A multi-park energy control method and system

Abstract

The present invention provides a multi-park energy control method and system, which relates to the field of energy control technology. The method includes: obtaining energy consumption data of each park in a preset current time period and resource flow data of each component structure in the park; taking each park as a central park, and determining other parks related to the central park according to the resource flow data, taking other parks as associated parks, and forming an associated network centered on the central park; obtaining future energy consumption data of the central park in a preset future time period through the associated network; generating an energy control strategy based on the future energy consumption data and performing energy control based on the energy control strategy. The present invention realizes refined control of energy consumption by comprehensively considering energy consumption data within the park and between parks, and predicting future energy consumption.

Description

Technical Field

[0001] The present invention relates to the technical field of energy regulation and control, and in particular to a multi-park energy regulation and control method and system. Background Art

[0002] At present, the energy regulation of the park usually focuses on examining the energy consumption data of the park's internal energy use, industrial production, building operation, transportation, infrastructure, etc. Based on the energy consumption data, the corresponding park energy regulation strategy is established to improve the carbon emissions of each park through the park energy regulation strategy.

[0003] In the existing technology, since the energy consumption of the park is affected by many factors, including the impact of industrial production, building operations, population activities and transportation in other parks, this impact is accompanied by the redistribution of energy and changes in consumption patterns, making the energy consumption in the park always fluctuate, which in turn leads to the energy control strategy based on carbon emissions generation failing to achieve the desired control effect. Summary of the invention

[0004] The problem solved by the present invention is how to improve the energy control accuracy of a park.

[0005] In order to solve the above problems, the present invention provides a multi-park energy control method and system.

[0006] In a first aspect, the present invention provides a multi-park energy control method, comprising:

[0007] Obtaining energy consumption data of each park in a preset current time period and resource flow data of each component structure in the park; wherein the energy consumption data includes energy consumption sub-data of each component structure in the park and emission factors corresponding to the component structure;

[0008] Obtaining total resource flow data of the park according to the resource flow data of the component structure of each park;

[0009] Taking each of the parks as a central park, and determining other parks related to the central park according to the total resource flow data of each central park, and taking the other parks as associated parks;

[0010] Forming an associated network centered on the central park according to all the associated parks and the central park;

[0011] According to the energy consumption data of each of the parks in the associated network, obtaining the future total energy consumption data of the associated network in a preset future time period;

[0012] Determine the weight value of each associated park in the associated network through the associated network;

[0013] Combining the future total energy consumption data with the weight value of the associated park, obtaining the future energy consumption data of the central park in the preset future time period;

[0014] An energy control strategy is generated according to the future energy consumption data, and energy control is performed according to the energy control strategy.

[0015] Optionally, taking each of the parks as a central park, and determining other parks related to the central park according to the total resource flow data of each of the central parks, and taking the other parks as associated parks, includes:

[0016] Take each of the parks as a central park, and determine the flow characteristic index between the component structure of the central park and the component structures of other parks according to the total resource flow direction data of the central park;

[0017] Obtaining the allocation weight corresponding to each of the component structures of the central park;

[0018] According to the association value model, the sum of the products of the allocation weight of each component structure and the flow characteristic index of the component structure is used as the association value between the central park and the other parks;

[0019] According to the association value, a plurality of associated parks are obtained by screening from the other parks.

[0020] Optionally, the selecting, according to the association value, from other parks to obtain a plurality of the associated parks includes:

[0021] Determining whether the correlation value is greater than or equal to a preset correlation value;

[0022] If yes, the park corresponding to the associated value is used as the associated park of the central park;

[0023] If not, the park corresponding to the associated value is excluded from the associated parks of the central park.

[0024] Optionally, forming an associated network centered on the central park according to all the associated parks and the central park includes:

[0025] Determining the centrality between the associated parks and the central park according to the associated values ​​of all the associated parks of the central park;

[0026] Determining the relative position of each of the associated parks and the central park according to the centrality;

[0027] According to the relative position corresponding to each of the associated parks, the associated network centered on the central park is generated.

[0028] Optionally, obtaining the future total energy consumption data of the associated network in a preset future time period according to the energy consumption data of each of the parks in the associated network includes:

[0029] Integrate the energy consumption data of each of the parks in the associated network to determine the total energy consumption data of the associated network;

[0030] The total energy consumption data is input into an energy consumption prediction model, and the future total energy consumption data of the associated network in the preset future time period is determined through the output of the energy consumption prediction model.

[0031] Optionally, determining the weight value of each associated park in the associated network through the associated network includes:

[0032] According to the total resource flow data of each park in the associated network, obtaining the future total resource flow data of each park in the preset future time period;

[0033] Determining resource change data of the park according to the future total resource flow data;

[0034] According to the resource change data of the park, adjusting the centrality of the park with the central park in the associated network;

[0035] The weight value of the park in the associated network is determined according to the adjusted centrality.

[0036] Optionally, combining the future total energy consumption data and the weight value of the associated park to obtain the future energy consumption data of the central park in the preset future time period includes:

[0037] Determine the prediction sub-data of each associated park in the preset future time period according to multiplying the future total energy consumption data of the associated network by the weight value of the associated park;

[0038] Obtain the predicted sub-data of the associated park in other associated networks;

[0039] Averaging the predicted sub-data of the associated park in all the associated networks to obtain the energy consumption data of the associated park in the preset future time period;

[0040] The future energy consumption data of the central park in the preset future time period is determined based on the energy consumption data of each of the associated parks and the future total energy consumption data.

[0041] Optionally, generating an energy regulation strategy according to the future energy consumption data and performing energy regulation according to the energy regulation strategy includes:

[0042] Determining the expected control target of the central park according to the future energy consumption data;

[0043] Generate energy control strategies through reinforcement learning combined with expected control targets;

[0044] The energy consumption data of the central park in the preset current time period is taken as the state, the initial energy regulation strategy is taken as the action, and the error value between the energy consumption data after the initial energy regulation strategy is applied and the expected regulation target is taken as the reward;

[0045] When the error value is less than a preset reward threshold, the energy regulation strategy at the corresponding moment is determined to be the final energy regulation strategy, and energy regulation is performed according to the final energy regulation strategy.

