A High-Dimensional Multi-Objective Hydropower Big Data Intelligent Decision-Making Method and System

Abstract

This application discloses a high-dimensional, multi-objective, intelligent decision-making method and system for hydropower big data, belonging to the field of artificial intelligence. The method includes: building an intelligent decision analysis platform based on the target company; collecting data on the target company's hydropower energy planning and uploading it to the intelligent decision analysis platform; generating a primary energy allocation decision model; training the primary energy allocation decision model one by one using the energy digital demands of each target branch company to obtain various gain training parameters; determining model optimization parameters; optimizing the primary energy allocation decision model to obtain the target energy allocation decision model; and making intelligent decisions on energy allocation based on the target hydropower big data. This solves the technical problems of low intelligence level in hydropower big data analysis and low efficiency and accuracy in energy allocation decisions in existing technologies. It achieves the technical effect of improving energy supply and demand coordination capabilities and energy allocation decision efficiency, thereby enhancing the economic benefits of enterprises.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence, and in particular to a high-dimensional, multi-objective, big data intelligent decision-making method and system for hydropower. Background Technology

[0002] With rapid economic development and steady improvement in science and technology, manufacturing is rapidly moving towards informatization. Informatization has been integrated into the entire process of production and operation, greatly promoting the development of enterprises.

[0003] Currently, a massive amount of data has been accumulated during the process of information technology construction and application. By utilizing system platforms for data sharing, storage, and processing, and by building different information platforms in conjunction with application scenarios, we can effectively extract and utilize data information, thereby improving the level of data service for production.

[0004] However, in practical applications, the value of data cannot be fully utilized due to the bottlenecks of traditional information technology. In the process of collecting, storing, analyzing, managing, and comprehensively utilizing massive amounts of data, the siloed information system architecture leads to a digital divide, lacking information technology support and failing to provide efficient data analysis results for project implementation. Existing technologies suffer from low levels of intelligence in hydropower big data analysis and low efficiency and accuracy in energy allocation decisions. Summary of the Invention

[0005] The purpose of this application is to provide a high-dimensional, multi-objective intelligent decision-making method and system for hydropower big data, in order to solve the technical problems of low intelligence level in hydropower big data analysis and low efficiency and accuracy of energy allocation decision-making in the existing technology.

[0006] In view of the above problems, this application provides a high-dimensional multi-objective hydropower big data intelligent decision-making method and system.

[0007] Firstly, this application provides a high-dimensional, multi-objective, intelligent decision-making method for hydropower big data. The method includes: building an intelligent decision analysis platform based on a target company, wherein the intelligent decision analysis platform includes a topology structure of the target company and its various target branches; collecting data on the target company's hydropower energy planning to obtain energy planning data; uploading the target company's energy planning data to the intelligent decision analysis platform to generate a preliminary energy allocation decision model; distributing the preliminary energy allocation decision model to each target branch and training the preliminary energy allocation decision model using the energy digital needs of each target branch; obtaining the gain training parameters of each target branch and sending the gain training parameters back to the target company for gain learning to determine model optimization parameters; and optimizing the preliminary energy allocation decision model using the model optimization parameters to obtain a target energy allocation decision model for intelligent energy allocation decision-making based on target hydropower big data.

[0008] On the other hand, this application also provides a high-dimensional, multi-objective hydropower big data intelligent decision-making system, wherein the system includes: a platform construction module, which is used to build an intelligent decision analysis platform based on a target company, wherein the intelligent decision analysis platform includes a topology structure of the target company and its various branch companies; a data acquisition module, which is used to collect data on the target company's hydropower energy planning to obtain the target company's energy planning data; a model generation module, which is used to upload the target company's energy planning data to the intelligent decision analysis platform to generate a preliminary energy allocation decision model; and a model training module, which... The model training module distributes the primary energy allocation decision model to each target branch company and trains the primary energy allocation decision model one by one using the energy digital requirements of each target branch company; the parameter determination module obtains the gain training parameters of each target branch company and sends the gain training parameters back to the target company for gain learning to determine the model optimization parameters; the intelligent decision-making module optimizes the primary energy allocation decision model using the model optimization parameters to obtain the target energy allocation decision model for intelligent decision-making on energy allocation based on target hydropower big data.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0010] This application establishes an intelligent decision analysis platform based on the specific circumstances of the target company. This platform includes the topology of the target company and its various branch companies. It then collects the target company's hydropower energy planning data, uploads this data to the intelligent decision analysis platform, generates a primary energy allocation decision model, and sends it to each branch company. The primary energy allocation decision model is then trained using the energy demand data of each branch company, yielding gain training parameters for each branch company. These gain training parameters are then fed back to the target company for gain learning, resulting in optimized model parameters. These optimized parameters are then used to further optimize the primary energy allocation decision model, resulting in the target energy allocation decision model. Finally, intelligent energy allocation decisions are made based on the target hydropower big data. This achieves the technical effect of intelligent energy allocation decision-making, improving allocation efficiency and accuracy. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0012] Figure 1 A flowchart illustrating a high-dimensional, multi-objective, big data-driven intelligent decision-making method for hydropower, provided as an embodiment of this application;

