Clean energy consumption optimization method and system considering computerized collaboration

Abstract

The invention provides a clean energy consumption optimization method considering computerized collaboration. The method comprises the following steps: constructing a computerized collaboration system architecture; constructing a multi-subject joint optimization multi-objective function; performing multi-period joint optimization scheduling and computing power center site selection on the basis of a multi-body joint optimization objective function in combination with power generation constraints and power grid constraints; and outputting the optimized power dispatching strategy and the computing power resource configuration scheme to realize cross-regional consumption of clean energy and minimization of the total cost of the power grid. According to the invention, two modules of a receiving-end power grid and a sending-end power grid are constructed, and cross-regional and cross-time optimization configuration of clean energy is realized through intelligent scheduling and control of an information network; the computing power center is responsible for coordinating resources of all parties and ensuring that the clean energy can be conveyed to the demand end in a most economical and most efficient mode, so that wide consumption and utilization of the clean energy are effectively promoted, and powerful support is provided for constructing a green and low-carbon energy system.

Description

Technical Field

[0001] This invention belongs to the field of energy absorption capacity assessment technology, specifically relating to a clean energy absorption optimization method and system that considers computer-aided collaboration. Background Technology

[0002] With the advancement of the "dual carbon" target, the proportion of clean energy sources such as wind and solar power in the power system is constantly increasing. However, their output is intermittent and fluctuating, making it difficult to balance power supply and demand in real time. This results in low utilization rates of clean energy, and the structural contradiction of "wind and solar curtailment" coexisting with power shortages in the east is becoming increasingly prominent. In terms of resource matching, computing infrastructure is mainly concentrated in eastern load centers, relying on fossil fuel power, while western renewable energy-rich areas face insufficient computing power demand, affecting the absorption of green electricity. At the system coordination level, computing power scheduling is performance-oriented, while the power system aims for frequency stabilization and peak regulation. The lack of a unified optimization framework results in a 3%-5% loss in renewable energy utilization. Technically, the heterogeneity of computing power systems and the volatility of power systems are difficult to reconcile using traditional control models, leading to low efficiency in cross-domain coordination. Therefore, how to optimize the absorption of clean energy and reduce the total operating cost of the power grid through scheduling and computing center location is an urgent problem to be solved.

[0003] A search revealed Chinese invention patent CN118137572A, which discloses a method for improving clean energy absorption capacity based on power load allocation. First, a support vector regression model is used to predict power generation, and a clustering algorithm is used to handle data anomalies and missing data. The prediction error is corrected using the normal distribution assumption. Then, based on the corrected power, the real-time energy state of the energy storage system is calculated, taking into account parameters such as self-discharge rate and charge / discharge efficiency. Finally, by combining the energy state values ​​during peak and off-peak periods with the time difference and power utilization rate, an absorption rate index is constructed. Optimizing energy storage configuration and scheduling strategies maintains the absorption rate above a preset value, thereby suppressing load fluctuations, reducing wind and solar curtailment, and improving the efficiency of clean energy utilization. This technical solution suffers from several application shortcomings: the prediction model relies excessively on data quality and precise parameter settings, and noise in the actual data can easily lead to performance degradation; it assumes that the prediction error follows a normal distribution, but in reality, factors such as weather and equipment can cause the error distribution to deviate from the assumption, affecting the reliability of correction; the energy storage state calculation involves multiple time-varying parameters, and inaccurate calibration can lead to estimation bias; the method focuses on offline processing and lacks a real-time response mechanism, making it difficult to cope with sudden fluctuations; the algorithm integrates clustering, regression, and probabilistic correction, resulting in computational complexity and difficulty in deployment in resource-constrained scenarios. These limitations collectively restrict its practical application effectiveness in complex power grid environments.