[0046] Optionally, the multi-park energy control method further includes:

[0047] The carbon emissions of the park are predicted through a distributed system, wherein the distributed system includes a plurality of nodes, each of which processes a park and the associated network corresponding to the park.

[0048] In a second aspect, the present invention further provides a multi-park energy control system, comprising:

[0049] A data acquisition unit, used to acquire energy consumption data of each park in a preset current time period and resource flow data of each component structure in the park; wherein the energy consumption data includes energy consumption sub-data of each component structure in the park and emission factors corresponding to the component structures;

[0050] A resource analysis unit, configured to obtain total resource flow data of the park according to the resource flow data of the component structure of each park;

[0051] An association analysis unit, used to take each of the parks as a central park, and determine other parks related to the central park according to the total resource flow data of each central park, and take the other parks as associated parks;

[0052] A cluster unit, used for forming an associated network centered on the central park according to all the associated parks and the central park;

[0053] A first data processing unit, configured to obtain future total energy consumption data of the associated network in a preset future time period according to the energy consumption data of each of the parks in the associated network;

[0054] A second data processing unit is used to determine the weight value of each associated park in the associated network through the associated network;

[0055] An energy consumption analysis unit, configured to combine the future total energy consumption data and the weight value of the associated park to obtain the future energy consumption data of the central park in the preset future time period;

[0056] The control unit is used to generate an energy control strategy according to the future energy consumption data and perform energy control according to the energy control strategy.

[0057] The energy control method and system of multiple parks of the present invention can obtain energy consumption data and resource flow data of each park within a preset time period, so as to have a detailed understanding of the energy use of each park. These data include energy consumption sub-data of each component structure in the park and the corresponding emission factor, thereby providing a basis for subsequent analysis. Then, by analyzing the resource flow data of each park, the total resource flow data of the park is obtained, which helps to understand the flow of resources within the park and provides a macro perspective for further analysis. Then, each park is taken as a central park, and the relationship with other parks is determined according to its total resource flow data, and an association network is constructed. This network includes not only the central park, but also the associated parks related to it. In this way, the mutual dependence and influence between parks can be identified, so as to perform energy control at a more macro level. By constructing an association network, not only the energy consumption and resource flow of a single park are paid attention to, but the park is regarded as a node in an overall network, and the energy flow and consumption of each node are interrelated with other nodes. On the basis of the association network, by determining the weight value of each associated park, energy consumption can be allocated more accurately to ensure the reasonable distribution of energy between parks. Finally, by combining the future total energy consumption data and the weight values ​​of the associated parks, the energy consumption data of the central park in the future time period can be obtained. This enables the control strategy to more accurately guide the use and allocation of energy, thereby improving the accuracy and efficiency of energy control. The present invention can effectively avoid the problem of a single park being affected by other parks and reducing the control accuracy through a multi-park energy control method. By comprehensively considering the energy consumption data within the park and between parks, and predicting future energy consumption, refined control of energy consumption is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 This is one of the flow charts of the multi-park energy control method of the present invention;

[0059] Figure 2 This is the second flow chart of the multi-park energy control method of the present invention;

[0060] Figure 3 This is the third flow chart of the multi-park energy control method of the present invention;

[0061] Figure 4 This is the fourth flow chart of the multi-park energy control method of the present invention;

[0062] Figure 5 This is the fifth flow chart of the multi-park energy control method of the present invention;

[0063] Figure 6 It is a schematic diagram of the structure of the multi-park energy control system of the present invention. DETAILED DESCRIPTION

[0064] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0065] Combination Figure 1 As shown, the present invention provides a multi-park energy control method, comprising:

[0066] Obtain energy consumption data of each park in a preset current time period and resource flow data of each component structure in the park; wherein the energy consumption data includes energy consumption sub-data of each component structure in the park and emission factors corresponding to the component structures.

[0067] Specifically, the energy consumption data of each park in the preset current time period and the resource flow data of each component structure in the park are collected. Among them, the energy consumption data includes the energy consumption sub-data of each component structure and the corresponding emission factor. These data are the basis of energy regulation. Through these data, the energy usage and environmental impact of each park can be understood. Therefore, in the preferred embodiment of the present invention, implementing this step usually requires the deployment of sensors and monitoring equipment, and the use of software systems to collect and store these data.

[0068] Exemplarily, in the present invention, the composition structure of the park refers to different units or parts within the park, each of which has a direct impact on energy consumption and resource flow; these composition structures may include but are not limited to the following: Buildings: including residential buildings, office buildings, schools, hospitals, etc., and the energy consumption sub-data of these buildings may include energy usage in heating, cooling, lighting, elevator operation, etc. Industrial facilities: such as factories, production workshops, warehouses, etc., these facilities consume a lot of energy during the production process and generate corresponding emissions. Transportation system: including roads, public transportation systems (such as electric buses, shared bicycles), logistics and transportation, etc. within the park, and the operation of these systems requires energy support. Infrastructure: such as water supply systems, power supply systems (including power grids), gas supply systems (such as natural gas pipelines), communication networks and waste treatment systems, etc., these infrastructures provide support for the normal operation of the park. Commercial and retail areas: including shopping malls, supermarkets, catering and entertainment facilities, etc., the operation of these areas requires energy support, such as refrigeration, lighting, air conditioning, etc. Green space and open space: such as parks, gardens, stadiums, etc., these areas will involve energy consumption such as irrigation, lighting and maintenance. Energy production facilities: such as solar photovoltaic panels, wind turbines, biomass power plants, etc., which produce renewable energy within the park. Energy storage and distribution systems: such as battery energy storage systems, smart grids, regional heating and cooling networks, etc., which affect the efficient use and distribution of energy. Residential areas: including employee dormitories, residential buildings, etc., the energy consumption in these areas comes from the daily life of residents, such as heating, cooling, cooking and hot water supply. At the same time, in order to collect energy consumption data and resource flow data of these components, the following devices can be used: smart meters: used to measure and record electricity consumption; gas meters: used to measure and record gas consumption; water meters: used to measure and record water resource consumption; heat meters: used to measure and record heat energy consumption; sensors: used to monitor indoor environmental parameters such as temperature, humidity, light intensity; emission monitoring equipment: used to measure greenhouse gas emissions and other pollutant emissions. Software system: used to integrate, store and analyze the data collected by the above devices. Through these devices and systems, the energy use and environmental impact of each park can be understood in detail, providing accurate data support for energy regulation.