[0013] Figure 2 A flowchart illustrating the process of building an intelligent decision analysis platform based on a target company in a high-dimensional multi-objective hydropower big data intelligent decision-making method provided in this application embodiment;

[0014] Figure 3 A flowchart illustrating the generation of a primary energy allocation decision model in a high-dimensional multi-objective hydropower big data intelligent decision-making method provided in this application embodiment;

[0015] Figure 4 This is a schematic diagram of the structure of a high-dimensional multi-objective hydropower big data intelligent decision-making system according to this application;

[0016] Figure labeling: Platform building module 11, data acquisition module 12, model generation module 13, model training module 14, parameter determination module 15, intelligent decision-making module 16. Detailed Implementation

[0017] This application provides a high-dimensional, multi-objective intelligent decision-making method and system for hydropower big data, solving the technical problems of low intelligence level in hydropower big data analysis and low efficiency and accuracy in energy allocation decisions in existing technologies. It achieves the technical effect of improving energy supply and demand coordination capabilities and energy allocation decision-making efficiency, thereby enhancing enterprise economic benefits.

[0018] The acquisition, storage, use, and processing of data in this application all comply with the relevant provisions of national laws and regulations.

[0019] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. It should be understood that this application is not limited to the exemplary embodiments described herein. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application. It should also be noted that, for ease of description, only the parts related to this application are shown in the accompanying drawings, not all of them.

[0020] Example 1

[0021] like Figure 1 As shown, this application provides a high-dimensional, multi-objective, big data intelligent decision-making method for hydropower, wherein the method includes:

[0022] Step S100: Build an intelligent decision analysis platform based on the target company, wherein the intelligent decision analysis platform includes the topology of the target company and each target branch;

[0023] Furthermore, such as Figure 2 As shown, the intelligent decision analysis platform built based on the target company, step S100 of this application embodiment includes:

[0024] Step S110: Construct the target company's computing resource supply center;

[0025] Step S120: Pool the supply assets of the computing resource supply center to build a data aggregation center;

[0026] Step S130: Perform intelligent analysis on the aggregated data of the data aggregation center to construct a data analysis service center;

[0027] Step S140: Based on the computing resource supply center and the analysis data of the data analysis service center, a planning decision is made to construct a data decision center;

[0028] Step S150: Based on the data aggregation center, the data analysis service center, and the data decision center, build the intelligent decision analysis platform.

[0029] Specifically, the intelligent decision analysis platform is a platform for making overall energy allocation decisions based on the resource situation of the target company and the needs of each target branch. The target company is any company that needs to allocate energy resources. Each target branch is a branch that needs to obtain resources from the target company. The intelligent decision analysis platform includes the topology of the target company and the target branches, used to systematically organize the structure between the company and its branches. This provides a clear transmission path for data transmission.

[0030] Specifically, the computing resource supply center is a central platform for calculating the supply of resources possessed by the target company, used to calculate the amount of resources the target company can provide. The supply assets are assets that the target company can provide to the target branch; optionally, the supply assets can be: hydropower resources, thermal power resources, and new energy resources. The pooling management of supply assets involves managing assets based on the dynamic performance of the total supply assets during the supply process. This allows for demand-oriented integration and improvement of supply assets, realizing a shift from "production-driven sales" to "sales-driven production," improving supply efficiency, and maximizing benefits.

