Engineering project management system and method based on big data analysis

Abstract

The invention relates to the field of intelligent management of water conservancy projects, and provides an engineering project management system and method based on big data analysis. According to the method, historical hydrological data including specific indexes such as water level, reservoir inflow and reservoir outflow and historical meteorological data including specific indexes such as rainfall, evaporation capacity and air temperature of a target management reservoir are comprehensively analyzed by adopting a data analysis technology based on deep learning; and a reservoir dispatching operation scheme is generated based on the features obtained through comprehensive analysis. In this way, the problems that a traditional decision-making mode lacks data support and scientific analysis is insufficient can be solved, and the scientificity and rationality of reservoir management decision-making can be improved.

Description

Technical Field

[0001] The present application relates to the field of intelligent management of water conservancy projects, and more specifically, to an engineering project management system and method based on big data analysis. Background Art

[0002] The scientificity and rationality of water conservancy project management decisions are directly related to whether the project can fully realize its due benefits. Among the many water conservancy projects, reservoirs are particularly important as comprehensive water conservancy hubs that integrate multiple functions such as flood control, power generation, irrigation, water supply and ecological protection.

[0003] Reservoir management is not only related to flood control safety in downstream areas, but also can effectively regulate water volume and reduce the impact of flood disasters when floods come. It also shoulders the responsibility of providing clean energy for society, reducing dependence on traditional energy through hydropower generation, and promoting green and low-carbon development. At the same time, reservoirs are also an important source of water for agricultural irrigation. In addition, with the acceleration of urbanization, reservoirs also provide a stable and reliable source of water for urban water supply, ensuring the basic living needs of residents.

[0004] However, in the actual management process, traditional reservoir management decisions often rely on the experience and judgment of managers. Although this method can cope with daily management needs to a certain extent, it is powerless when faced with complex situations such as extreme weather conditions and water shortages. Due to the lack of sufficient data support and scientific analysis, decisions are often difficult to make accurately and effectively.

[0005] Therefore, an engineering project management solution based on big data analysis is needed. Summary of the invention

[0006] In view of the shortcomings of the prior art, the present application provides an engineering project management system and method based on big data analysis.

[0007] A project management method based on big data analysis, comprising:

[0008] Obtaining hydrological data and meteorological data of the target management reservoir at multiple historical time points, wherein the hydrological data includes water level, inflow and outflow, and the meteorological data includes rainfall, evaporation, and temperature;

[0009] Performing data analysis on the hydrological data and meteorological data at multiple historical time points to obtain a time series characteristic matrix of parameters related to the reservoir scheduling operation plan;

[0010] Performing association coding on the time series characteristic matrix of the parameters related to the reservoir scheduling operation plan to obtain an association characteristic vector of the parameters related to the reservoir scheduling operation plan;

[0011] Based on the associated characteristic vectors of the parameters related to the reservoir scheduling operation plan, a scheduling plan for the target managed reservoir is obtained.

[0012] An engineering project management system based on big data analysis, comprising:

[0013] Reservoir-related data acquisition module, used to obtain hydrological data and meteorological data of the target management reservoir at multiple historical time points, the hydrological data includes water level, inflow and outflow, and the meteorological data includes rainfall, evaporation, and temperature;

[0014] A reservoir-related data analysis module, used to analyze the hydrological data and meteorological data at multiple historical time points to obtain a time series characteristic matrix of parameters related to the reservoir scheduling operation plan;

[0015] A feature association coding module is used to perform association coding on the time series feature matrix of the parameters related to the reservoir scheduling operation plan to obtain an association feature vector of the parameters related to the reservoir scheduling operation plan;

[0016] The reservoir dispatching scheme generating module is used to obtain the dispatching scheme of the target managed reservoir based on the associated characteristic vector of the parameters related to the reservoir dispatching operation scheme.

[0017] This application has significant technical effects due to the adoption of the above technical solutions:

[0018] The project management system and method based on big data analysis provided by this application uses data analysis technology based on deep learning to conduct a comprehensive analysis of the historical hydrological data of the target management reservoir, including specific indicators such as water level, inflow, outflow, and historical meteorological data including specific indicators such as rainfall, evaporation, and temperature, and generates a reservoir dispatching operation plan based on the characteristics obtained from the comprehensive analysis. In this way, the problem of lack of data support and insufficient scientific analysis in traditional decision-making methods can be overcome, which is conducive to improving the scientificity and rationality of reservoir management decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0020] Figure 1 It is a flowchart of an engineering project management method based on big data analysis according to an embodiment of the present application.

[0021] Figure 2This is a flowchart of step S120 in the engineering project management method based on big data analysis according to an embodiment of the present application.

[0022] Figure 3 This is a flowchart of step S121 in the engineering project management method based on big data analysis according to an embodiment of the present application.

[0023] Figure 4 This is a flowchart of step S122 in the engineering project management method based on big data analysis according to an embodiment of the present application.

[0024] Figure 5 This is a flowchart of step S130 in the engineering project management method based on big data analysis according to an embodiment of the present application.

[0025] Figure 6 It is a system block diagram of an engineering project management system based on big data analysis according to an embodiment of the present application. DETAILED DESCRIPTION

[0026] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0027] As mentioned in the above background technology, water conservancy projects are important infrastructures, and the scientificity and rationality of their management decisions have an important impact on the project benefits. As a multifunctional water conservancy hub project, the reservoir has the functions of flood control, power generation, irrigation, and water supply. Reservoir management faces multiple goals such as flood control safety, power generation benefits, and optimal allocation of water resources. Traditional reservoir management decisions mainly rely on experience judgment, lack of data support and scientific analysis, and the scientificity and rationality of decisions need to be improved. Therefore, a project management solution based on big data analysis is expected.