[0004] A search revealed Chinese invention patent CN120013108A, which discloses a method and system for assessing clean energy absorption capacity based on transmission-receiver collaboration. Centered on "transmission-receiver collaboration," the system aggregates multi-source data from power generation, grid, load, and storage in real time. After cleaning, alignment, and fusion, it extracts time-frequency features and constructs a graph attention network to dynamically assess the renewable energy generation capacity at the sending end and simultaneously predict multi-scale loads at the receiving end. The system quantifies the absorption space online using the difference between "generation forecast" and "load forecast," and continuously optimizes model parameters. Based on this, the system generates scheduling strategies within seconds: during surplus periods, it increases clean energy output, charges energy storage, and shifts peak demand; during shortage periods, it reduces output, releases energy storage, activates reserves, and optimizes power flow, achieving closed-loop collaboration between power generation, grid, load, and storage, significantly improving renewable energy utilization and grid stability. Although the patent constructs a clean energy consumption assessment system that coordinates the sending and receiving ends, it still has five shortcomings: First, the algorithm stack is complex, with graph neural networks, time series analysis, and optimization coupling leading to a large computational load, making it difficult for edge devices or provincial dispatch centers to support real-time operation; second, it is highly dependent on high-quality synchronous data, and the absence, jumps, or delays of on-site sensors and smart meters will directly amplify prediction errors and lead to misjudgments; third, the model has weak generalization ability, with significant differences in regional power grid structures and renewable energy composition, requiring retraining when changing regions, resulting in high migration costs; fourth, the dispatch strategy is difficult to implement, as electricity price guidance and user peak shifting involve market mechanisms and policy support, and simple technical instructions are unlikely to be effective quickly; fifth, extreme scenarios and interpretability are lacking, and the system's robustness is not verified in the face of fluctuations such as typhoons and equipment cascading failures, and the black-box decision-making of deep learning lacks transparent basis, making it difficult for operators to trace and adjust, or affecting supervision and trust. Summary of the Invention

[0005] To address the aforementioned issues, this invention provides a method and system for optimizing clean energy consumption by considering computer-based collaboration. The aim is to optimize the allocation and utilization of clean energy through the synergistic effect of highly integrated information technology and power technology. By constructing a "computer-based collaboration" model, and utilizing the deep integration of information networks and power networks, joint optimization scheduling and computing center location are performed on the sending and receiving end power grids. The system comprehensively considers the operating costs of various energy sources such as thermal power, new energy power generation, and hydropower (including pumped storage), as well as the costs of purchasing and selling electricity, the cost of curtailment assessment, and the cost of inter-regional communication, minimizing the net cost of the power grid. Thus, while satisfying various power grid constraints, it improves the inter-regional consumption level of clean energy and reduces the total operating cost of the power grid.

[0006] The specific plan is as follows:

[0007] A clean energy consumption optimization method considering computer-aided collaboration, the method includes:

[0008] S1. Construct a computer-aided collaborative system architecture, including a sending-end power grid module and a receiving-end power grid module. Deploy data centers in the sending-end power grid module and the receiving-end power grid module respectively, and perform data interaction and computing power scheduling.

[0009] S2. Construct a multi-entity joint optimization multi-objective function, including thermal power operation cost, new energy power generation operation cost, hydropower operation cost, net cost of purchasing and selling electricity, curtailment assessment cost, and cross-regional communication cost;

[0010] S3. Based on the multi-entity joint optimization objective function, and combining power generation constraints and grid constraints, perform multi-time period joint optimization scheduling and computing center location selection;

[0011] S4. Output the optimized power dispatch strategy and computing resource allocation scheme to achieve cross-regional consumption of clean energy and minimize the total grid cost. Further, the sending-end grid module is deployed in a clean energy-rich area, and the receiving-end grid module is deployed in a load-concentrated area. Further, the multi-objective function for multi-entity joint optimization in step S2 is as follows:

[0012] ;

[0013] ;

[0014] in, For the power grid serial number, Indicates the net cost of the power grid. , , , , , These represent the operating costs of thermal power plants, renewable energy generation plants, hydropower plants, net electricity purchase and sale costs, curtailment assessment costs, and inter-regional communication costs, respectively. Further, the operating costs of thermal power plants include ignition point fuel costs and start-up and shutdown costs, as follows: ;

[0015] ;

[0016] ;

