A method, apparatus and system for configuring integrated energy.

Abstract

This invention discloses a method, apparatus, and system for configuring integrated energy. The method includes: acquiring data of a system to be configured; analyzing the data of the system to be configured based on the data similarity between the data of the system to be configured and various historical configuration schemes in a historical configuration scheme library, adjusting the historical configuration schemes, and providing a target configuration scheme; obtaining the configuration structure of integrated energy according to the target configuration scheme, and acquiring changes in the configuration structure of integrated energy in real time, adjusting the integrated energy operation model, and obtaining a target operation model; obtaining real-time operation data of integrated energy based on the target operation model; extracting features from the real-time operation data by combining a pre-constructed evaluation index library, obtaining target evaluation features; fusing the evaluation weights corresponding to the target evaluation features, and providing an operation evaluation result of integrated energy; adjusting the target operation model according to the operation evaluation result, completing the configuration of integrated energy, and improving the stability of integrated energy.

Description

Technical Field

[0001] This invention belongs to the technical field of integrated energy systems, specifically relating to a method, apparatus, and system for configuring integrated energy. Background Technology

[0002] An integrated energy system refers to an energy system that utilizes advanced physical information technology and innovative management models within a specific area to integrate multiple energy sources, achieving coordinated planning and optimized operation among various energy subsystems. This system aims to meet diverse energy demands while effectively improving energy efficiency and promoting sustainable energy development. The management of an integrated energy system involves regulating energy flow through information flow, optimizing the control of energy equipment within the system, reducing overall operating costs, improving energy efficiency, and ensuring safe and reliable operation. This is a crucial guarantee for the stable and efficient operation of an integrated energy system and the core of its flexible operation. Simultaneously, accurate assessment of integrated energy resources is a prerequisite for improving integrated energy utilization efficiency.

[0003] Patent application CN112182887A discloses a simulation method for integrated energy system planning and optimization, comprising: step S1, establishing an efficiency analysis model based on EQC according to energy characteristics; step S2, establishing a four-layer energy center model based on multi-energy coupling analysis of the integrated energy system; step S3, establishing an improved convex programming model based on non-convex factors in the integrated energy system; and step S4, establishing an integrated energy system planning and optimization model based on economy and efficiency. This multi-objective programming optimization, with economy and efficiency as its goals, achieves qualitative matching in the process of integrated energy utilization, reaching a comprehensive performance balance in all aspects.

[0004] Currently, integrated energy planning is based on the energy characteristics of the integrated energy system. However, after operating in different environments for a period of time, integrated energy systems will undergo certain changes. If a fixed planning scheme is adopted, the utilization efficiency of integrated energy cannot be guaranteed, and there are certain risks and uncertainties. How to formulate and adjust integrated energy planning schemes and improve the stability of integrated energy is a problem that needs to be solved. Summary of the Invention

[0005] To address the shortcomings of the existing technologies, this invention provides a method, apparatus, and system for configuring integrated energy. The method includes: acquiring system data to be configured; analyzing the system data based on the similarity between the system data to be configured and various historical configuration schemes in a historical configuration scheme library, adjusting the historical configuration schemes, and providing a target configuration scheme; obtaining the configuration structure of integrated energy according to the target configuration scheme, acquiring changes in the configuration structure of integrated energy in real time, adjusting the integrated energy operation model, and obtaining a target operation model; obtaining real-time operation data of integrated energy based on the target operation model; extracting features from the real-time operation data using a pre-constructed evaluation index library to obtain target evaluation features; fusing the evaluation weights corresponding to the target evaluation features to provide an operation evaluation result of integrated energy; and adjusting the target operation model according to the operation evaluation result to complete the configuration of integrated energy and improve the stability of integrated energy. By analyzing the data of the system to be configured, a preliminary plan for integrated energy is completed, a target configuration scheme is obtained, and the operation model of integrated energy is adjusted based on the target configuration scheme. The target operation model is given, and adjustments are made in real time based on the changes in the operation and configuration of integrated energy. To a certain extent, the utilization efficiency of integrated energy is guaranteed, the risks and uncertainties of integrated energy are reduced, and the stability of integrated energy is improved.