[0028] Figure 1 FIG. 1 is a flowchart of a project management method based on big data analysis according to an embodiment of the present application. Figure 1As shown, according to the engineering project management method based on big data analysis of the embodiment of the present application, the method includes: S110, obtaining the hydrological data and meteorological data of the target management reservoir at multiple historical time points, the hydrological data including water level, inflow and outflow, and the meteorological data including rainfall, evaporation, and temperature; S120, performing data analysis on the hydrological data and meteorological data at multiple historical time points to obtain a time series feature matrix of parameters related to the reservoir scheduling and operation plan; S130, performing association coding on the time series feature matrix of parameters related to the reservoir scheduling and operation plan to obtain an associated feature vector of parameters related to the reservoir scheduling and operation plan; S140, obtaining the scheduling plan of the target management reservoir based on the associated feature vector of parameters related to the reservoir scheduling and operation plan.

[0029] In step S110, the hydrological data and meteorological data of the target management reservoir at multiple historical time points are obtained, the hydrological data including water level, inflow and outflow, and the meteorological data including rainfall, evaporation and temperature. It should be understood that the water level refers to the water surface height corresponding to the water storage capacity of the reservoir at a certain point in time, and the water level data is an important indicator reflecting the water storage status of the reservoir; the inflow refers to the water flow entering the reservoir at a certain point in time, which usually includes the flow from natural water sources such as upstream rivers and rainfall, as well as possible artificially regulated flow (such as the flow introduced through the diversion channel); the outflow refers to the water flow flowing out of the reservoir at a certain point in time, which is usually used for irrigation, power generation, water supply and other purposes, and may also be used for flood discharge operations during flood control. Precipitation refers to the total amount of rainfall in a certain area within a certain period of time near a certain point in time. Precipitation data is an important basis for evaluating the amount of water inflow to the reservoir; evaporation refers to the amount of water lost by the water surface of the reservoir due to evaporation within a certain period of time near a certain point in time. Evaporation data can be used to evaluate the amount of water reduction in the reservoir in the future; temperature refers to the temperature data at a certain point in time. Although the temperature itself does not directly affect the reservoir operation, it is closely related to other meteorological factors (such as rainfall, evaporation, etc.). In general, hydrological data and meteorological data provide important information support for reservoir operation decisions. By analyzing and mining these data, useful features and information can be extracted to guide future decisions. This can make the operation plan more scientific, reasonable and effective. For example, in the flood control season, the water storage and outflow of the reservoir can be adjusted in advance according to the forecast data of rainfall and inflow to ensure flood control safety; in the dry season, irrigation and water supply plans can be reasonably arranged according to the evaporation and rainfall data to make full use of water resources.

[0030] In step S120, data analysis is performed on the hydrological data and meteorological data at multiple historical time points to obtain a time series characteristic matrix of parameters related to the reservoir operation scheme. Specifically, Figure 2FIG. 1 is a flowchart of step S120 in the engineering project management method based on big data analysis according to an embodiment of the present application. Figure 2 As shown, the step S120 includes: S121, time-series encoding the hydrological data of the historical multiple time points to obtain a reservoir hydrological parameter time-series characteristic matrix; S122, time-series encoding the meteorological data of the historical multiple time points to obtain a reservoir meteorological parameter time-series characteristic matrix; S123, fusing the reservoir hydrological parameter time-series characteristic matrix and the reservoir meteorological parameter time-series characteristic matrix to obtain the reservoir scheduling operation plan related parameter time-series characteristic matrix.

[0031] In step S121, the hydrological data at the multiple historical time points are time-series encoded to obtain the reservoir hydrological parameter time-series characteristic moments. Specifically, Figure 3 FIG. 1 is a flowchart of step S121 in the engineering project management method based on big data analysis according to an embodiment of the present application. Figure 3 As shown, the step S121 includes: S1211, arranging the hydrological data of the multiple historical time points into a reservoir hydrological parameter time series input matrix according to the time dimension; S1212, passing the reservoir hydrological parameter time series input matrix through a reservoir data encoder to obtain the reservoir hydrological parameter time series feature matrix.

[0032] In step S1211, the hydrological data at the multiple historical time points are arranged into a reservoir hydrological parameter time series input matrix according to the time dimension. It should be understood that the hydrological data at multiple historical time points reflect the hydrological state of the reservoir at different time points. In order to effectively use these data, they need to be arranged according to the time dimension. By arranging them according to the time dimension, the law of change of the hydrological state over time can be better captured and analyzed, which helps to more accurately predict the future operation state of the reservoir, thereby formulating a more scientific and reasonable scheduling plan, which in turn helps to improve the operation efficiency and safety of the reservoir and reduce losses caused by improper decision-making.

[0033] In step S1212, the reservoir hydrological parameter time series input matrix is ​​passed through the reservoir data encoder to obtain the reservoir hydrological parameter time series feature matrix. It should be understood that the original reservoir hydrological parameter time series input matrix contains a large amount of redundant information and noise. In order to extract the characteristic information that is more critical to subsequent analysis from the original time series input matrix, such as the periodicity of water level changes, the volatility of inflow and outflow flows, it is necessary to encode the reservoir hydrological parameter time series input matrix. Specifically, the reservoir hydrological parameter time series input matrix is ​​input into the reservoir data encoder. Here, the reservoir data encoder specifically refers to a convolutional neural network model in which adjacent layers use mutually transposed convolution kernels. The ordinary convolutional neural network model can extract local features in the input matrix through the convolution operation of the convolution layer, and here, by using the convolution kernels of the adjacent layers that are mutually transposed, the model's learning ability for features can be further enhanced, so that the model can capture more complex and abstract reservoir hydrological time series state characteristic information. And by using this structure, the information loss caused by the convolution process can be compensated to a certain extent, so that more original information can be retained in the encoded feature matrix. The reservoir hydrological parameter time series feature matrix obtained after encoding processing contains the time series change characteristics of water level and inflow and outflow flow. These dynamic feature information can provide a scientific reference for subsequent reservoir scheduling decisions.