[0017] in, Indicates the cost of fuel for thermal power generation. Indicates start-up and shutdown costs. Indicates the unit of time. Indicates the serial number of the thermal power unit. This indicates the total number of thermal power units. This indicates the unit's status, where 0 represents shutdown and 1 represents startup. Indicates the output power of the thermal power unit. , , They represent the fitting parameters, , , These represent the unit price of coal, the cost of a single unit start-up, and the cost of a single unit shutdown, respectively. The operating costs of the new energy power generation are the operating costs of wind power and photovoltaic power generation, as follows: ;

[0018] in, Indicates wind power output. This indicates the unit operating cost of wind power. Indicates photovoltaic power generation output. This represents the unit operating cost of photovoltaic power generation; the hydropower operating cost refers to the operating cost of pumped storage hydropower stations, as follows: ;

[0019] in, Indicates the number of pumped storage power stations. Indicates the pumping power of the hydroelectric power station. This represents the unit operating cost of pumped storage. Indicates the operating power of the hydropower station. This represents the unit operating cost of the hydropower station; the net cost of purchasing and selling electricity is as follows: ;

[0020] in, , These represent the purchase price and the sales price of electricity, respectively. , These represent the receiving power and transmitting power of the communication channel, respectively; the cost of power curtailment assessment is as follows: ;

[0021] in, This indicates the maximum available hydropower generation. , These represent the predicted maximum output of hydropower, wind power, and photovoltaic power generation, respectively.

[0022] The cross-regional communication cost refers to the additional cost of the power grid where the data center is located. Assuming the communication cost of the data center in the eastern region is 0, as follows:

[0023] ;

[0024] ;

[0025] In the formula, This indicates the power consumption of the computing center.

[0026] Furthermore, the power generation constraints in step S3 include thermal power constraints, hydropower constraints, termination of new energy sources, and pumped storage constraints; the thermal power constraints are as follows: ;

[0027] ;

[0028] ;

[0029] in, , These represent the maximum ramp rate and maximum descent rate of the thermal power unit, respectively. , These represent the number of consecutive start-up and consecutive shutdown periods of a thermal power unit from the last start-up / shutdown state transition time to time period t, respectively. , These represent the minimum number of consecutive operating periods and the minimum number of consecutive downtime periods, respectively; the hydropower constraints are as follows: ;

[0030] ;

[0031] ;

[0032] ;

[0033] in, , These represent the upper and lower limits of water consumption for hydropower generation, , These represent the upper and lower limits of the water level, respectively. , These represent the initial and final water storage volumes, respectively. , These represent the inflow and outflow of water, respectively; the constraints for new energy sources are as follows: ;

[0034] ;

[0035] in, , These represent the minimum output of wind power and the minimum output of photovoltaic power, respectively; the pumped storage constraints are as follows: ;

[0036] ;

[0037] ;

[0038] in, , These represent the minimum and maximum pumping power, respectively. , These represent the minimum and maximum power generation, respectively. This represents the energy conversion efficiency of the pumped storage power station. Further, the grid constraints in step S3 include power balance constraints, spinning reserve constraints, and channel constraints; the power balance constraints are as follows: ; in, This represents the real-time power of the power grid; the spinning reserve constraints are as follows: ;

[0039] ;

[0040] in, , These represent the positive and negative spinning reserve capacity requirements, respectively; the channel constraints are as follows: ;

[0041] ;

[0042] in, , These represent the minimum and maximum power limits allowed for grid interconnection channels, respectively.

[0043] This invention also provides a clean energy consumption optimization system considering computer-computer collaboration, comprising: a receiving-end grid module, deployed in a load-concentrated area, consisting of several grid nodes, each equipped with a data center for collecting, processing, and analyzing local power data, and interconnected with each other; a sending-end grid module, deployed in a clean energy-rich area, also consisting of several grid nodes, each equipped with a data center for monitoring the operating status of local power generation resources and interacting with the receiving-end grid's data center; and a computing center module for performing multi-agent joint optimization of multi-objective functions, outputting a clean energy consumption optimization scheme, and performing power dispatch.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] 1. Two main modules, the receiving-end power grid and the sending-end power grid, were constructed. Within these modules, the network is closely connected to various power generation resources. Through intelligent scheduling and control of the information network, the cross-regional and cross-time-period optimized allocation of clean energy was achieved. 2. As the core manager of the information network, the computing center is responsible for coordinating resources from all parties to ensure that clean energy can be delivered to the demand side in the most economical and efficient manner. This effectively promotes the widespread consumption and utilization of clean energy and provides strong support for building a green and low-carbon energy system. Attached Figure Description