[0006] In a first aspect, the present invention provides a method for configuring integrated energy, specifically including the following steps:

[0007] Obtain the system data to be configured;

[0008] Based on the data similarity between the system data to be configured and the historical configuration schemes in the historical configuration scheme library, the system data to be configured is analyzed, the historical configuration schemes are adjusted, and the target configuration scheme is given.

[0009] Based on the target configuration scheme, the configuration structure of the integrated energy is obtained, and the changes in the configuration structure of the integrated energy are acquired in real time. The integrated energy operation model is then adjusted to obtain the target operation model.

[0010] Based on the target operation model, real-time operation data of integrated energy is obtained;

[0011] By combining a pre-built evaluation index library, feature extraction is performed on real-time operational data to obtain target evaluation features;

[0012] By integrating the assessment weights corresponding to the assessment characteristics of the target, a comprehensive energy operation assessment result is given;

[0013] Based on the operational assessment results, the target operational model is adjusted to complete the configuration of integrated energy resources.

[0014] Furthermore, the data similarity between the system data to be configured and the data of each historical configuration scheme in the historical configuration scheme library is obtained through the following steps:

[0015] Based on the type of system data to be configured, a similarity index analysis strategy is used to analyze the local similarity with the historical data corresponding to each historical configuration scheme.

[0016] By combining the similarity weights corresponding to different types, the local similarity of each historical configuration scheme is fused to obtain the data similarity of each historical configuration scheme.

[0017] Furthermore, the types of data to be configured include discrete and continuous types, and the local similarity includes discrete similarity and continuous similarity;

[0018] Based on the type of system data to be configured, a similarity index analysis strategy is used to analyze the local similarity with historical data corresponding to each historical configuration scheme, specifically including:

[0019] The first similarity index is used to analyze the discrete similarity between discrete system data to be configured and corresponding discrete data in historical configuration schemes;

[0020] The second similarity index is used to analyze the continuous similarity between the continuous data of the system to be configured and the corresponding continuous data in the historical configuration scheme.

[0021] Furthermore, similarity weights include discrete similarity weights and continuous similarity weights;

[0022] By combining the similarity weights corresponding to different types, the local similarity scores of each historical configuration scheme are fused to obtain the data similarity scores for each historical configuration scheme, specifically including:

[0023] The discrete similarity and continuous similarity are normalized respectively to obtain the standard discrete similarity and standard continuous similarity;

[0024] By combining discrete similarity weights and continuous similarity weights, the standard discrete similarity and standard continuous similarity are weighted and summed to obtain the data similarity.

[0025] Furthermore, based on the data similarity between the system data to be configured and various historical configuration schemes in the historical configuration scheme library, the system data to be configured is analyzed, and the historical configuration schemes are adjusted to provide a target configuration scheme, specifically including:

[0026] Based on data similarity, the historical configuration schemes in the historical configuration scheme library are sorted, and pending historical schemes are given based on the sorting results.

[0027] Based on the pre-obtained configuration mapping function, the system data to be configured is mapped to obtain the target configuration scheme. The pre-obtained configuration mapping function is obtained by analyzing and correcting the historical scheme to be configured.

[0028] Furthermore, based on the target configuration scheme, the integrated energy configuration structure is obtained, specifically including:

[0029] From the target configuration scheme, obtain the various energy devices and loads in the integrated energy system, and determine the nodes of the integrated directed graph;

[0030] Analyze the operating status of each energy device and load in the integrated energy system to determine the attributes of each node in the integrated directed graph;

[0031] By combining the connection relationships and connection status of various energy devices and loads in the integrated energy system, the edges and edge attributes corresponding to each node in the integrated directed graph are determined.

[0032] By integrating the nodes, attributes, edges, and attributes of the integrated directed graph, the integrated directed graph is constructed, resulting in the configuration structure of the integrated energy.

[0033] Furthermore, by acquiring real-time changes in the configuration structure of the integrated energy system and adjusting the integrated energy operation model, a target operation model is obtained, which specifically includes:

[0034] Based on the changes in the configuration structure of integrated energy, the changes in the configuration structure are quantified, and the state changes of integrated energy are given.