[0034] In step S122, the meteorological data at the historical multiple time points are time-series encoded to obtain a reservoir meteorological parameter time-series characteristic matrix. Specifically, Figure 4 FIG. 1 is a flowchart of step S122 in the engineering project management method based on big data analysis according to an embodiment of the present application. Figure 4 As shown, the step S122 includes: S1221, arranging the meteorological data of the multiple historical time points into a reservoir meteorological parameter time series input matrix according to the time dimension; S1222, passing the reservoir meteorological parameter time series input matrix through the reservoir data encoder to obtain the reservoir meteorological parameter time series feature matrix.

[0035] In step S1221, the meteorological data at the multiple historical time points are arranged into a reservoir meteorological parameter time series input matrix according to the time dimension. It should be understood that meteorological data changes with time and has obvious time series characteristics. By arranging the meteorological data at multiple historical time points according to the time dimension, the changing patterns of meteorological elements (such as rainfall, evaporation, temperature, etc.) over time can be clearly reflected. This arrangement method helps to better understand and analyze meteorological data, thereby revealing the potential information and patterns hidden behind the meteorological data.

[0036] In step S1222, the reservoir meteorological parameter time series input matrix is ​​passed through the reservoir data encoder to obtain the reservoir meteorological parameter time series feature matrix. It should be understood that the reservoir meteorological parameter time series input matrix contains a large amount of raw data, which contains both useful information for subsequent analysis and redundancy or noise. Through the reservoir data encoder, these raw data can be effectively reduced in dimension and feature extracted to retain the most critical and useful information. The reservoir meteorological parameter time series feature matrix obtained by encoding contains the change trend information, periodicity information, abnormal values, etc. of different meteorological elements. Based on this reservoir meteorological parameter time series feature matrix, future meteorological conditions can be predicted more accurately, thereby providing a more reliable basis for scheduling plans.

[0037] In step S123, the reservoir hydrological parameter time series feature matrix and the reservoir meteorological parameter time series feature matrix are fused to obtain the reservoir scheduling operation plan related parameter time series feature matrix. It should be understood that the formulation of the reservoir scheduling operation plan needs to take into account a variety of factors, among which hydrological parameters (such as water level, flow, etc.) and meteorological parameters (such as rainfall, evaporation, etc.) are the two core elements. In order to comprehensively consider various key information when formulating the reservoir scheduling plan, it is necessary to fuse the reservoir hydrological parameter time series feature matrix and the reservoir meteorological parameter time series feature matrix in the technical solution of the present application. Based on the fused feature matrix, the subsequent prediction model can more accurately predict the relevant parameters of the reservoir scheduling operation plan, such as water level changes and flow changes in the future period of time, thereby helping to improve the accuracy and reliability of the scheduling plan.

[0038] In step S130, the reservoir scheduling operation scheme related parameter time series feature matrix is ​​associated coded to obtain the reservoir scheduling operation scheme related parameter associated feature vector. Specifically, Figure 5 FIG. 1 is a flowchart of step S130 in the engineering project management method based on big data analysis according to an embodiment of the present application. Figure 5As shown, the step S130 includes: S131, passing the time series characteristic matrix of the parameters related to the reservoir scheduling and operation plan through a local correlation feature capturer to obtain a local correlation characteristic matrix of the parameters related to the reservoir scheduling and operation plan; S132, expanding the local correlation characteristic matrix of the parameters related to the reservoir scheduling and operation plan to obtain a local correlation characteristic vector of the parameters related to the reservoir scheduling and operation plan; S133, passing the time series characteristic matrix of the parameters related to the reservoir scheduling and operation plan through a global correlation feature capturer to obtain a global correlation characteristic matrix of the parameters related to the reservoir scheduling and operation plan; S134, expanding the global correlation characteristic matrix of the parameters related to the reservoir scheduling and operation plan to obtain a global correlation characteristic vector of the parameters related to the reservoir scheduling and operation plan; S135, performing target domain scale modulation based on hidden disturbance optimization on the local correlation characteristic vector of the parameters related to the reservoir scheduling and operation plan and the global correlation characteristic vector of the parameters related to the reservoir scheduling and operation plan to obtain the correlation characteristic vector of the parameters related to the reservoir scheduling and operation plan.

[0039] In step S131, the reservoir scheduling operation scheme related parameter time series feature matrix is ​​passed through a local correlation feature capturer to obtain the reservoir scheduling operation scheme related parameter local correlation feature matrix. It should be understood that although the reservoir scheduling operation scheme related parameter time series feature matrix obtained after fusion contains various key information required for reservoir scheduling operation, this information often exists in a complex and obscure way. By capturing the correlation features of the fused matrix, the deep correlation between these information can be deeply excavated, thereby revealing the laws and patterns hidden behind the data. The captured correlation features can provide a basis for making more scientific and reasonable scheduling decisions. For example, by analyzing the correlation between parameters such as water level, flow, and rainfall, the water level change trend and flow distribution requirements in the future can be predicted, so as to formulate a more accurate scheduling plan. Specifically, in this application, the information in the reservoir scheduling operation scheme related parameter time series feature matrix is ​​captured by using a local correlation feature capturer. Here, the local correlation feature capturer refers to a convolutional neural network model using a two-dimensional convolution kernel. The two-dimensional convolution kernel can slide on the input matrix and extract local features, thereby revealing the correlation between different parameters at different time points.