[0069] According to the resource flow data of the component structure of each of the parks, the total resource flow data of the park is obtained.

[0070] Specifically, after collecting the resource flow data of each component structure, these data are aggregated to obtain the total resource flow data of each park. This involves data processing and analysis to ensure that the data of all component structures are correctly aggregated. In a preferred embodiment of the present invention, a data integration tool and a database management system are used to facilitate the integration and query of data.

[0071] Each of the parks is regarded as a central park, and based on the total resource flow data of each central park, other parks related to the central park are determined, and the other parks are regarded as associated parks.

[0072] Specifically, each park is regarded as a central park, and other parks related to it are determined based on its total resource flow data. These related parks are called associated parks. In a preferred embodiment of the present invention, network analysis technology in graph theory can be applied to identify and define the mutual relationship between parks. In this way, an associated network with each central park as the core can be constructed.

[0073] An associated network centered on the central park is formed based on all the associated parks and the central park.

[0074] Specifically, after determining the central park and associated parks, a comprehensive association network needs to be constructed to take into account the energy flow relationships between all parks and parks. This network can be a directed graph, in which nodes represent parks and directed edges represent the direction and size of energy flows. This network can be constructed and analyzed using network modeling tools such as NetworkX or igraph. This network includes not only direct energy flows between parks, but may also include indirect energy dependencies, such as connections through supply chains or shared energy infrastructure.

[0075] According to the energy consumption data of each of the parks in the associated network, the future total energy consumption data of the associated network in a preset future time period is obtained.

[0076] Specifically, in order to predict future total energy consumption data, time series analysis or machine learning models are needed to analyze historical energy consumption data and predict future consumption trends. This may involve the use of algorithms such as ARIMA models, neural networks, or random forests. The forecasting model needs to be updated regularly to reflect the latest energy consumption data and market changes. In addition, external factors such as weather changes, economic fluctuations, or policy changes need to be considered, which may affect energy consumption. The implementation of this step requires professional data analysis and modeling skills, as well as corresponding software tools.

[0077] The weight value of each associated park in the associated network is determined through the associated network.

[0078] Specifically, in the association network, the weight value of each park represents its importance in energy consumption and resource flow. These weight values ​​can be determined by multi-criteria decision analysis methods, where each criterion represents a specific attribute of the park, such as energy consumption, emission factor or resource utilization efficiency. Using methods such as weighted summation or hierarchical analysis, a comprehensive score can be assigned to each park to determine its weight value. The implementation of this step requires a deep understanding of decision analysis methods and the use of professional decision support systems.

[0079] By combining the future total energy consumption data and the weight value of the associated park, the future energy consumption data of the central park in the preset future time period is obtained.

[0080] Specifically, the energy consumption data of the central park in the future time period can be calculated by combining the future total energy consumption data and the weight values ​​of the associated parks. This involves a weighted average calculation, in which the energy consumption data of each associated park is weighted according to its weight value. In a preferred embodiment of the present invention, data analysis tools such as Excel, R or Python can be used to perform these calculations. The key to this step is to ensure the accuracy and rationality of the weight value in order to provide reliable forecast data for energy regulation.

[0081] An energy control strategy is generated according to the future energy consumption data, and energy control is performed according to the energy control strategy.

[0082] Specifically, energy control strategies are generated based on future energy consumption data. This involves using optimization algorithms, such as linear programming, dynamic programming, or genetic algorithms, to determine the most effective energy allocation scheme. These strategies need to consider factors such as energy costs, environmental impact, and reliability. The generated strategies should be operational and can guide energy use and allocation in actual operations. The implementation of this step requires interdisciplinary knowledge, including energy management, economics, and environmental science, as well as the practical ability to implement these strategies.

[0083] The energy control method of multiple parks of the present invention can obtain energy consumption data and resource flow data of each park within a preset time period, so as to have a detailed understanding of the energy use of each park. These data include energy consumption sub-data of each component structure in the park and the corresponding emission factor, thereby providing a basis for subsequent analysis. Then, by analyzing the resource flow data of each park, the total resource flow data of the park is obtained, which helps to understand the flow of resources within the park and provides a macro perspective for further analysis. Then, each park is taken as a central park, and the relationship with other parks is determined according to its total resource flow data, and an association network is constructed. This network includes not only the central park, but also the associated parks related to it. In this way, the mutual dependence and influence between parks can be identified, so as to perform energy control at a more macro level. By constructing an association network, not only the energy consumption and resource flow of a single park are focused on, but the park is regarded as a node in an overall network, and the energy flow and consumption of each node are interrelated with other nodes. On the basis of the association network, by determining the weight value of each associated park, energy consumption can be allocated more accurately to ensure the reasonable distribution of energy between parks. Finally, by combining the future total energy consumption data and the weight values ​​of the associated parks, the energy consumption data of the central park in the future time period can be obtained. This enables the control strategy to more accurately guide the use and allocation of energy, thereby improving the accuracy and efficiency of energy control. The present invention can effectively avoid the problem of a single park being affected by other parks and reducing the control accuracy through a multi-park energy control method. By comprehensively considering the energy consumption data within the park and between parks, and predicting future energy consumption, refined control of energy consumption is achieved.

[0084] Optionally, combined Figure 2 As shown, each of the parks is taken as a central park, and other parks related to the central park are determined according to the total resource flow data of each central park, and the other parks are taken as associated parks, including:

[0085] Take each of the parks as a central park, and determine the flow characteristic index between the component structure of the central park and the component structures of other parks according to the total resource flow direction data of the central park;

[0086] Obtaining the allocation weight corresponding to each of the component structures of the central park;

[0087] According to the association value model, the sum of the products of the allocation weight of each component structure and the flow characteristic index of the component structure is used as the association value between the central park and the other parks;

[0088] According to the association value, a plurality of associated parks are obtained by screening from the other parks.