[0046] Figure 1 This is a flowchart of the optimization method of the present invention; Figure 2 This is a diagram of the optimized system architecture of the present invention. Detailed Implementation

[0047] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. The specific implementation methods of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0048] Example 1:

[0049] like Figure 1 As shown in the figure, this embodiment is a clean energy consumption optimization method considering computer-aided collaboration. The method includes:

[0050] S1. Construct a computer-aided collaborative system architecture, including a sending-end power grid module and a receiving-end power grid module. Deploy data centers in the sending-end power grid module and the receiving-end power grid module respectively, and perform data interaction and computing power scheduling. The sending-end power grid module is deployed in a clean energy-rich area, and the receiving-end power grid module is deployed in a load-concentrated area.

[0051] S2. Construct a multi-entity joint optimization multi-objective function, including thermal power operating costs, renewable energy power generation operating costs, hydropower operating costs, net electricity purchase and sale costs, curtailment assessment costs, and inter-regional communication costs; the multi-entity joint optimization multi-objective function is as follows:

[0052] ;

[0053] ;

[0054] in, For the power grid serial number, Indicates the net cost of the power grid. , , , , , These represent the operating costs of thermal power plants, renewable energy generation, hydropower, net electricity purchase and sale, curtailment assessment costs, and inter-regional communication costs, respectively. Thermal power plant operating costs include ignition point fuel costs and start-up and shutdown costs, as follows: ;

[0055] ;

[0056] ;

[0057] in, Indicates the cost of fuel for thermal power generation. Indicates start-up and shutdown costs. Indicates the unit of time. Indicates the serial number of the thermal power unit. This indicates the total number of thermal power units. This indicates the unit's status, where 0 represents shutdown and 1 represents startup. Indicates the output power of the thermal power unit. , , They represent the fitting parameters, , , These represent the unit price of coal, the cost of a single unit start-up, and the cost of a single unit shutdown, respectively. The operating costs of new energy power generation are the operating costs of wind power and photovoltaic power generation, as follows: ;

[0058] in, Indicates wind power output. This indicates the unit operating cost of wind power. Indicates photovoltaic power generation output. This represents the unit operating cost of photovoltaic power generation; the operating cost of hydropower is the operating cost of pumped storage hydropower stations, as follows: ;

[0059] in, Indicates the number of pumped storage power stations. Indicates the pumping power of the hydroelectric power station. This represents the unit operating cost of pumped storage. Indicates the operating power of the hydropower station. This represents the unit operating cost of a hydropower station; the net cost of purchasing and selling electricity is as follows: ;

[0060] in, , These represent the purchase price and the sales price of electricity, respectively. , These represent the receiving power and transmitting power of the connection channel, respectively; the cost of power curtailment assessment is as follows: ;

[0061] in, This indicates the maximum available hydropower generation. , These represent the predicted maximum output of hydropower, wind power, and photovoltaic power generation, respectively.

[0062] Cross-regional communication costs are additional costs incurred by the power grid where the data center is located. Assuming the data center's communication costs are zero in the eastern region, the following is an example:

[0063] ;

[0064] ;

[0065] In the formula, This indicates the power consumption of the computing center.