[0040] In step S132, the local correlation feature matrix of the parameters related to the reservoir scheduling operation plan is expanded to obtain the local correlation feature vector of the parameters related to the reservoir scheduling operation plan. It should be understood that matrix operations are often more complicated. By expanding the local correlation feature matrix of the parameters related to the reservoir scheduling operation plan to obtain a one-dimensional vector, the complexity of data processing can be reduced and the computational efficiency of data processing can be improved. In addition, during the expansion process, the key feature information in the original matrix is ​​not lost, so the rationality of the finally generated reservoir scheduling plan will not be affected.

[0041] In step S133, the reservoir scheduling operation scheme related parameter time series feature matrix is ​​passed through a global correlation feature capturer to obtain the reservoir scheduling operation scheme related parameter global correlation feature matrix. Here, the global correlation feature capturer is a convolutional neural network model that uses a global conversion matrix as a filter. It should be understood that the global conversion matrix is ​​a special filter that acts on the entire input data, rather than acting only on a local area by sliding like a traditional convolution kernel. This matrix can capture the global information in the input data, that is, considering the feature points at all positions, thereby generating global correlation features. Through the global correlation feature capturer, the global dependency between the feature points at different positions in the reservoir scheduling operation scheme related parameter time series feature matrix can be captured, which is conducive to understanding the complex interactions between the reservoir scheduling operation related parameters, thereby improving the model's feature representation ability for the reservoir scheduling operation scheme related parameters. By capturing global dependencies and generating global correlation features, the relevant parameters can be accurately predicted, which can help make more scientific and reasonable scheduling decisions.

[0042] In step S134, the global correlation feature matrix of the parameters related to the reservoir scheduling operation plan is expanded to obtain the global correlation feature vector of the parameters related to the reservoir scheduling operation plan. Similarly, by expanding the global correlation feature matrix of the parameters related to the reservoir scheduling operation plan, the efficiency of data processing can be improved while retaining key feature information.

[0043] In step S135, the local correlation feature vector of the parameters related to the reservoir scheduling operation scheme and the global correlation feature vector of the parameters related to the reservoir scheduling operation scheme are modulated by the target domain scale based on hidden disturbance optimization to obtain the correlation feature vector of the parameters related to the reservoir scheduling operation scheme. It should be understood that the local correlation feature vector of the parameters related to the reservoir scheduling operation scheme mainly captures the local spatiotemporal characteristics of the related parameters, while the global correlation feature vector of the parameters related to the reservoir scheduling operation scheme focuses on the global dependency. In order to comprehensively capture the various characteristics of the parameters related to the reservoir scheduling operation scheme, including local details and global trends, it is necessary to fuse the local correlation feature vector of the parameters related to the reservoir scheduling operation scheme and the global correlation feature vector of the parameters related to the reservoir scheduling operation scheme in the technical solution of the present application. The feature vector obtained after fusion contains local correlation and global correlation features, which can more comprehensively reflect the complex relationship of the parameters related to the reservoir scheduling operation scheme. In contrast, if it is not fused, some important features may be omitted, resulting in inaccurate prediction, which will in turn affect the rationality of the scheduling scheme.

[0044] In particular, considering that the local correlation feature vectors of the parameters related to the reservoir scheduling and operation plan are more focused on capturing short-term fluctuations and local patterns in the input time series, while the global correlation feature vectors of the parameters related to the reservoir scheduling and operation plan are more focused on capturing the long-term trends and overall patterns in the input data, when the two vectors are directly fused, the more complex relationship between the two feature vectors may be ignored. If the complex relationship between local and global features is not fully considered in the feature fusion process, the final generated reservoir scheduling plan may lack sensitivity to local changes or adaptability to long-term trends. This may affect the efficiency and effectiveness of the scheduling plan, making it impossible to achieve the optimal water resources management effect in practical applications. Therefore, in the technical solution of the present application, it is necessary to perform target domain scale modulation based on hidden perturbation optimization on the local correlation feature vectors of the parameters related to the reservoir scheduling and operation plan and the global correlation feature vectors of the parameters related to the reservoir scheduling and operation plan to obtain the correlation feature vectors of the parameters related to the reservoir scheduling and operation plan.

[0045] Specifically, in an embodiment of the present application, the local correlation feature vector of the parameters related to the reservoir scheduling and operation plan and the global correlation feature vector of the parameters related to the reservoir scheduling and operation plan are modulated on the target domain scale based on hidden disturbance optimization to obtain the correlation feature vector of the parameters related to the reservoir scheduling and operation plan, including: mapping the local correlation feature vector of the parameters related to the reservoir scheduling and operation plan and the global correlation feature vector of the parameters related to the reservoir scheduling and operation plan to the intrinsic decomposition space to obtain the local correlation intrinsic feature vector of the parameters related to the reservoir scheduling and operation plan and the global correlation intrinsic feature vector of the parameters related to the reservoir scheduling and operation plan; mapping the local correlation intrinsic feature vector of the parameters related to the reservoir scheduling and operation plan and the global correlation intrinsic feature vector of the parameters related to the reservoir scheduling and operation plan Perform adaptive feature scaling of the target domain to obtain the local associated modulation intrinsic feature vector of the parameters related to the reservoir scheduling and operation plan and the global associated modulation intrinsic feature vector of the parameters related to the reservoir scheduling and operation plan; construct a fine-grained interaction semantic matrix of the parameters related to the reservoir scheduling and operation plan between the local associated modulation intrinsic feature vector of the parameters related to the reservoir scheduling and operation plan and the global associated modulation intrinsic feature vector of the parameters related to the reservoir scheduling and operation plan; based on the fine-grained interaction semantic matrix of the parameters related to the reservoir scheduling and operation plan, fuse the local associated modulation intrinsic feature vector of the parameters related to the reservoir scheduling and operation plan and the global associated modulation intrinsic feature vector of the parameters related to the reservoir scheduling and operation plan to obtain the associated feature vector of the parameters related to the reservoir scheduling and operation plan.