[0031] Specifically, the data aggregation center is a central platform for implementing pooled management, used to aggregate and store real-time dynamic fluctuation data of the supply assets during the allocation process. The aggregated data reflects changes in the supply assets, including hydropower scheduling data, hydropower production economic operation data, and fuel storage data. Then, the data analysis service center performs intelligent analysis on the aggregated data, analyzing data fluctuation rates, asset inflow and outflow changes, etc. Next, based on the analysis data obtained from the computing resource supply center and the data analysis service center, decisions are made regarding resource supply, constructing the data decision center. The data decision center is used to analyze and make decisions about resource supply and fluctuations. This achieves the goal of comprehensively understanding the supply resources, resulting in improved resource utilization efficiency and decision-making accuracy.

[0032] Step S200: Collect data on the target company's hydropower energy plan to obtain the target company's energy planning data;

[0033] Specifically, the target company's hydropower energy plan includes plans for the application of watershed hydropower and plans for thermal power operations. The target company's energy planning data reflects its planning for existing resources, including: thermal power generation, planned heat supply, planned hydropower supply, hydrological data, fuel procurement plans, fuel inventory data, equipment operating status data, and power production operation data. By understanding the company's energy planning data, the company can gain insight into resource utilization, thus providing the technical benefit of foundational data for subsequent energy allocation.

[0034] Step S300: Upload the target company's energy planning data to the intelligent decision analysis platform to generate a preliminary energy allocation decision model;

[0035] Furthermore, such as Figure 3 As shown, in the embodiment of this application, step S300 of generating the primary energy allocation decision model further includes:

[0036] Step S310: Historical data of the target company is collected to determine historical power generation flow.

[0037] Step S320: By performing in-depth analysis on the historical power generation flow, the distribution of upstream water level, downstream water level, and power generation water quality corresponding to each historical time node is determined;

[0038] Step S330: Perform mathematical statistics on the influencing factors of the upstream water level distribution and the historical power generation flow, the downstream water level distribution and the historical power generation flow, and the power generation water quality distribution and the historical power generation flow, respectively;

[0039] Step S340: Obtain the upstream water level impact parameters, downstream water level impact parameters, and water quality impact parameters.

[0040] Furthermore, step S300 in this embodiment of the application also includes:

[0041] Step S350: Construct a planning data calibration model using the upstream water level influence parameters, downstream water level influence parameters, and water quality influence parameters;

[0042] Step S360: Input the target company's energy planning data into the planning data calibration model to perform intelligent calibration of the influencing parameters, in order to generate the primary energy allocation decision model.

[0043] Specifically, historical data collection of the target company's basic data involves obtaining information on the company's resource utilization during past hydropower generation. This basic data includes: power generation, power output of the turbine generator units, output data of a specific cross-section of the water flow, water flow rate, normal water level of the hydropower station reservoir, and installed capacity of the hydropower station. Based on this basic data, the target company's historical power generation flow rate can be determined. This historical power generation flow rate refers to the power generation flow rate during the target company's historical operations.

[0044] Specifically, by acquiring historical power generation flow over a certain period, further analysis can reveal the water level distribution at each historical time point. The upstream water level distribution represents the chronological arrangement of upstream water level values ​​at each historical time point. The downstream water level distribution represents the chronological arrangement of downstream water level values ​​at each historical time point. The power generation water quality distribution represents the chronological arrangement of water quality data collected during the power generation process.

[0045] Specifically, the mathematical statistics of the influencing factors refer to obtaining the degree of influence of upstream and downstream water levels and water quality on power generation flow by combining historical power generation flow rates at various historical time points. The upstream water level influence parameter reflects the degree of influence of upstream water level on hydropower generation. The downstream water level influence parameter reflects the degree of influence of downstream water level on hydropower generation. The water quality influence parameter reflects the degree of influence of water quality on hydropower generation.

[0046] Specifically, the planning data calibration model is a functional model that intelligently calibrates energy planning data after determining the degree of influence of each influencing factor on hydropower generation using upstream water level influence parameters, downstream water level influence parameters, and water quality influence parameters. By intelligently calibrating the influence parameters of the target company's energy planning data, the target company's energy planning data is adjusted to output more accurate planning data after considering influencing factors, resulting in the primary energy allocation decision model. This primary energy allocation decision model is a functional model used to initially allocate energy to the target company. Thus, the goal of initially allocating energy based on historical data is achieved, resulting in improved allocation accuracy, reduced allocation errors, and increased energy utilization efficiency.