[0089] Specifically, each park is first regarded as a central park, and the flow characteristic indicators of its internal composition structure are identified by analyzing the total resource flow data of the central park. Flow characteristic indicators refer to key indicators used to determine the resource flow relationship between parks. These indicators help to quantify and evaluate the resource flow within the park and between the park and external parks, thereby providing decision support for energy regulation. Exemplarily, it usually includes the following aspects: Resource flow vector: refers to the amount of resource flow between parks, such as the flow of energy, materials, costs, etc. Flow frequency: The frequency of resource flow between parks, reflecting the activeness of resource flow. Flow direction: The direction of resource flow, which can be directed, indicating which park the resource flows from to another park. Flow efficiency: The efficiency of resource flow, which may involve the efficiency of resource utilization, energy conversion efficiency, etc. Flow cost: The cost involved in the process of resource flow, including transportation cost, processing cost, etc. Environmental impact: The impact of resource flow on the environment, such as carbon emissions, pollutant emissions, etc. Dependence: The degree of dependence of one park on the resources of another park, reflecting the mutual dependence between parks. Complementarity: The complementarity of resources between parks, that is, whether the resources of one park can supplement the resource needs of another park. Stability: The stability of resource flow reflects the predictability and reliability of resource flow. Response time: The response time of the park to changes in resource demand reflects the park's adaptability to market changes. The association value between the central park and other parks is then determined by the association value model, combining the allocation weights of the park's component structures with the flow characteristic indicators, where the association value model includes one or more of the above-mentioned flow characteristic indicators to quantify the degree of association between the parks. In this way, the associated parks that have a significant impact on the energy consumption of the central park can be identified, so that they can be given priority attention in the energy regulation strategy. This step involves detailed monitoring and analysis of energy and resource flows within the park to identify energy and resource usage patterns, efficiency, and potential optimization points. Next, the allocation weights of each component structure in the central park are obtained, which may be determined based on factors such as the energy consumption, economic contribution, or environmental impact of each structure. Then, the association value between the central park and other parks is calculated using the association value model, which multiplies the allocation weights of each component structure with the flow characteristic indicators and sums them to quantify the interdependence and influence between the parks. Finally, based on the calculated correlation values, several parks with the highest correlation are selected from all other parks as associated parks, which have a significant impact on the energy regulation of the central park.

[0090] In an embodiment of the present invention, the interdependence between parks is identified by quantitative methods, thereby providing more accurate decision support for energy regulation. By identifying flow characteristic indicators and assigning weights, the contribution of each component structure to the energy consumption and emissions of the park can be more accurately evaluated, thereby providing a basis for optimizing resource allocation and energy use. In addition, the associated parks screened out by the associated value model help to determine the parks that should be focused on and coordinated in energy regulation, thereby improving the pertinence and effectiveness of energy regulation.

[0091] Optionally, the selecting, according to the association value, from other parks to obtain a plurality of the associated parks includes:

[0092] Determining whether the correlation value is greater than or equal to a preset correlation value;

[0093] If yes, the park corresponding to the associated value is used as the associated park of the central park;

[0094] If not, the park corresponding to the associated value is excluded from the associated parks of the central park.

[0095] Specifically, it is necessary to first evaluate the correlation value between each park and the central park. This evaluation is based on the data previously calculated by the correlation value model, in which the sum of the product of the allocation weight of each component structure and the flow characteristic index is used to quantify the interdependence between parks. The core of the evaluation is to set a preset correlation value as a criterion for judging the closeness of the relationship between parks. If the correlation value of a park reaches or exceeds this preset value, then the park is considered to be an associated park with a significant impact on the central park and needs to be included in the scope of energy regulation. Conversely, if the correlation value is lower than the preset value, the park is considered to have a small impact on the central park and can be excluded from the list of associated parks. This screening process ensures that only those parks that really have a significant impact on the energy consumption of the central park are included in the formulation of the regulation strategy through a quantitative method, thereby improving the pertinence and effectiveness of energy regulation.

[0096] In the embodiment of the present invention, by setting a clear threshold to screen the associated parks, it can be ensured that only those truly important parks are taken into consideration, thereby avoiding waste of resources and unnecessary regulatory measures. At the same time, this method also improves the flexibility and adaptability of the system.

[0097] Optionally, combined Figure 3 As shown, the association network centered on the central park is formed based on all the associated parks and the central park, including:

[0098] Determining the centrality between the associated parks and the central park according to the associated values ​​of all the associated parks of the central park;

[0099] Determining the relative position of each of the associated parks and the central park according to the centrality;

[0100] According to the relative position corresponding to each of the associated parks, the associated network centered on the central park is generated.

[0101] Specifically, in the process of constructing an association network with the central park as the core, it is first necessary to determine the centrality between these parks and the central park based on the association values ​​of all the associated parks of the central park. Centrality is an indicator of the importance of a park in the network, which reflects the core role of the park in energy flow and resource allocation. By calculating the association value, the direct and indirect connections between each associated park and the central park can be quantified, which may include energy supply, resource sharing or infrastructure interconnection. Next, the relative position of each associated park to the central park is determined based on the calculated centrality. This step involves further analysis of the centrality, and network analysis tools may be used to identify the hierarchical structure and mutual relationship of the parks in the network. The determination of relative positions helps to understand the interdependence and potential synergy between parks. Finally, based on the relative position corresponding to each associated park, an association network with the central park as the center is generated. This network includes not only direct connections between parks, but also indirect connections through third-party parks. Visualization tools, such as network diagrams, can intuitively display the connection relationship and network structure between parks.

[0102] In a preferred embodiment of the present invention, for a park , its degree centrality It can be expressed as the number of directly connected parks. The formula is as follows:

[0103] = ;

[0104] in, Indicates the park With other parks The number of connecting edges between .