[0046] More specifically, in an embodiment of the present application, a fine-grained interaction semantic matrix of the parameters related to the reservoir scheduling and operation scheme is constructed between the local correlation modulation intrinsic feature vector of the parameters related to the reservoir scheduling and operation scheme and the global correlation modulation intrinsic feature vector of the parameters related to the reservoir scheduling and operation scheme, including: multiplying the transposed vector of the local correlation modulation intrinsic feature vector of the parameters related to the reservoir scheduling and operation scheme and the global correlation modulation intrinsic feature vector of the parameters related to the reservoir scheduling and operation scheme, and then dividing the multiplication by the scale of the global correlation modulation intrinsic feature vector of the parameters related to the reservoir scheduling and operation scheme to obtain the fine-grained interaction matrix of the parameters related to the reservoir scheduling and operation scheme; performing a convolution coding operation based on a convolution kernel of 3×3 on the fine-grained interaction matrix of the parameters related to the reservoir scheduling and operation scheme to obtain the fine-grained interaction semantic matrix of the parameters related to the reservoir scheduling and operation scheme.

[0047] More specifically, in an embodiment of the present application, based on the fine-grained interactive semantic matrix of the parameters related to the reservoir scheduling and operation plan, the local associated modulated intrinsic feature vectors of the parameters related to the reservoir scheduling and operation plan and the global associated modulated intrinsic feature vectors of the parameters related to the reservoir scheduling and operation plan are fused to obtain the associated feature vectors of the parameters related to the reservoir scheduling and operation plan, including: multiplying the fine-grained interactive semantic matrix of the parameters related to the reservoir scheduling and operation plan by the local associated modulated intrinsic feature vectors of the parameters related to the reservoir scheduling and operation plan to obtain the local associated feature vectors of the parameters related to the reservoir scheduling and operation plan after information field mapping; multiplying the fine-grained interactive semantic matrix of the parameters related to the reservoir scheduling and operation plan by the global associated modulated intrinsic feature vectors of the parameters related to the reservoir scheduling and operation plan to obtain the global associated feature vectors of the parameters related to the reservoir scheduling and operation plan after information field mapping; weighted fusion of the local associated feature vectors of the parameters related to the reservoir scheduling and operation plan after the information field mapping and the global associated feature vectors of the parameters related to the reservoir scheduling and operation plan after the information field mapping to obtain the associated feature vectors of the parameters related to the reservoir scheduling and operation plan.

[0048] In an embodiment of the present application, specifically, the local correlation feature vector of the parameters related to the reservoir scheduling and operation scheme and the global correlation feature vector of the parameters related to the reservoir scheduling and operation scheme are modulated on a target domain scale based on hidden disturbance optimization to obtain the correlation feature vector of the parameters related to the reservoir scheduling and operation scheme, including: processing the local correlation feature vector of the parameters related to the reservoir scheduling and operation scheme and the global correlation feature vector of the parameters related to the reservoir scheduling and operation scheme according to the following formula to obtain the correlation feature vector of the parameters related to the reservoir scheduling and operation scheme; wherein the formula is:

[0049]

[0050] v c =αv kt +βv st

[0051] Where V1 represents the local correlation eigenvector of the parameters related to the reservoir operation scheme, V2 represents the global correlation eigenvector of the parameters related to the reservoir operation scheme, PCA(·) represents the principal component analysis, U1 represents the sequence of the local correlation eigendecomposition vectors of the parameters related to the reservoir operation scheme, Λ1 represents the local correlation diagonal matrix of the parameters related to the reservoir operation scheme, λ 11 and λ 1m They represent the eigenvalues ​​of the first and mth positions of the local correlation diagonal matrix of the parameters related to the reservoir operation scheme, T represents the transpose of the vector, concat{·;·} represents the vector concatenation, and v 11 、v 12 、v 1mmThey represent the first, second and mth eigenvectors of the sequence of local correlation eigendecomposition vectors of the parameters related to the reservoir operation scheme, Vk represents the local correlation eigenvector of the parameters related to the reservoir operation scheme, U2 represents the sequence of global correlation eigendecomposition vectors of the parameters related to the reservoir operation scheme, Λ2 represents the local correlation diagonal matrix of the parameters related to the reservoir operation scheme, λ 21 and λ 2m They represent the eigenvalues ​​of the first and mth positions of the global correlation diagonal matrix of the parameters related to the reservoir operation scheme, respectively. 21 、v 22 、v 2m They represent the first, second and mth eigenvectors of the sequence of global correlation eigendecomposition vectors of the parameters related to the reservoir operation scheme, V s represents the global correlation intrinsic eigenvector of the parameters related to the reservoir operation scheme, v i represents the ith eigenvalue of the local correlation intrinsic eigenvector of the parameters related to the reservoir operation scheme, v j represents the jth eigenvalue of the global associated intrinsic eigenvector of the parameters related to the reservoir operation scheme, represents the square of the vector bi-norm, exp represents the natural exponential function, V′k represents the intrinsic eigenvector after local correlation modulation of the parameters related to the reservoir operation scheme, V′ s It represents the intrinsic eigenvector after global correlation modulation of the parameters related to the reservoir operation scheme, represents matrix multiplication, S represents the scale of the vector, M x Represents the fine-grained interaction matrix of parameters related to the reservoir operation scheme, Conv 3×3 (M x ) indicates that M x A convolutional coding operation based on a convolution kernel of 3×3 is performed, Ω represents the fine-grained interaction semantic matrix of parameters related to the reservoir operation scheme, v kt represents the local correlation feature vector of the parameters related to the reservoir operation scheme after the information field is mapped, v st represents the global correlation feature vector of the parameters related to the reservoir operation scheme after information field mapping, α and β represent weighted hyperparameters, and v c It represents the associated characteristic vector of parameters related to the reservoir operation scheme.

