400 groups of historical test data were randomly selected to conduct the effectiveness test of the dynamic multi-source heterogeneous data fusion module using three test modules respectively; among them, the historical test data includes: historical multi-source preprocessed data, historical analysis requirement vectors, and historical river channel state assessment vectors; since the dynamic fusion data cannot intuitively reflect the differences, the pre-trained river channel state collaborative prediction module of the river channel state collaborative prediction unit is used for verification; inputting the historical dynamic fusion data output by each module into the pre-trained river channel state collaborative prediction module to obtain historical state prediction results and comparing them with the true values to obtain the proportion of the prediction results within a reasonable range; The test modules are as follows: the dynamic multi-source heterogeneous data fusion module proposed in the present invention, denoted as test module one; removing the attention weighting layer in the dynamic multi-source heterogeneous data fusion module and directly sending the output vector of the data embedding layer to the comprehensive feature fusion layer for fusion, denoted as test module two; removing the comprehensive feature fusion layer in the dynamic multi-source heterogeneous data fusion module and directly sending the multi-source weighted feature vector output by the attention weighting layer to the output layer, denoted as test module three; The test results of the effectiveness of the dynamic multi-source heterogeneous data fusion module can be referred to Table 3.

[0037] Table 3. Test results of the effectiveness of the dynamic multi-source heterogeneous data fusion module

[0038] As can be seen from Table 3, using test module one, that is, the dynamic multi-source heterogeneous data fusion module proposed in the present invention, the obtained test results of the module effectiveness are better than those using other modules, indicating that the output data of the module proposed in the present invention helps to improve the accuracy of river state prediction, thereby further improving the river data management level.

[0039] The river state collaborative prediction unit is used to input the dynamic fusion data into the river state collaborative prediction module with cross-domain association to obtain the collaborative prediction result of the river state; Further, the process by which the river state collaborative prediction unit obtains the collaborative prediction result of the river state is as follows: Define the nodes and edges in the graph structure and construct the node feature vectors based on the dynamic fusion data; Use the historical dynamic fusion data to train the river state collaborative prediction module, and perform graph data processing and river state prediction according to the spatio-temporal graph convolutional layer and the prediction output layer in the river state collaborative prediction module respectively to obtain the pre-trained river state collaborative prediction module; Input the dynamic fusion data into the pre-trained river state collaborative prediction module to obtain the collaborative prediction result of the river state.

[0040] The multi-objective collaborative optimization unit is used to input the dynamic fusion data and the collaborative prediction result into the multi-objective collaborative optimization module, dynamically adjust the objective weights and decision priorities according to the dynamic fusion data and the collaborative prediction result, use the multi-objective optimization algorithm to solve, obtain the optimized decision parameters and perform optimization according to the optimized decision parameters.

[0041] Further, the process by which the multi-objective collaborative optimization unit obtains the optimized decision parameters includes: Define the objective function and constraint conditions of the multi-objective collaborative optimization module and determine the decision variables; Input the dynamic fusion data and collaborative prediction results into the multi-objective collaborative optimization module, and dynamically adjust the objective weights and decision priorities according to the dynamic fusion data and collaborative prediction results; Combine the adjusted objective weights and decision priorities, and use the multi-objective optimization algorithm to solve the optimization problem to obtain the optimized decision parameters.

[0042] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A collaborative optimization method for river comprehensive management and control data based on the Internet of Things, characterized in that: include: Obtain multi-source heterogeneous data collected by IoT sensors, analyze demand reports and river status assessment reports; Preprocessing the multi-source heterogeneous data, the analysis requirement report and the river channel status assessment report to obtain multi-source preprocessed data, an analysis requirement vector and a river channel status assessment vector; Inputting the multi-source pre-processed data, the analysis requirement vector and the river channel state assessment vector into a dynamic multi-source heterogeneous data fusion module based on an attention mechanism to generate dynamic fusion data; Inputting the dynamic fusion data into a cross-domain associated river channel state collaborative prediction module to obtain a collaborative prediction result of the river channel state; Construct a multi-objective collaborative optimization module, input the dynamic fusion data and the collaborative prediction results into the multi-objective collaborative optimization module, dynamically adjust the target weights and decision priorities according to the dynamic fusion data and the collaborative prediction results, use a multi-objective optimization algorithm to solve, obtain optimized decision parameters, and optimize according to the optimized decision parameters.

2. According to the method of collaborative optimization of river comprehensive management and control data based on the Internet of Things in claim 1, it is characterized in that: The multi-source heterogeneous data include: river hydrological status data, river water quality status data and river engineering status data.