[0105] Taking the association value into consideration, weighted degree centrality can be used to reflect the influence of each park:

[0106] = ;

[0107] in, It is a park With other parks The association value between them, i and j are used to represent the index variables of different parks in the park network. Specifically: i is used to represent the park currently considered, that is, the central park. In the calculation of degree centrality, is the campus whose connectivity is being analyzed. j is used to traverse the All other directly connected parks. In the degree centrality calculation formula: Indicates that the park All other parks The connecting edges between Add the quantities to get the park The total number of connected edges. In the calculation of weighted degree centrality, Indicates that the park All other parks The correlation value between Add together to get the park In short, i is the campus currently being analyzed, and j is the index of the other campuses connected to i, which is used for the summation operation in the formula.

[0108] Based on the calculated centrality, the relative position of each associated park to the central park can be further determined. The relative position can be expressed in the following ways:

[0109] RelativePosition( )= ;

[0110] in, Represents each of the associated parks, The index variable used to traverse all associated parks, m is the number of all associated parks, that is, The value range is from 1 to m. This formula represents the relative importance of each associated campus in the central campus network.

[0111] In the embodiments of the present invention, by determining the centrality and relative position through quantitative methods, the role and influence of each park in the energy network can be more accurately evaluated. This helps to optimize energy distribution, improve the efficiency and reliability of energy use, and thus provide data support for the formulation of energy regulation strategies.

[0112] Optionally, obtaining the future total energy consumption data of the associated network in a preset future time period according to the energy consumption data of each of the parks in the associated network includes:

[0113] Integrate the energy consumption data of each of the parks in the associated network to determine the total energy consumption data of the associated network;

[0114] The total energy consumption data is input into an energy consumption prediction model, and the future total energy consumption data of the associated network in the preset future time period is determined through the output of the energy consumption prediction model.

[0115] Specifically, in the process of predicting the future total energy consumption data of the associated network, the energy consumption data of each park in the associated network needs to be integrated first. This step involves collecting and organizing the energy consumption records of each park in a specific time period, including the use of different forms of energy such as electricity, gas, and water, to determine the total energy consumption data of the associated network. The purpose of integrating the data is to obtain a comprehensive view of energy consumption, which helps to identify patterns and trends in energy consumption. Next, the collected total energy consumption data is input into the energy consumption prediction model. The energy consumption prediction model is a statistical or machine learning model that can predict future energy consumption based on historical data and possible external factors (such as weather conditions, market changes, etc.). The output of the model will provide the predicted total energy consumption data of the associated network in a preset future time period. The key to this step is to select or develop a prediction model that can accurately capture energy consumption patterns and trends. According to the search results, there are a variety of models that can be used for energy consumption prediction, including ARIMA model, random forest, neural network, etc., among which the random forest model shows higher prediction accuracy in some cases.

[0116] In the embodiments of the present invention, by predicting energy consumption, energy managers are helped to understand future energy demand in advance, so as to more effectively plan and schedule energy supply, optimize energy distribution, reduce waste, and improve energy utilization efficiency.

[0117] Optionally, combined Figure 4 As shown, determining the weight value of each associated park in the associated network through the associated network includes:

[0118] According to the total resource flow data of each park in the associated network, obtaining the future total resource flow data of each park in the preset future time period;

[0119] Determining resource change data of the park according to the future total resource flow data;

[0120] According to the resource change data of the park, adjusting the centrality of the park with the central park in the associated network;

[0121] The weight value of the park in the associated network is determined according to the adjusted centrality.

[0122] Specifically, in the process of determining the weight value of each associated park in the associated network, it is first necessary to predict the future total resource flow data of these parks in a preset future time period based on the total resource flow data of each park in the associated network. This step involves the analysis of historical resource flow data, and it may be necessary to use time series analysis, machine learning or statistical models to predict future resource flow trends. Next, based on the predicted future total resource flow data, the resource change data of each park is determined. Exemplarily, first, the historical resource flow data of each park is collected, which may include data such as energy consumption, material flow, cost flow, etc., and then time series analysis (such as ARIMA model), machine learning model (such as random forest, neural network) or statistical model is used to analyze these data to identify the trend and pattern of resource flow. Based on historical data and trend analysis, a prediction model is established to predict the total resource flow data of each park in the future preset time period. By comparing the predicted future resource flow data with the current resource flow data, the resource change data is calculated, thereby reflecting the increase or decrease of the resource flow of each park. Resource change data reflects the increase or decrease in resource flows in the park, which is crucial to understanding the changes in energy demand and supply of the park in the future time period. Then, based on the resource change data, the centrality of each park in the associated network with the central park is adjusted. Exemplarily, centrality is used to measure the importance of the park in the network. Before considering the resource change data, the centrality of each park may be based on the number and quality of its connections with other parks in the network. Taking the resource change data into account, the centrality of each park is adjusted. For example, if a park has a significant increase in resource outflow, it may be necessary to increase its centrality in the network to reflect its increased influence. Centrality can be adjusted by using network analysis tools (such as the NetworkX library), including recalculating indicators such as degree centrality, closeness centrality, or betweenness centrality for each park in the network. Centrality is an indicator of the importance of a park in the network, which reflects the central role of the park in energy flow and resource allocation. By adjusting the centrality, the status and influence of the park in the network can be more accurately reflected. Finally, the weight value of each park in the associated network is determined based on the adjusted centrality. Exemplarily, a multi-criteria decision analysis (MCDM) method, such as the analytic hierarchy process (AHP) or the topology of technology evaluation (TOPSIS), is used to determine the weight value of each park; these methods allow for consideration of multiple factors, such as resource consumption, environmental impact, economic contribution, etc. A weight value is assigned to each park based on the centrality of each park and the results of the multi-criteria decision analysis. The weight value reflects the relative importance of the park in the associated network. The quantification of the weight value may involve the weighted summation or other mathematical operations of the resource change data and centrality of each park to obtain a comprehensive score, which is the weight value of the park.The weight value is a quantitative representation of the importance of the park, which can be used to guide the formulation of energy regulation strategies. The determination of the weight value may involve multi-criteria decision analysis or similar methods to evaluate and quantify the influence of each park.

[0123] In the embodiments of the present invention, by predicting the future resource flow and adjusting the centrality, the role and influence of each park in the energy network can be more accurately evaluated. This helps to optimize energy distribution, improve the efficiency and reliability of energy use, and promote collaboration between parks.