[0052] That is, in the technical solution of the present application, the local correlation feature vector of the parameters related to the reservoir scheduling operation scheme and the global correlation feature vector of the parameters related to the reservoir scheduling operation scheme are modulated on the target domain scale based on hidden disturbance optimization. The process first maps the local correlation feature vector of the parameters related to the reservoir scheduling operation scheme and the global correlation feature vector of the parameters related to the reservoir scheduling operation scheme to the intrinsic decomposition space to obtain the local correlation intrinsic feature vector of the parameters related to the reservoir scheduling operation scheme and the global correlation intrinsic feature vector of the parameters related to the reservoir scheduling operation scheme. Those of ordinary skill in the art should know that the process of mapping to the intrinsic decomposition space is a combination of dimensionality reduction and feature extraction. In a specific example, the eigenvalue decomposition technology of principal component analysis (PCA) is used to find the intrinsic decomposition space, and the intrinsic decomposition space mapping is performed. The mapped local correlation intrinsic feature vector of the parameters related to the reservoir scheduling operation scheme and the global correlation intrinsic feature vector of the parameters related to the reservoir scheduling operation scheme not only represent the projection of the original feature vector in the new coordinate system, but also reflect the main direction of change of the data, which helps to avoid the "dimensionality disaster", that is, the problem that the sample density drops sharply as the number of features increases.

[0053] Next, after mapping the feature vector to the intrinsic decomposition space, it is necessary to perform adaptive feature scaling of the target domain on the local associated intrinsic feature vectors of the parameters related to the reservoir scheduling and operation plan and the global associated intrinsic feature vectors of the parameters related to the reservoir scheduling and operation plan to adjust the source domain feature vectors to better match the data distribution of the target domain. This operation is intended to capture and compensate for the systematic deviations between the two domains, so as to reduce the hidden disturbance of the class expression of the local feature distribution of the local associated intrinsic feature vectors of the parameters related to the reservoir scheduling and operation plan and the global associated intrinsic feature vectors of the parameters related to the reservoir scheduling and operation plan to the overall class expression. After modulation, the intrinsic feature vectors after local associated modulation of the parameters related to the reservoir scheduling and operation plan and the intrinsic feature vectors after global associated modulation of the parameters related to the reservoir scheduling and operation plan can retain the characteristics of the source domain while being more in line with the actual needs of the target domain, thereby enhancing the model's understanding and application of the target domain data.

[0054] After completing the scale modulation of the feature vector, a fine-grained interactive semantic matrix of the parameters related to the reservoir scheduling and operation scheme is constructed between the intrinsic feature vector after local correlation modulation of the parameters related to the reservoir scheduling and operation scheme and the intrinsic feature vector after global correlation modulation of the parameters related to the reservoir scheduling and operation scheme. The fine-grained interactive semantic matrix of the parameters related to the reservoir scheduling and operation scheme is used to describe a more detailed relationship between feature vectors. The introduction of the fine-grained interactive semantic matrix of the parameters related to the reservoir scheduling and operation scheme enables the model to go beyond simple point-to-point comparison and consider more complex patterns and contextual information, thereby enhancing the model's sensitivity to subtle differences and improving the quality of feature representation. Finally, based on the fine-grained interactive semantic matrix of the parameters related to the reservoir scheduling and operation scheme, the intrinsic feature vector after local correlation modulation of the parameters related to the reservoir scheduling and operation scheme and the intrinsic feature vector after global correlation modulation of the parameters related to the reservoir scheduling and operation scheme are fused to obtain the associated feature vector of the parameters related to the reservoir scheduling and operation scheme.

[0055] In step S140, based on the reservoir scheduling operation scheme related parameter associated feature vector, obtain the scheduling scheme of the target management reservoir. Specifically, in the embodiment of the present application, the step S140 includes: passing the reservoir scheduling operation scheme related parameter associated feature vector through the scheduling scheme generator to obtain a generation result, and the generation result is used to represent the scheduling scheme of the target management reservoir. It should be understood that the reservoir scheduling operation scheme related parameter associated feature vector is obtained by processing and analyzing a large amount of historical data. In order to convert the features in this feature vector into an easy-to-understand scheduling scheme, it is necessary to use a scheduling scheme generator. The scheduling scheme generator can automatically generate a reasonable scheduling scheme based on the information in the newly input feature vector through a large amount of data training. The generated scheduling scheme includes the reservoir's water storage target, water storage strategy, flood discharge plan, etc. in different time periods. Through a reasonable scheduling scheme, it can be ensured that the reservoir maintains a reasonable water storage capacity and water supply capacity under different seasons and meteorological conditions, thereby ensuring the safety and stable supply of water resources.