[0047] Step S400: Distribute the primary energy allocation decision model to each target branch, and train the primary energy allocation decision model one by one using the energy digital requirements of each target branch;

[0048] Furthermore, the primary energy allocation decision model is trained one by one. In this embodiment, step S400 further includes:

[0049] Step S410: Obtain the primary energy allocation plan for each target branch in the primary energy allocation decision model;

[0050] Step S420: By performing a one-to-one demand-allocation analysis on the energy digital demands and the primary energy allocation plan of each target branch, the distribution of energy demand profit and loss parameters of each target branch is determined.

[0051] Step S430: Perform parameter transformation on the energy demand surplus / deficit parameter distribution to generate the various gain training parameters.

[0052] Specifically, each target branch has different energy requirements due to its different production goals. The energy demand figures for each target branch reflect its actual energy needs. By using these energy demand figures to train the primary energy allocation decision model, each model is trained to allocate energy to its respective target branch. The training parameters are then generated by combining the energy allocation figures with the model's own needs.

[0053] Specifically, based on the primary energy allocation decision model, the primary energy allocation plan for each target branch is obtained by the target company. This primary energy allocation plan clearly defines the types and quantities of resources available to each target branch during its production process. This includes: power generation, power output of the hydroelectric generator set, water flow, and fuel consumption. Furthermore, by matching each energy demand with the primary energy allocation plan, a demand-allocation analysis is performed to obtain the demand-allocation matching status. The energy demand surplus / deficit parameter distribution of the target branch refers to the parameter distribution of the degree to which each target branch meets its demand. This energy demand surplus / deficit parameter reflects whether the energy demand in the primary energy allocation plan exceeds or falls below the target branch's energy demand. By averaging the parameter distribution, the gain training parameters are obtained to adjust the allocation plan based on the demand situation of each target branch. By rapidly responding to the energy demand of each branch, the overall plan can be adjusted in a timely manner, achieving the technical effect of improving the accuracy of decision-making.

[0054] Furthermore, regarding the primary energy allocation plan for each target branch company, step S410 of this application embodiment also includes:

[0055] Step S411: Collect and obtain the historical energy planning distribution of the target company for the target branch;

[0056] Step S412: Collect and obtain the historical energy usage feedback data distribution of the target branch company;

[0057] Step S413: Based on the distribution of historical energy usage feedback data, the historical energy planning distribution is corrected, and the primary energy allocation plan is determined based on the correction result.

[0058] Specifically, the historical energy planning distribution represents the historical energy resource acquisition situation of the target branch. The historical energy usage feedback data distribution represents the utilization of acquired energy resources by the target branch. The historical energy planning distribution is revised based on the feedback data distribution, correcting for resource shortages or surpluses. Thus, the primary energy allocation plan is obtained based on the revision results. After implementing the planned energy allocation, and considering the actual resource utilization of each target branch, feedback control is applied to the allocation plan, achieving the technical effect of improving resource utilization efficiency and planning accuracy.

[0059] Step S500: Obtain the gain training parameters of each target branch, and send the gain training parameters back to the target company for gain learning to determine the model optimization parameters;

[0060] Furthermore, the gain training parameters are fed back to the target company for gain learning. In this embodiment, step S500 further includes:

[0061] Step S510: Within the data decision center, gain learning is performed on the primary energy allocation plan using the gain training parameters to determine the adjusted energy allocation plan for each branch company.

[0062] Step S520: Input the adjusted energy allocation plans of each branch company into the primary energy allocation decision model for optimization training to obtain the optimized parameters of the trained model.

[0063] Specifically, the gain training parameters are used to correct the planned energy allocation. Therefore, the gain training parameters reflecting the performance of each target branch are fed back to the target company to obtain the deviation between the planned allocation and actual demand. Gain learning is then performed to obtain the adjusted energy allocation plan for each branch. This adjusted energy allocation plan for each branch considers both supply and actual demand, ensuring that it meets the needs of each branch. Furthermore, by optimizing and training the primary energy allocation decision model, the imbalanced allocation is improved, enhancing the model's allocation accuracy. This determines the model optimization parameters that characterize the optimization scale. This achieves the technical effect of optimizing the model and improving its decision-making accuracy.