[0124] Optionally, combined Figure 5 As shown, the future energy consumption data of the central park in the preset future time period is obtained by combining the future total energy consumption data and the weight value of the associated park, including:

[0125] Determine the prediction sub-data of each associated park in the preset future time period according to multiplying the future total energy consumption data of the associated network by the weight value of the associated park;

[0126] Obtain the predicted sub-data of the associated park in other associated networks;

[0127] Averaging the predicted sub-data of the associated park in all the associated networks to obtain the energy consumption data of the associated park in the preset future time period;

[0128] The future energy consumption data of the central park in the preset future time period is determined based on the energy consumption data of each of the associated parks and the future total energy consumption data.

[0129] Specifically, in the process of combining the future total energy consumption data and the weight value of the associated park to determine the energy consumption data of the central park in the future time period, it is first necessary to multiply the future total energy consumption data of the associated network by the weight value of the associated park to determine the predicted sub-data of each associated park in the preset future time period. This step is actually to weight the energy consumption of each associated park to reflect its relative importance in the network. Next, it is necessary to obtain the predicted sub-data of these associated parks in other associated networks, which may involve cross-network data integration and analysis. Then, the predicted sub-data of these associated parks in all associated networks are averaged to obtain the energy consumption data of each associated park in the preset future time period. This step helps to smooth out the outliers in individual networks and provide more robust prediction results. Finally, the energy consumption data of each associated park and the future total energy consumption data are combined to determine the energy consumption data of the central park in the preset future time period. This step integrates the information of all associated parks and provides a comprehensive energy consumption forecast for the central park. In a preferred embodiment of the present invention, a machine learning model, such as a random forest or a neural network, can be used to determine an energy consumption growth factor based on energy consumption data, and based on the energy consumption growth factor and the energy consumption data of the associated park, the energy consumption data in a preset future time period can be determined.

[0130] For example, suppose we have a central park A, which is connected to three associated parks B, C and D. Through analysis, we obtain the total energy consumption forecast data of these parks in the future time period, and determine the weight value of each associated park relative to the central park A. These weight values ​​reflect their relative importance in energy consumption. For example, the weight values ​​may be: B=0.5, C=0.3, D=0.2. Then obtain the predicted sub-data across networks: Assuming that parks B, C and D are not only connected to park A, but also have energy consumption associations with other parks, collect the predicted sub-data of these parks in other associated networks. For example, if park B is also connected to parks E and F, it is necessary to obtain the predicted data of B in these networks. Then average the predicted sub-data: For each associated park, average its predicted sub-data in all associated networks. For example, if the predicted consumption of campus B in the network with campus A is 100 units, in the network with campus E is 120 units, and in the network with campus F is 90 units, we calculate the average: (100 + 120 + 90) / 3 = 100 units. Similarly, the average predicted consumption data of campuses C and D are obtained. Finally, the energy consumption data of the central campus A is determined by combining the average predicted consumption data of each associated campus and their weight values. For example, if the average predicted consumption data of associated campuses B, C, and D are 100 units, 150 units, and 200 units, respectively, the predicted consumption of the central campus A is calculated according to the weight values: predicted consumption of the central campus A = (100×0.5)+(150×0.3)+(200×0.2)=50+45+40=135 units. Through this process, the energy consumption forecast data of the central campus A in the future time period can be obtained, which integrates the information of all associated campuses and takes into account their relative importance in the network.

[0131] In the embodiment of the present invention, by comprehensively considering the weight values ​​of the associated parks and the future energy consumption data, a more comprehensive and accurate energy consumption forecast is provided. This forecast not only considers the direct energy consumption of the central park, but also considers the mutual influence between it and the associated parks, thereby providing a more accurate energy consumption forecast.

[0132] Optionally, generating an energy regulation strategy according to the future energy consumption data and performing energy regulation according to the energy regulation strategy includes:

[0133] Determining the expected control target of the central park according to the future energy consumption data;

[0134] Generate energy control strategies through reinforcement learning combined with expected control targets;

[0135] The energy consumption data of the central park in the preset current time period is taken as the state, the initial energy regulation strategy is taken as the action, and the error value between the energy consumption data after the initial energy regulation strategy is applied and the expected regulation target is taken as the reward;

[0136] When the error value is less than a preset reward threshold, the energy regulation strategy at the corresponding moment is determined to be the final energy regulation strategy, and energy regulation is performed according to the final energy regulation strategy.

[0137] Specifically, in the process of generating energy control strategies and performing energy control, it is first necessary to determine the expected control target of the central park based on future energy consumption data. This step is the basis for the formulation of the entire control strategy, which involves the prediction and analysis of the energy consumption of the central park, as well as the setting of energy consumption targets. Then, the energy control strategy is generated by learning in combination with the expected control target through the reinforcement learning method. In this process, the energy consumption data of the central park in the preset current time period is regarded as the state, the initial energy control strategy is regarded as the action, and the error value between the energy consumption data after the initial energy control strategy and the expected control target is used as the reward signal. The key to this step is to continuously adjust and optimize the strategy through the reinforcement learning algorithm to minimize the error value, thereby achieving effective control of energy consumption. Finally, when the error value is less than the preset reward threshold, the initial energy control strategy is determined to be the final energy control strategy, and energy control is performed according to this strategy. This judgment criterion ensures that the strategy will only be actually adopted when it can achieve the expected control effect. The advantage of this method is that it can dynamically adapt to changes in energy consumption, and continuously optimize the control strategy through real-time feedback and learning, thereby achieving more accurate and effective energy management.

[0138] In an embodiment of the present invention, by learning historical data and predicting future trends, real-time monitoring and dynamic adjustment of energy consumption can be achieved, thereby improving the flexibility and response speed of energy use. At the same time, a more scientific and accurate control strategy can be provided through a reinforcement learning algorithm.

[0139] Optionally, the multi-park energy control method further includes:

[0140] The carbon emissions of the park are predicted through a distributed system, wherein the distributed system includes a plurality of nodes, each of which processes a park and the associated network corresponding to the park.