3. According to the method of collaborative optimization of river comprehensive management and control data based on the Internet of Things in claim 1, it is characterized in that: The process of preprocessing the multi-source heterogeneous data, the analysis requirement report and the river channel status assessment report to obtain multi-source preprocessed data, an analysis requirement vector and a river channel status assessment vector includes: Performing data cleaning, format standardization, and time-space alignment on the multi-source heterogeneous data to obtain the multi-source pre-processed data; The analysis requirement report and the river channel status assessment report are respectively subjected to text cleaning, text analysis and vectorization to obtain the analysis requirement vector and the river channel status assessment vector.

4. According to the method of collaborative optimization of river comprehensive management and control data based on the Internet of Things in claim 1, it is characterized in that: The multi-source pre-processed data, the analysis requirement vector and the river state assessment vector are input into a dynamic multi-source heterogeneous data fusion module based on an attention mechanism to generate dynamic fusion data, comprising: Input the multi-source pre-processed data, the analysis requirement vector and the river channel state assessment vector into the data embedding layer of the dynamic multi-source heterogeneous data fusion module to obtain a multi-source embedding vector, an analysis requirement embedding vector and a river channel state assessment embedding vector; Inputting the multi-source embedding vector, the analysis requirement embedding vector and the river state assessment embedding vector into the attention weighted layer of the dynamic multi-source heterogeneous data fusion module to obtain a multi-source weighted feature vector; Inputting the multi-source weighted feature vector, the analysis requirement embedding vector and the river channel status assessment embedding vector into the comprehensive feature fusion layer of the dynamic multi-source heterogeneous data fusion module to obtain a fused feature vector; The fused feature vector is input into the output layer of the dynamic multi-source heterogeneous data fusion module to obtain the dynamic fused data.

5. According to the method of collaborative optimization of river comprehensive management and control data based on the Internet of Things in claim 1, it is characterized in that: The process of inputting the dynamic fusion data into the cross-domain associated river state collaborative prediction module to obtain the collaborative prediction result of the river state includes: Defining nodes and edges in a graph structure, and constructing node feature vectors based on the dynamic fusion data; The river channel state collaborative prediction module is trained using historical dynamic fusion data, and graph data processing and river channel state prediction are performed according to the spatiotemporal graph convolution layer and the prediction output layer in the river channel state collaborative prediction module, respectively, to obtain a pre-trained river channel state collaborative prediction module; The dynamic fusion data is input into the pre-trained river channel state collaborative prediction module to obtain the collaborative prediction result of the river channel state.

6. The method for collaborative optimization of river comprehensive management and control data based on the Internet of Things according to claim 1 is characterized in that: The process of inputting the dynamic fusion data and the collaborative prediction result into the multi-objective collaborative optimization module, dynamically adjusting the objective weight and decision priority according to the dynamic fusion data and the collaborative prediction result, and solving by using the multi-objective optimization algorithm to obtain the optimized decision parameters includes: Defining the objective function and constraint conditions of the multi-objective collaborative optimization module and determining the decision variables; Inputting the dynamic fusion data and the collaborative prediction results into the multi-objective collaborative optimization module, and dynamically adjusting the objective weight and decision priority according to the dynamic fusion data and the collaborative prediction results; The adjusted target weights are combined with the decision priorities, and a multi-objective optimization algorithm is used to solve the optimization problem to obtain the optimization decision parameters.

7. A river comprehensive management and control data collaborative optimization system based on the Internet of Things, characterized in that: include: Data acquisition unit, data processing unit, data dynamic fusion unit, river status collaborative prediction unit and multi-objective collaborative optimization unit; The data acquisition unit is used to obtain multi-source heterogeneous data, analysis demand reports and river status assessment reports collected by Internet of Things sensors; the data processing unit is used to pre-process the multi-source heterogeneous data, the analysis demand report and the river status assessment report to obtain multi-source pre-processed data, analysis demand vectors and river status assessment vectors; the data dynamic fusion unit is used to input the multi-source pre-processed data, the analysis demand vectors and the river status assessment vectors into a dynamic multi-source heterogeneous data fusion module based on an attention mechanism to generate dynamic fusion data; the river status collaborative prediction unit is used to input the dynamic fusion data into a cross-domain associated river status collaborative prediction module to obtain a collaborative prediction result of the river status; the multi-objective collaborative optimization unit is used to input the dynamic fusion data and the collaborative prediction result into a multi-objective collaborative optimization module, dynamically adjust the target weight and decision priority according to the dynamic fusion data and the collaborative prediction result, solve using a multi-objective optimization algorithm, obtain optimized decision parameters and optimize according to the optimized decision parameters.

Images