[0056] In summary, the project management method based on big data analysis based on the embodiment of the present application is explained, which uses data analysis technology based on deep learning to comprehensively analyze the historical hydrological data of the target management reservoir including specific indicators such as water level, inflow, outflow, etc. and the historical meteorological data including specific indicators such as rainfall, evaporation, and temperature, and generates a reservoir dispatching operation plan based on the characteristics obtained by the comprehensive analysis. In this way, the problem of lack of data support and insufficient scientific analysis in traditional decision-making methods can be overcome, which is conducive to improving the scientificity and rationality of reservoir management decisions.

[0057] Figure 6: is a system block diagram of an engineering project management system based on big data analysis according to an embodiment of the present application. Figure 6 As shown, according to the engineering project management system 100 based on big data analysis of the embodiment of the present application, it includes: a reservoir-related data acquisition module 110, which is used to obtain the hydrological data and meteorological data of the target management reservoir at multiple historical time points, the hydrological data including water level, inflow and outflow, and the meteorological data including rainfall, evaporation, and temperature; a reservoir-related data analysis module 120, which is used to perform data analysis on the hydrological data and meteorological data at multiple historical time points to obtain a time series feature matrix of parameters related to the reservoir scheduling and operation plan; a feature association coding module 130, which is used to perform association coding on the time series feature matrix of parameters related to the reservoir scheduling and operation plan to obtain a parameter association feature vector related to the reservoir scheduling and operation plan; a reservoir scheduling plan generation module 140, which is used to obtain the scheduling plan of the target management reservoir based on the parameter association feature vector related to the reservoir scheduling and operation plan.

[0058] Here, those skilled in the art can understand that the specific functions and operations of each unit and module in the above-mentioned engineering project management system based on big data analysis 100 have been referred to above. Figures 1 to 5 The description of the engineering project management method based on big data analysis has been introduced in detail, and therefore, its repeated description will be omitted.

[0059] In summary, the engineering project management system 100 based on big data analysis according to the embodiment of the present application is explained, which uses data analysis technology based on deep learning to comprehensively analyze the historical hydrological data of the target management reservoir including specific indicators such as water level, inflow, outflow, etc. and the historical meteorological data including specific indicators such as rainfall, evaporation, and temperature, and generates a reservoir dispatching operation plan based on the characteristics obtained by the comprehensive analysis. In this way, the problem of lack of data support and insufficient scientific analysis in traditional decision-making methods can be overcome, which is conducive to improving the scientificity and rationality of reservoir management decisions.

[0060] As described above, the engineering project management system 100 based on big data analysis according to the embodiment of the present application can be implemented in various wireless terminals, such as a server for engineering project management based on big data analysis. In one example, the engineering project management system 100 based on big data analysis according to the embodiment of the present application can be integrated into the wireless terminal as a software module and / or a hardware module. For example, the engineering project management system 100 based on big data analysis can be a software module in the operating system of the wireless terminal, or can be an application developed for the wireless terminal; of course, the engineering project management system 100 based on big data analysis can also be one of the many hardware modules of the wireless terminal.

[0061] Alternatively, in another example, the engineering project management system 100 based on big data analysis and the wireless terminal may also be separate devices, and the engineering project management system 100 based on big data analysis may be connected to the wireless terminal via a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.

Claims

1. A project management method based on big data analysis, characterized in that: include: Obtaining hydrological data and meteorological data of the target management reservoir at multiple historical time points, wherein the hydrological data includes water level, inflow and outflow, and the meteorological data includes rainfall, evaporation, and temperature; Performing data analysis on the hydrological data and meteorological data at multiple historical time points to obtain a time series characteristic matrix of parameters related to the reservoir scheduling operation plan; Performing association coding on the time series characteristic matrix of the parameters related to the reservoir scheduling operation plan to obtain an association characteristic vector of the parameters related to the reservoir scheduling operation plan; Based on the associated characteristic vectors of the parameters related to the reservoir scheduling operation plan, a scheduling plan for the target managed reservoir is obtained.

2. The engineering project management method based on big data analysis according to claim 1 is characterized in that: The hydrological data and meteorological data at the historical multiple time points are analyzed to obtain a time series characteristic matrix of parameters related to the reservoir operation scheme, including: Performing time series coding on the hydrological data at the multiple historical time points to obtain a reservoir hydrological parameter time series characteristic matrix; Performing time series coding on the meteorological data at the plurality of historical time points to obtain a reservoir meteorological parameter time series characteristic matrix; The reservoir hydrological parameter time series characteristic matrix and the reservoir meteorological parameter time series characteristic matrix are integrated to obtain the reservoir scheduling operation plan related parameter time series characteristic matrix.

3. The engineering project management method based on big data analysis according to claim 2 is characterized in that: The hydrological data at the multiple historical time points are time-series encoded to obtain a reservoir hydrological parameter time-series characteristic matrix, including: Arranging the hydrological data of the multiple historical time points into a reservoir hydrological parameter time series input matrix according to the time dimension; The reservoir hydrological parameter time series input matrix is ​​passed through a reservoir data encoder to obtain the reservoir hydrological parameter time series characteristic matrix.

4. The engineering project management method based on big data analysis according to claim 3 is characterized in that: The meteorological data of the multiple historical time points are time-series encoded to obtain a reservoir meteorological parameter time-series characteristic matrix, including: Arranging the meteorological data of the multiple historical time points into a reservoir meteorological parameter time series input matrix according to the time dimension; The reservoir meteorological parameter time series input matrix is ​​passed through the reservoir data encoder to obtain the reservoir meteorological parameter time series characteristic matrix.