[0141] Specifically, under the distributed system architecture, each node independently processes the data of a park and its corresponding associated network, thereby processing the data of multiple parks in parallel, greatly reducing the overall computing time. For example, if a large industrial park is composed of several small parks, each node can specifically process the data of a small park while considering the mutual influence between the park and other parks. This distributed processing method not only improves the computing speed, but also allows the system to flexibly add or reduce nodes to adapt to associated networks of different sizes, thereby achieving scalability. In addition, distributed systems can also reduce the risk of single point failures and improve the stability and reliability of the overall system.

[0142] In an embodiment of the present invention, a large-scale complex data set is processed through a distributed system while maintaining high computational efficiency and accuracy.

[0143] In general, the multi-park energy control method of the present invention provides comprehensive basic information for energy control by collecting energy consumption data and resource flow data of each component structure in each park, including energy consumption and emission factors; by analyzing these data, the energy use and environmental impact of each park can be understood in detail, providing data support for subsequent energy management and optimization. And each park is taken as a central park, and the relationship with other parks is determined according to the resource flow data, and an associated network is constructed. This network includes not only the direct energy flow between parks, but also the indirect energy dependency relationship; the construction of the network helps to identify the mutual dependence and influence between parks, so as to carry out energy control at a more macro level. At the same time, using the energy consumption data in the associated network, combined with the energy consumption prediction model, the total energy consumption in the future time period is predicted, which helps energy managers understand the future energy demand in advance; through prediction, energy supply can be more effectively planned and scheduled, energy distribution can be optimized, waste can be reduced, and energy utilization efficiency can be improved. The centrality of the park is adjusted by the resource change data in the associated network, and the weight value of each associated park is determined to reflect its relative importance in energy consumption; The determination of the weight value helps to allocate energy consumption more accurately in the energy control strategy and ensure the reasonable distribution of energy among the parks. In addition, the energy consumption data of the central park in the future time period is calculated by combining the future total energy consumption data and the weight value of the associated park, and a refined energy control strategy is generated to guide the energy use and allocation in actual operation and improve the accuracy and efficiency of energy control. Through reinforcement learning, an energy control strategy is generated according to the expected control target, and the strategy is dynamically adjusted to minimize the error between the actual energy consumption and the target; the advantage of this method is that it can dynamically adapt to changes in energy consumption, and continuously optimize the control strategy through real-time feedback and learning to achieve more accurate and effective energy management. In addition, the prediction of carbon emissions in the park through a distributed system also improves computing efficiency and accuracy, while reducing the risk of single point failure; this distributed processing method not only improves the computing speed, but also allows the system to flexibly add or reduce nodes to adapt to associated networks of different sizes, thereby achieving scalability. The present invention can reduce energy waste and improve energy efficiency through refined energy regulation; more accurate energy management helps reduce unnecessary energy consumption, thereby reducing greenhouse gas emissions and other environmental impacts. By predicting and optimizing energy distribution, energy supply during critical periods can be ensured and system reliability can be enhanced; optimizing energy consumption can reduce energy costs and improve the economic efficiency of the park. The application of reinforcement learning and distributed systems enables the energy regulation system to adapt to changes and flexibly respond to different energy demands and market conditions.The present invention realizes fine-grained regulation of energy consumption by comprehensively considering energy consumption data within and between parks and predicting future energy consumption, improves the accuracy and efficiency of energy regulation, and provides strong support for the sustainable development of the park.

[0144] Combination Figure 6 As shown, the present invention also provides a multi-park energy control system, including:

[0145] A data acquisition unit, used to acquire energy consumption data of each park in a preset current time period and resource flow data of each component structure in the park; wherein the energy consumption data includes energy consumption sub-data of each component structure in the park and emission factors corresponding to the component structures;

[0146] A resource analysis unit, configured to obtain total resource flow data of the park according to the resource flow data of the component structure of each park;

[0147] An association analysis unit, used to take each of the parks as a central park, and determine other parks related to the central park according to the total resource flow data of each central park, and take the other parks as associated parks;

[0148] A cluster unit, used for forming an associated network centered on the central park according to all the associated parks and the central park;

[0149] A first data processing unit, configured to obtain future total energy consumption data of the associated network in a preset future time period according to the energy consumption data of each of the parks in the associated network;

[0150] A second data processing unit is used to determine the weight value of each associated park in the associated network through the associated network;

[0151] An energy consumption analysis unit, configured to combine the future total energy consumption data and the weight value of the associated park to obtain the future energy consumption data of the central park in the preset future time period;

[0152] The control unit is used to generate an energy control strategy according to the future energy consumption data and perform energy control according to the energy control strategy.

[0153] The carbon emission prediction system of multiple parks of the present invention can obtain energy consumption data and resource flow data of each park within a preset time period, so as to have a detailed understanding of the energy use of each park. These data include energy consumption sub-data of each component structure in the park and the corresponding emission factor, thereby providing a basis for subsequent analysis. Then, by analyzing the resource flow data of each park, the total resource flow data of the park is obtained, which helps to understand the flow of resources within the park and provides a macro perspective for further analysis. Then, each park is taken as a central park, and the relationship with other parks is determined according to its total resource flow data, and an association network is constructed. This network includes not only the central park, but also the associated parks related to it. In this way, the mutual dependence and influence between parks can be identified, so as to perform energy regulation at a more macro level. By constructing an association network, not only the energy consumption and resource flow of a single park are focused on, but the park is regarded as a node in an overall network, and the energy flow and consumption of each node are interrelated with other nodes. On the basis of the association network, by determining the weight value of each associated park, energy consumption can be allocated more accurately to ensure the reasonable distribution of energy between parks. Finally, by combining the future total energy consumption data and the weight values ​​of the associated parks, the energy consumption data of the central park in the future time period can be obtained. This enables the control strategy to more accurately guide the use and allocation of energy, thereby improving the accuracy and efficiency of energy control. The present invention can effectively avoid the problem of a single park being affected by other parks and reducing the control accuracy through a multi-park energy control method. By comprehensively considering the energy consumption data within the park and between parks, and predicting future energy consumption, refined control of energy consumption is achieved.