[0064] Step S600: Optimize the primary energy allocation decision model using the model optimization parameters to obtain the target energy allocation decision model, which is used for intelligent energy allocation decision-making based on the target hydropower big data.

[0065] Specifically, the target energy allocation decision model is a functional model for allocating energy resources for the target company. It is obtained by optimizing the imbalanced portion of the primary energy allocation decision model among the various branches. The target energy allocation decision model is used to allocate energy based on the obtained target hydropower big data, resulting in intelligent decision-making. This achieves the goal of efficient data analysis and processing, intelligently deriving scientific decisions, and realizing the technical effects of intelligent production analysis and planning, improving energy utilization efficiency and decision-making accuracy.

[0066] In summary, the high-dimensional multi-objective hydropower big data intelligent decision-making method provided in this application has the following technical effects:

[0067] 1. This application embodiment utilizes an intelligent decision-making allocation platform to obtain the target company's hydropower energy planning data, perform initial allocation, and obtain an initial energy allocation decision model. Then, it trains the initial energy allocation decision model using the energy digital demands of each target branch company to determine the adaptability of each branch company to the initial allocation. Finally, it optimizes the model based on the gain training parameters to obtain the target energy allocation decision model, achieving intelligent decision-making for energy allocation based on the target hydropower big data. This achieves the technical effects of improving the intelligence level of hydropower big data allocation, increasing intelligent allocation efficiency, improving resource balance utilization, and maximizing enterprise benefits.

[0068] 2. This application constructs a computing resource supply center for the target company, pools and manages the supply assets of the computing resource supply center, constructs a data aggregation center, and intelligently analyzes the aggregated data of the data aggregation center by constructing a data analysis service center. Then, based on the computing resource supply center and the analysis data from the data analysis service center, planning and decision-making are carried out to construct a data decision center. Finally, based on the data aggregation center, the data analysis service center, and the data decision center, the intelligent decision analysis platform is built. This achieves the goal of resource integration and utilization, and achieves the technical effect of improving resource utilization efficiency and decision accuracy.

[0069] Example 2

[0070] Based on the same inventive concept as the high-dimensional multi-objective hydropower big data intelligent decision-making method in the foregoing embodiments, such as Figure 4 As shown, this application also provides a high-dimensional, multi-objective, big data intelligent decision-making system for hydropower, wherein the system includes:

[0071] Platform building module 11, which is used to build an intelligent decision analysis platform based on the target company, wherein the intelligent decision analysis platform includes the topology of the target company and each target branch;

[0072] Data acquisition module 12 is used to collect data on the target company's hydropower energy plan in order to obtain the target company's energy planning data.

[0073] Model generation module 13 is used to upload the target company's energy planning data to the intelligent decision analysis platform to generate a primary energy allocation decision model.

[0074] Model training module 14 is used to distribute the primary energy allocation decision model to each target branch, and to train the primary energy allocation decision model one by one using the energy digital requirements of each target branch.

[0075] The parameter determination module 15 is used to obtain the gain training parameters of each target branch and send the gain training parameters back to the target company for gain learning in order to determine the model optimization parameters.

[0076] The intelligent decision-making module 16 is used to optimize the primary energy allocation decision model using the model optimization parameters to obtain a target energy allocation decision model for intelligent decision-making on energy allocation of target hydropower big data.

[0077] Furthermore, the system also includes:

[0078] A central construction unit, which is used to construct the target company's computing resources for the center;

[0079] A data aggregation center construction unit is used to pool and manage the supply assets of the computing resource supply center in order to build a data aggregation center.

[0080] An analysis center construction unit is used to perform intelligent analysis on the aggregated data of the data aggregation center in order to build a data analysis service center;

[0081] A decision center construction unit is used to make planning decisions based on the computing resource supply center and the analysis data of the data analysis service center, so as to construct a data decision center;

[0082] An analysis platform construction unit is used to build the intelligent decision analysis platform based on the data aggregation center, the data analysis service center, and the data decision center.