[0035] Based on the pre-acquired state change threshold, the state change amount is judged, and a dynamic exploration factor is given. The dynamic exploration factor is used to influence the action selection of the comprehensive energy at the next moment.

[0036] Based on dynamic exploration factors and combined with the system loss function, the integrated energy operation model is trained until convergence, thus obtaining the target operation model.

[0037] Furthermore, based on the changes in the configuration structure of integrated energy, the changes in the configuration structure are quantified, and the state changes of integrated energy are given, specifically including:

[0038] Obtain the composite directed graph at two adjacent time points, analyze the changes in the number of nodes and edges in the composite directed graph at the corresponding two time points, and obtain the change results;

[0039] Based on the changes, we analyze the changes of each node and edge in the comprehensive directed graph at two adjacent time points to obtain the changes in nodes and edges.

[0040] By combining the changes in node values ​​and edge values, we obtain the state change values.

[0041] Furthermore, the construction of the integrated energy operation model is determined through the following steps:

[0042] Based on the initial deep learning model, combined with the integrated directed graph and the system action space, the expected cumulative reward values ​​corresponding to different states and actions are obtained. The system action space includes various actions performed on each energy device and load in the integrated energy system.

[0043] Based on the expected cumulative reward value corresponding to different states and actions, the mean and variance of the expected cumulative reward value are given, and the state distribution of the expected cumulative reward value corresponding to different states and actions is obtained.

[0044] Based on the state distribution of the expected cumulative reward value, the model parameters of the integrated energy operation model are determined, and the construction of the integrated energy operation model is completed.

[0045] Furthermore, by combining a pre-built evaluation index library, feature extraction is performed on real-time operational data to obtain target evaluation features, specifically including:

[0046] Based on a pre-built comprehensive energy index library, corresponding initial assessment features are selected from real-time operation data;

[0047] Principal component analysis was used to analyze the initial evaluation features and identify the core features.

[0048] The target evaluation features are determined based on the similarity between each initial evaluation feature and the core feature.

[0049] Furthermore, principal component analysis is used to analyze the initial evaluation features and identify the core features, which specifically include:

[0050] Analyze the degree of linear correlation between the initial evaluation features and construct the covariance matrix of the initial evaluation features;

[0051] The covariance matrix is ​​subjected to eigenvalue decomposition, which yields multiple eigenvalues.

[0052] Based on the variance explanation ratio of each feature value and combined with a preset explanation threshold, the core features are obtained by filtering among multiple feature values.

[0053] Furthermore, the linear correlation between the initial evaluation features is analyzed, and the covariance matrix of the initial evaluation features is constructed, specifically including:

[0054] Based on the elements in each initial evaluation feature and the feature mean of the corresponding initial evaluation feature, the deviation feature of each initial evaluation feature is given.

[0055] Based on the deviation features of any two initial evaluation features, the mean of the product of the two deviation features is given, and the degree of linear correlation between the initial evaluation features is obtained.

[0056] By integrating the degree of linear correlation among the various initial evaluation features, a covariance matrix of the initial evaluation features is constructed.

[0057] Furthermore, based on the variance explanation ratio of each feature value and a preset explanation threshold, core features are obtained by filtering among multiple feature values, specifically including:

[0058] Based on the variance explained by each eigenvalue, the cumulative variance explained by each eigenvalue is given.

[0059] The cumulative variance explained by each feature value is compared with the preset explanation threshold to obtain the comparison results;

[0060] Based on the comparison results, the feature values ​​are filtered, and the feature vectors corresponding to the filtered feature values ​​are given.

[0061] The core features are obtained by projecting the standard data matrix onto a vector matrix composed of the feature vectors corresponding to the filtered feature values.

[0062] Furthermore, the target operation model includes a reward function, which consists of a comprehensive energy generation component, a comprehensive energy storage component, and a comprehensive energy low-carbon operation component.