[0066] S3. Based on a multi-agent joint optimization objective function, and combining generation constraints and grid constraints, perform multi-time-period joint optimization scheduling and computing center location selection; Generation constraints include thermal power constraints, hydropower constraints, new energy termination constraints, and pumped storage constraints; Thermal power constraints are as follows: ;

[0067] ;

[0068] ;

[0069] in, , These represent the maximum ramp rate and maximum descent rate of the thermal power unit, respectively. , These represent the number of consecutive start-up and consecutive shutdown periods of a thermal power unit from the last start-up / shutdown state transition time to time period t, respectively. , These represent the minimum number of consecutive operating periods and the minimum number of consecutive downtime periods, respectively; the hydropower constraints are as follows: ;

[0070] ;

[0071] ;

[0072] ;

[0073] in, , These represent the upper and lower limits of water consumption for hydropower generation, , These represent the upper and lower limits of the water level, respectively. , These represent the initial and final water storage volumes, respectively. , These represent the inflow and outflow of water, respectively; the constraints for new energy sources are as follows: ;

[0074] ;

[0075] in, , These represent the minimum output of wind power and the minimum output of photovoltaic power, respectively; the pumped storage constraints are as follows: ;

[0076] ;

[0077] ;

[0078] in, , These represent the minimum and maximum pumping power, respectively. , These represent the minimum and maximum power generation, respectively. This represents the energy conversion efficiency of the pumped storage power station. Grid constraints include power balance constraints, spinning reserve constraints, and channel constraints; the power balance constraints are as follows: ; in, This represents the real-time power of the power grid; the spinning reserve constraints are as follows: ;

[0079] ;

[0080] in, , These represent the positive and negative spinning reserve capacity requirements, respectively; the channel constraints are as follows: ;

[0081] ;

[0082] in, , These represent the minimum and maximum power limits allowed for grid interconnection channels, respectively.

[0083] S4. Output the optimized power dispatch strategy and computing resource allocation scheme to achieve cross-regional consumption of clean energy and minimize the total grid cost. Example 2: This example provides a clean energy consumption optimization system considering computer-computer collaboration, used to implement the optimization method in Example 1, such as... Figure 2 As shown, the system includes: a receiving-end grid module, deployed in a load-concentrated area, consisting of several grid nodes, each equipped with a data center for collecting, processing, and analyzing local power data, and interconnected with each other; a sending-end grid module, deployed in a clean energy-rich area, also consisting of several grid nodes, each equipped with a data center for monitoring the operating status of local power generation resources and interacting with the receiving-end grid's data center; and a computing center module, used to perform multi-agent joint optimization and multi-objective function optimization, output clean energy consumption optimization schemes, and perform power dispatching.

[0084] The above embodiments are for illustrative purposes only and are not intended to limit the scope of this invention. Although this invention has been described in detail with reference to the embodiments, those skilled in the art should understand that various combinations, modifications, or equivalent substitutions of the technical solutions of this invention do not depart from the spirit and scope of the technical solutions of this invention and should be covered within the scope of the claims of this invention.

Claims

1. A method for optimizing clean energy consumption considering computer-aided collaboration, characterized in that the method... include: S1. Construct a computer-aided collaborative system architecture, including a sending-end power grid module and a receiving-end power grid module. Deploy data centers in the sending-end power grid module and the receiving-end power grid module respectively, and perform data interaction and computing power scheduling. S2. Construct a multi-entity joint optimization multi-objective function, including thermal power operation cost, new energy power generation operation cost, hydropower operation cost, net cost of purchasing and selling electricity, curtailment assessment cost, and cross-regional communication cost; S3. Based on the multi-entity joint optimization objective function, and combining power generation constraints and grid constraints, perform multi-time period joint optimization scheduling and computing center location selection; S4 outputs optimized power dispatch strategies and computing resource allocation schemes to achieve cross-regional consumption of clean energy and minimize the total cost of the power grid.

2. The method for optimizing clean energy consumption considering computer-aided collaboration according to claim 1, characterized in that, The sending-end power grid module is deployed in a clean energy-rich area, and the receiving-end power grid module is deployed in a load-concentrated area.

3. The method for optimizing clean energy consumption considering computer-aided collaboration according to claim 2, characterized in that, The multi-agent joint optimization of the multi-objective function in step S2 is as follows: ； ； in, For the power grid serial number, Indicates the net cost of the power grid. , , , , , These represent the operating costs of thermal power, the operating costs of new energy power generation, the operating costs of hydropower, the net cost of purchasing and selling electricity, the cost of curtailment assessment, and the cost of inter-regional communication, respectively.