[0083] Furthermore, the system also includes:

[0084] A historical flow acquisition unit is used to collect historical basic data of the target company to determine historical power generation flow.

[0085] The analysis unit is used to determine the distribution of upstream water level, downstream water level, and power generation water quality at each historical time point by performing in-depth analysis of the historical power generation flow.

[0086] The statistical unit is used to perform mathematical statistics on the influencing factors of the upstream water level distribution and the historical power generation flow, the downstream water level distribution and the historical power generation flow, and the power generation water quality distribution and the historical power generation flow, respectively.

[0087] An influence parameter acquisition unit is used to acquire upstream water level influence parameters, downstream water level influence parameters, and water quality influence parameters.

[0088] Furthermore, the system also includes:

[0089] A calibration model construction unit is used to construct a planning data calibration model using the upstream water level influence parameters, downstream water level influence parameters, and water quality influence parameters.

[0090] A primary allocation model generation unit is used to input the target company's energy planning data into the planning data calibration model for intelligent calibration of the influencing parameters, in order to generate the primary energy allocation decision model.

[0091] Furthermore, the system also includes:

[0092] The allocation planning unit is used to obtain the primary energy allocation plan for each target branch in the primary energy allocation decision model.

[0093] The profit and loss parameter distribution determination unit is used to determine the energy demand profit and loss parameter distribution of each target branch by performing a one-to-one demand-allocation analysis on the energy digital demand of each target branch and the primary energy allocation plan.

[0094] A gain parameter generation unit is used to perform parameter transformation on the energy demand surplus / deficit parameter distribution to generate the various gain training parameters.

[0095] Furthermore, the system also includes:

[0096] Historical planning unit, the historical planning unit is used to collect and obtain the historical energy planning distribution of the target company for the target branch;

[0097] Feedback unit, the feedback unit is used to collect and obtain the historical energy usage feedback data distribution of the target branch;

[0098] The correction unit is used to correct the historical energy planning distribution based on the historical energy use feedback data distribution, and to determine the primary energy allocation plan based on the correction result.

[0099] Furthermore, the system also includes:

[0100] A gain learning unit is used in the data decision center to perform gain learning on the primary energy allocation plan using the gain training parameters, in order to determine the adjusted energy allocation plan for each branch.

[0101] An optimization training unit is used to input the adjusted energy allocation plans of each branch company into the primary energy allocation decision model for optimization training, thereby obtaining the optimized parameters of the trained model.

[0102] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Figure 1 The high-dimensional multi-objective hydropower big data intelligent decision-making method and specific examples in Embodiment 1 are also applicable to the high-dimensional multi-objective hydropower big data intelligent decision-making system of this embodiment. Through the foregoing detailed description of the high-dimensional multi-objective hydropower big data intelligent decision-making method, those skilled in the art can clearly understand the high-dimensional multi-objective hydropower big data intelligent decision-making system of this embodiment; therefore, for the sake of brevity, it will not be described in detail here. As for the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant details can be found in the method section.

[0103] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A high-dimensional multi-objective hydroelectric power big data intelligent decision method, characterized in that, The method includes: An intelligent decision analysis platform is built based on the target company, wherein the intelligent decision analysis platform includes the topological structure of the target company and each target branch company; Data was collected on the target company's hydropower energy plan to obtain the target company's energy planning data; The target company's energy planning data is uploaded to the intelligent decision analysis platform to generate a preliminary energy allocation decision model. The primary energy allocation decision model is distributed to each target branch, and the primary energy allocation decision model is trained one by one using the energy digital requirements of each target branch. Obtain the gain training parameters for each target branch, and send the gain training parameters back to the target company for gain learning to determine the model optimization parameters; The primary energy allocation decision model is optimized using the model optimization parameters to obtain a target energy allocation decision model, which is used for intelligent energy allocation decisions based on target hydropower big data. The process of training the primary energy allocation decision model includes: Obtain the primary energy allocation plan for each target branch in the primary energy allocation decision model; By conducting a one-to-one demand-allocation analysis of the energy digital demands and the primary energy allocation plan of each target branch, the distribution of energy demand profit and loss parameters of each target branch is determined. The energy demand profit and loss parameters are parameters that reflect the situation where the energy digital demands of the target branch exceed or fall below those of the target branch in the primary energy allocation plan. The energy demand surplus / deficit parameter distribution is transformed to generate the gain training parameters, which are used to correct the planned energy allocation. Specifically, the process of feeding back the gain training parameters to the target company for gain learning includes: Within the data decision center, the gain training parameters are used to perform gain learning on the primary energy allocation plan in order to determine the adjusted energy allocation plan for each branch. The adjusted energy allocation plans of each branch company are input into the primary energy allocation decision model for optimization training, and the optimized parameters of the trained model can be obtained.