[0154] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided by the present invention can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0155] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0156] The foregoing is merely a specific embodiment of the present invention, which enables those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be 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 the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A multi-park energy control method, characterized in that: include: Obtaining energy consumption data of each park in a preset current time period and resource flow data of each component structure in the park; wherein the energy consumption data includes energy consumption sub-data of each component structure in the park and emission factors corresponding to the component structure; Obtaining total resource flow data of the park according to the resource flow data of the component structure of each park; Taking each of the parks as a central park, and determining other parks related to the central park based on the total resource flow data of each central park, and taking the other parks as associated parks; specifically including: taking each of the parks as a central park, and determining the flow characteristic index between the component structure of the central park and the component structures of other parks based on the total resource flow data of the central park; Obtaining the allocation weight corresponding to each of the component structures of the central park; According to the association value model, the sum of the products of the allocation weight of each component structure and the flow characteristic index of the component structure is used as the association value between the central park and the other parks; According to the association value, a plurality of associated parks are obtained by screening from the other parks; Forming an associated network centered on the central park according to all the associated parks and the central park; According to the energy consumption data of each of the parks in the associated network, obtaining the future total energy consumption data of the associated network in a preset future time period; Determine the weight value of each associated park in the associated network through the associated network; Combining the future total energy consumption data with the weight value of the associated park, obtaining the future energy consumption data of the central park in the preset future time period; An energy control strategy is generated according to the future energy consumption data, and energy control is performed according to the energy control strategy.

2. The multi-park energy control method according to claim 1, characterized in that: The step of selecting a plurality of associated parks from other parks according to the associated value includes: Determining whether the correlation value is greater than or equal to a preset correlation value; If yes, the park corresponding to the associated value is used as the associated park of the central park; If not, the park corresponding to the associated value is excluded from the associated parks of the central park.

3. The multi-park energy control method according to claim 1, characterized in that: The forming of an associated network centered on the central park based on all the associated parks and the central park includes: Determining the centrality between the associated parks and the central park according to the associated values ​​of all the associated parks of the central park; Determining the relative position of each of the associated parks and the central park according to the centrality; According to the relative position corresponding to each of the associated parks, the associated network centered on the central park is generated.

4. The multi-park energy control method according to claim 3 is characterized in that: The obtaining, according to the energy consumption data of each of the parks in the associated network, future total energy consumption data of the associated network in a preset future time period comprises: Integrate the energy consumption data of each of the parks in the associated network to determine the total energy consumption data of the associated network; The total energy consumption data is input into an energy consumption prediction model, and the future total energy consumption data of the associated network in the preset future time period is determined through the output of the energy consumption prediction model.

5. The multi-park energy control method according to claim 4 is characterized in that: Determining the weight value of each associated park in the associated network through the associated network includes: According to the total resource flow data of each park in the associated network, obtaining the future total resource flow data of each park in the preset future time period; Determining resource change data of the park according to the future total resource flow data; According to the resource change data of the park, adjusting the centrality of the park with the central park in the associated network; The weight value of the park in the associated network is determined according to the adjusted centrality.

6. The multi-park energy control method according to claim 1, characterized in that: The combining of the future total energy consumption data and the weight value of the associated park to obtain the future energy consumption data of the central park in the preset future time period includes: Determine the prediction sub-data of each associated park in the preset future time period according to multiplying the future total energy consumption data of the associated network by the weight value of the associated park; Obtain the predicted sub-data of the associated park in other associated networks; Averaging the predicted sub-data of the associated park in all the associated networks to obtain the energy consumption data of the associated park in the preset future time period; The future energy consumption data of the central park in the preset future time period is determined based on the energy consumption data of each of the associated parks and the future total energy consumption data.

7. The multi-park energy control method according to claim 1, characterized in that: The step of generating an energy control strategy according to the future energy consumption data and performing energy control according to the energy control strategy includes: Determining the expected control target of the central park according to the future energy consumption data; Generate energy control strategies through reinforcement learning combined with expected control targets; The energy consumption data of the central park in the preset current time period is taken as the state, the initial energy regulation strategy is taken as the action, and the error value between the energy consumption data after the initial energy regulation strategy is applied and the expected regulation target is taken as the reward; When the error value is less than a preset reward threshold, the energy regulation strategy at the corresponding moment is determined to be the final energy regulation strategy, and energy regulation is performed according to the final energy regulation strategy.

8. The multi-park energy control method according to claim 1, characterized in that: The multi-park energy control method further includes: The carbon emissions of the park are predicted through a distributed system, wherein the distributed system includes a plurality of nodes, each of which processes a park and the associated network corresponding to the park.

9. A multi-park energy control system, characterized in that: include: A data acquisition unit, used to acquire energy consumption data of each park in a preset current time period and resource flow data of each component structure in the park; wherein the energy consumption data includes energy consumption sub-data of each component structure in the park and emission factors corresponding to the component structures; A resource analysis unit, configured to obtain total resource flow data of the park according to the resource flow data of the component structure of each park; An association analysis unit is used to take each of the parks as a central park, and determine other parks related to the central park according to the total resource flow data of each central park, and take other parks as associated parks; specifically including: taking each of the parks as a central park, and determining the flow characteristic index between the component structure of the central park and the component structures of other parks according to the total resource flow data of the central park; Obtaining the allocation weight corresponding to each of the component structures of the central park; According to the association value model, the sum of the products of the allocation weight of each component structure and the flow characteristic index of the component structure is used as the association value between the central park and the other parks; According to the association value, a plurality of associated parks are obtained by screening from the other parks; A cluster unit, used for forming an associated network centered on the central park according to all the associated parks and the central park; A first data processing unit, configured to obtain future total energy consumption data of the associated network in a preset future time period according to the energy consumption data of each of the parks in the associated network; A second data processing unit is used to determine the weight value of each associated park in the associated network through the associated network; An energy consumption analysis unit, configured to combine the future total energy consumption data and the weight value of the associated park to obtain the future energy consumption data of the central park in the preset future time period; The control unit is used to generate an energy control strategy according to the future energy consumption data and perform energy control according to the energy control strategy.

Images