4. The method for optimizing clean energy consumption considering computer-aided collaboration according to claim 3, characterized in that, The operating cost of thermal power plants includes ignition fuel cost and start-up and shutdown costs, as follows: ; ； ； in, Indicates the cost of fuel for thermal power generation. Indicates start-up and shutdown costs. Indicates the unit of time. Indicates the serial number of the thermal power unit. This indicates the total number of thermal power units. This indicates the unit's status, where 0 represents shutdown and 1 represents startup. Indicates the output power of the thermal power unit. , , They represent the fitting parameters, , , These represent the unit price of coal, the cost of a single unit start-up, and the cost of a single unit shutdown, respectively. The operating costs of the new energy power generation are the operating costs of wind power and photovoltaic power generation, as follows: ; in, Indicates wind power output. This indicates the unit operating cost of wind power. Indicates photovoltaic power generation output. This represents the unit operating cost of photovoltaic power generation; the hydropower operating cost refers to the operating cost of pumped storage hydropower stations, as follows: ; in, Indicates the number of pumped storage power stations. Indicates the pumping power of the hydroelectric power station. This represents the unit operating cost of pumped storage. Indicates the operating power of the hydropower station. This represents the unit operating cost of the hydropower station; the net cost of purchasing and selling electricity is as follows: ; in, , These represent the purchase price and the sales price of electricity, respectively. , These represent the receiving power and transmitting power of the communication channel, respectively; the cost of power curtailment assessment is as follows: ; in, This indicates the maximum available hydropower generation. , These represent the predicted maximum output of hydropower, wind power, and photovoltaic power generation, respectively. The cross-regional communication cost refers to the additional cost of the power grid where the data center is located. Assuming the communication cost of the data center in the eastern region is 0, as follows: ； ； In the formula, This indicates the power consumption of the computing center.

5. The method for optimizing clean energy consumption considering computer-aided collaboration according to claim 3, characterized in that, The power generation constraints in step S3 include thermal power constraints, hydropower constraints, termination of new energy sources, and pumped storage constraints; the thermal power constraints are as follows: ; ； ； in, , These represent the maximum ramp rate and maximum descent rate of the thermal power unit, respectively. , These represent the number of consecutive start-up and consecutive shutdown periods of a thermal power unit from the last start-up / shutdown state transition time to time period t, respectively. , These represent the minimum number of consecutive operating periods and the minimum number of consecutive downtime periods, respectively; the hydropower constraints are as follows: ; ； ； ； in, , These represent the upper and lower limits of water consumption for hydropower generation, , These represent the upper and lower limits of the water level, respectively. , These represent the initial and final water storage volumes, respectively. , These represent the inflow and outflow of water, respectively; New energy constraints are as follows: ; ； in, , These represent the minimum output of wind power and the minimum output of photovoltaic power, respectively; the pumped storage constraints are as follows: ; ； ； in, , These represent the minimum and maximum pumping power, respectively. , These represent the minimum and maximum power generation, respectively. This indicates the energy conversion efficiency of a pumped storage power station.

6. The method for optimizing clean energy consumption considering computer-aided collaboration according to claim 5, characterized in that, The grid constraints in step S3 include power balance constraints, spinning reserve constraints, and channel constraints; the power balance constraints are as follows: ; in, This represents the real-time power of the power grid; the spinning reserve constraints are as follows: ; ； in, , These represent the positive and negative spinning reserve capacity requirements, respectively; the channel constraints are as follows: ; ； in, , These represent the minimum and maximum power limits allowed for grid interconnection channels, respectively.

7. A clean energy consumption optimization system considering computer-aided collaboration for implementing the method as described in any one of claims 1 to 6, characterized in that, include: The receiving-end grid module, deployed in areas with concentrated loads, consists of several grid nodes, each equipped with a data center for collecting, processing, and analyzing local power data, and is interconnected. The sending-end grid module, deployed in areas rich in clean energy, also consists of several grid nodes, each equipped with a data center for monitoring the operational status of local power generation resources and interacting with the receiving-end grid's data center. The computing center module is used to perform multi-agent joint optimization and multi-objective function optimization, output optimized clean energy consumption schemes, and perform power dispatching.

Images