[0063] Based on the operational assessment results, the target operational model is adjusted to complete the configuration of integrated energy resources, specifically including:

[0064] If the operational evaluation result is unsatisfactory, then analyze the sub-scores of each part of the operational evaluation result;

[0065] Analyze the achievement status of each sub-score, and adjust the coefficients of the corresponding part of the reward function in the target running model according to the achievement status;

[0066] The target running model is trained based on the adjusted reward function until convergence;

[0067] Based on the target operating model after training, run the integrated energy system and complete the configuration of the integrated energy system.

[0068] Secondly, the present invention also provides an integrated energy configuration device, employing an integrated energy configuration method as described in any of the above claims, comprising:

[0069] The sensing module is used to acquire data about the system to be configured.

[0070] The storage module is used to store a library of historical configuration schemes.

[0071] The calculation module analyzes the system data to be configured based on the data similarity between the system data to be configured and the historical configuration schemes in the historical configuration scheme library, adjusts the historical configuration schemes, and provides a target configuration scheme. Based on the target configuration scheme, it obtains the configuration structure of the integrated energy system and acquires real-time changes in the configuration structure, adjusts the integrated energy system operation model, and obtains the target operation model. Based on the target operation model, it obtains real-time operation data of the integrated energy system. Combining a pre-built evaluation index library, it extracts features from the real-time operation data to obtain target evaluation features. It integrates the evaluation weights corresponding to the target evaluation features to provide the operation evaluation results of the integrated energy system. Based on the operation evaluation results, it adjusts the target operation model to complete the configuration of the integrated energy system.

[0072] Thirdly, the present invention also provides an integrated energy configuration system, comprising: a network device, an execution device, and an integrated energy configuration device as described above.

[0073] The present invention provides a method, apparatus, and system for configuring integrated energy, which has at least the following beneficial effects:

[0074] (1) By analyzing the data of the system to be configured, the preliminary planning of the integrated energy is completed, the target configuration scheme is obtained, and the operation model of the integrated energy is adjusted based on the target configuration scheme. The target operation model is given, and adjustments are made in real time based on the operation and configuration changes of the integrated energy. To a certain extent, the utilization efficiency of the integrated energy is guaranteed, the risk and uncertainty of the integrated energy are reduced, and the stability of the integrated energy is improved.

[0075] (2) By integrating the initial deep learning model, system action space and reward function, the state distribution of expected cumulative reward values ​​corresponding to different system actions and different system states is analyzed, so as to model the uncertainty of system state and system action and realize the dynamic adjustment of model parameters.

[0076] (3) By analyzing the data of the configuration system, based on the data similarity, the historical configuration schemes with high similarity are selected from the historical configuration scheme library to determine the target configuration scheme, so as to improve the configuration efficiency of the integrated energy while ensuring the integrated energy configuration effect. Attached Figure Description

[0077] Figure 1 A flowchart illustrating the integrated energy configuration method provided in this embodiment of the invention;

[0078] Figure 2 A flowchart illustrating the target configuration scheme provided in this embodiment of the invention;

[0079] Figure 3 A flowchart for obtaining the target running model provided in an embodiment of the present invention;

[0080] Figure 4 A flowchart for determining target evaluation features provided in an embodiment of the present invention;

[0081] Figure 5 A flowchart for determining evaluation weights provided in an embodiment of the present invention;

[0082] Figure 6 A flowchart for configuring integrated energy sources is provided as an embodiment of the present invention;

[0083] Figure 7 This is a structural block diagram of an integrated energy configuration device provided in an embodiment of the present invention. Detailed Implementation

[0084] To better understand the above technical solutions, a detailed description of the solutions will be provided below in conjunction with the accompanying drawings and specific embodiments. Obviously, the described embodiments are merely some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0085] The terminology used in the embodiments of this invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. The singular forms “a,” “the,” and “the” as used in the embodiments of this invention and the appended claims are also intended to include the plural forms, and “multiple” generally includes at least two unless the context clearly indicates otherwise.

[0086] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that an article or device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such an article or device. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the article or device that includes said element.