2. The method of claim 1, wherein, The intelligent decision analysis platform built based on the target company includes: Construct the target company's computing resource supply center; The supply assets of the computing resource supply center are pooled and managed to build a data aggregation center; Intelligent analysis is performed on the aggregated data in the data aggregation center to construct a data analysis service center; Based on the computing resource supply center, planning and decision-making are carried out on the analysis data of the data analysis service center to construct a data decision-making center; The intelligent decision analysis platform is built based on the data aggregation center, the data analysis service center, and the data decision center.

3. The method as described in claim 1, characterized in that, The generation of the primary energy allocation decision model includes: Historical data of the target company is collected to determine historical power generation flow. By performing in-depth analysis of the historical power generation flow, the distribution of upstream water level, downstream water level, and power generation water quality at each historical time point can be determined. Mathematical statistics were performed on the influencing factors of the upstream water level distribution and the historical power generation flow, the downstream water level distribution and the historical power generation flow, and the power generation water quality distribution and the historical power generation flow, respectively. Obtain parameters affecting upstream water level, downstream water level, and water quality.

4. The method as described in claim 3, characterized in that, The method includes: A planning data calibration model is constructed using the upstream water level impact parameters, downstream water level impact parameters, and water quality impact parameters. The target company's energy planning data is input into the planning data calibration model for intelligent calibration of the influencing parameters, in order to generate the primary energy allocation decision model.

5. The method as described in claim 1, characterized in that, The primary energy allocation plan for each target branch includes: The historical energy planning distribution of the target company for the target branch was collected and obtained; The historical energy usage feedback data distribution of the target branch was collected and obtained. Based on the distribution of historical energy use feedback data, the historical energy planning distribution is revised, and the primary energy allocation plan is determined based on the revision results.

6. A high-dimensional, multi-objective, big data intelligent decision-making system for hydropower, characterized in that: The system includes: The platform building module is used to build an intelligent decision analysis platform based on the target company, wherein the intelligent decision analysis platform includes the topology of the target company and each target branch company; A data acquisition module is used to collect data on the target company's hydropower energy plan in order to obtain the target company's energy planning data. A model generation module is used to upload the target company's energy planning data to the intelligent decision analysis platform to generate a primary energy allocation decision model. The model training module is used to distribute the primary energy allocation decision model to each target branch, and to train the primary energy allocation decision model one by one using the energy digital requirements of each target branch. The parameter determination module is used to obtain the gain training parameters of each target branch, and to send the gain training parameters back to the target company for gain learning in order to determine the model optimization parameters. The intelligent decision-making module is used to optimize the primary energy allocation decision model using the model optimization parameters to obtain a target energy allocation decision model for intelligent energy allocation decision-making based on target hydropower big data. The allocation planning unit is used to obtain the primary energy allocation plan for each target branch in the primary energy allocation decision model. The profit and loss parameter distribution determination unit is used to determine the energy demand profit and loss parameter distribution of each target branch by performing a one-to-one demand-allocation analysis on the energy digital demand of each target branch and the primary energy allocation plan. The energy demand profit and loss parameter is a parameter that reflects the situation where the energy digital demand of each target branch exceeds or falls below that of the target branch in the primary energy allocation plan. A gain parameter generation unit is used to perform parameter transformation on the energy demand surplus and deficit parameter distribution to generate the various gain training parameters, which are used to correct the planned energy allocation. A gain learning unit is used to perform gain learning on the primary energy allocation plan within the data decision center using the gain training parameters, in order to determine the adjusted energy allocation plan for each branch company. An optimization training unit is used to input the adjusted energy allocation plans of each branch company into the primary energy allocation decision model for optimization training, thereby obtaining the optimized parameters of the trained model.

Images