5. The engineering project management method based on big data analysis according to claim 4 is characterized in that: The reservoir scheduling operation scheme related parameter time series feature matrix is ​​associated with coding to obtain the reservoir scheduling operation scheme related parameter associated feature vector, including: The reservoir scheduling operation scheme related parameter time series feature matrix is ​​passed through a local correlation feature capturer to obtain a reservoir scheduling operation scheme related parameter local correlation feature matrix; Expanding the local correlation characteristic matrix of the parameters related to the reservoir scheduling operation plan to obtain the local correlation characteristic vector of the parameters related to the reservoir scheduling operation plan; The reservoir scheduling operation scheme related parameter time series feature matrix is ​​passed through a global correlation feature capturer to obtain a reservoir scheduling operation scheme related parameter global correlation feature matrix; Expanding the global correlation characteristic matrix of the parameters related to the reservoir scheduling operation plan to obtain a global correlation characteristic vector of the parameters related to the reservoir scheduling operation plan; The local correlation feature vector of the parameters related to the reservoir scheduling and operation plan and the global correlation feature vector of the parameters related to the reservoir scheduling and operation plan are modulated on a target domain scale based on hidden disturbance optimization to obtain the correlation feature vector of the parameters related to the reservoir scheduling and operation plan.

6. The engineering project management method based on big data analysis according to claim 5 is characterized in that: The method of performing target domain scale modulation based on hidden disturbance optimization on the local correlation feature vector of the parameters related to the reservoir scheduling operation plan and the global correlation feature vector of the parameters related to the reservoir scheduling operation plan to obtain the correlation feature vector of the parameters related to the reservoir scheduling operation plan includes: Mapping the local associated characteristic vector of the parameters related to the reservoir scheduling operation plan and the global associated characteristic vector of the parameters related to the reservoir scheduling operation plan to the intrinsic decomposition space to obtain the local associated intrinsic characteristic vector of the parameters related to the reservoir scheduling operation plan and the global associated intrinsic characteristic vector of the parameters related to the reservoir scheduling operation plan; Performing target domain adaptive feature scaling on the local associated intrinsic feature vector of the parameters related to the reservoir scheduling and operation scheme and the global associated intrinsic feature vector of the parameters related to the reservoir scheduling and operation scheme to obtain the local associated modulated intrinsic feature vector of the parameters related to the reservoir scheduling and operation scheme and the global associated modulated intrinsic feature vector of the parameters related to the reservoir scheduling and operation scheme; Constructing a fine-grained interaction semantic matrix of the parameters related to the reservoir scheduling and operation scheme between the intrinsic feature vectors after local correlation modulation of the parameters related to the reservoir scheduling and operation scheme and the intrinsic feature vectors after global correlation modulation of the parameters related to the reservoir scheduling and operation scheme; Based on the fine-grained interactive semantic matrix of the parameters related to the reservoir scheduling and operation plan, the intrinsic feature vectors after local correlation modulation of the parameters related to the reservoir scheduling and operation plan and the intrinsic feature vectors after global correlation modulation of the parameters related to the reservoir scheduling and operation plan are fused to obtain the associated feature vectors of the parameters related to the reservoir scheduling and operation plan.

7. The engineering project management method based on big data analysis according to claim 6 is characterized in that: Constructing a fine-grained interaction semantic matrix of the parameters related to the reservoir operation scheme between the intrinsic feature vector after local correlation modulation of the parameters related to the reservoir operation scheme and the intrinsic feature vector after global correlation modulation of the parameters related to the reservoir operation scheme, including: The transposed vector of the intrinsic feature vector after local correlation modulation of the parameters related to the reservoir scheduling and operation scheme and the intrinsic feature vector after global correlation modulation of the parameters related to the reservoir scheduling and operation scheme are multiplied and then divided by the scale of the intrinsic feature vector after global correlation modulation of the parameters related to the reservoir scheduling and operation scheme to obtain a fine-grained interaction matrix of the parameters related to the reservoir scheduling and operation scheme; A convolutional coding operation based on a convolution kernel of 3×3 is performed on the fine-grained interaction matrix of the parameters related to the reservoir scheduling and operation plan to obtain a fine-grained interaction semantic matrix of the parameters related to the reservoir scheduling and operation plan.

8. The engineering project management method based on big data analysis according to claim 7 is characterized in that: The reservoir data encoder is a convolutional neural network model that uses mutually transposed convolution kernels in adjacent layers, the local correlation feature capturer is a convolutional neural network model that uses two-dimensional convolution kernels, and the global correlation feature capturer is a convolutional neural network model that uses a global transformation matrix as a filter.

9. The engineering project management method based on big data analysis according to claim 8 is characterized in that: Based on the associated characteristic vector of the parameters related to the reservoir scheduling and operation plan, a scheduling plan for the target managed reservoir is obtained, including: passing the associated characteristic vector of the parameters related to the reservoir scheduling and operation plan through a scheduling plan generator to obtain a generation result, and the generation result is used to represent the scheduling plan for the target managed reservoir.

10. An engineering project management system based on big data analysis, characterized in that: include: Reservoir-related data acquisition module, used to obtain hydrological data and meteorological data of the target management reservoir at multiple historical time points, the hydrological data includes water level, inflow and outflow, and the meteorological data includes rainfall, evaporation, and temperature; A reservoir-related data analysis module, used to analyze the hydrological data and meteorological data at multiple historical time points to obtain a time series characteristic matrix of parameters related to the reservoir scheduling operation plan; A feature association coding module is used to perform association coding on the time series feature matrix of the parameters related to the reservoir scheduling operation plan to obtain an association feature vector of the parameters related to the reservoir scheduling operation plan; The reservoir dispatching scheme generating module is used to obtain the dispatching scheme of the target managed reservoir based on the associated characteristic vector of the parameters related to the reservoir dispatching operation scheme.

Images