[0087] Integrated energy systems couple multiple energy sources such as electricity, gas, and heat, and achieve multi-energy synergy in production, transmission, distribution, utilization, and storage. This can improve the overall energy utilization efficiency and system operation flexibility, and has been widely developed.

[0088] Currently, there is a lack of comprehensive control measures for integrated energy systems, from planning to management. This results in a failure to respond promptly to changes in the integrated energy system, leading to high energy consumption and poor intelligence in its operation. Ensuring comprehensive control of integrated energy systems in complex and dynamic environments is a problem that needs to be addressed.

[0089] Integrated energy planning involves data acquisition and analysis. A storage-computing integrated architecture is adopted, deeply integrating sensing, storage, and computing functions—capable of in-situ data processing (computation and storage fusion)—into small devices at the hardware level, forming integrated intelligent sensing devices that combine sensing, storage, and computing. These deeply integrated intelligent sensing devices are deployed in energy supply networks, energy coupling equipment, energy storage devices, and user terminals to achieve near-source, efficient processing of integrated energy system status data (critical calculations are performed within the storage medium), supporting system planning and control.

[0090] like Figure 1 As shown, this embodiment of the invention provides a method for configuring integrated energy, the specific steps of which are as follows:

[0091] S101: Obtain system data to be configured.

[0092] It's important to understand that the system data to be configured includes data required during the integrated energy planning process. This data includes, but is not limited to, construction area, geographical location, land use planning, photovoltaic resources, wind resources, geothermal resources, current energy supply and consumption data, environmental data (such as solar radiation intensity, wind speed, humidity, and temperature), energy market conditions (such as energy prices), and statistics on cooling, heating, and electricity loads. Cooling, heating, and electricity load statistics can be presented as duration curves for electricity, heat, cooling, and gas loads. It can also include basic infrastructure information for power distribution and gas networks, load demand on typical days in different seasons, and specific parameters for energy storage devices, photovoltaic power plants, substations, etc., with the appropriate data used depending on the specific focus.

[0093] S102: Based on the data similarity between the system data to be configured and the historical configuration schemes in the historical configuration scheme library, analyze the system data to be configured, adjust the historical configuration schemes, and give the target configuration scheme.

[0094] Before analyzing data similarity, it's essential to understand the historical configuration scheme database. Each historical configuration scheme in this database is based on planning data from different regions and under different circumstances, along with the corresponding configuration schemes. Configuration schemes involve analyzing and configuring planning data for the target area. Planning data includes construction area, geographical location, land use planning, photovoltaic resources, wind resources, geothermal resources, current energy supply and consumption data, environmental data (e.g., solar radiation intensity, wind speed, humidity, and temperature), energy market conditions (e.g., energy prices), and statistics on cooling, heating, and electricity loads. Cooling, heating, and electricity load statistics can be presented as duration curves for electricity, heat, cooling, and gas loads. It may also include basic construction information for power distribution and gas networks, load demand on typical days in different seasons, and specific parameters for energy storage devices, photovoltaic power plants, substations, etc., using the appropriate data based on different priorities. By analyzing planning data, the target area's electricity demand (e.g., peak hours, base load), heat demand (e.g., heating and hot water supply), and cooling demand (e.g., air conditioning and refrigeration) are determined, along with available energy resources within the target area, such as renewable energy sources like solar, wind, and geothermal energy, traditional energy sources like natural gas and electricity, and waste heat recovery from industrial processes. Based on the target area's conditions, appropriate energy technologies and equipment are selected, such as solar photovoltaic panels for power generation, wind turbines for wind power generation, ground-source heat pumps for heating and cooling, and energy storage devices for balancing supply and demand. Different energy technologies and equipment are integrated into a unified system, combining multiple energy sources to improve energy supply stability. Energy supply is adjusted according to changes in energy demand, and future energy demand is predicted to optimize energy production and dispatch, forming a complete configuration scheme.

[0095] Different planning data lead to different configuration schemes. The greater the difference in planning data, the lower the reliability of the corresponding configuration scheme. Therefore, the reliability of each historical configuration scheme in the historical configuration scheme library is determined by calculating the data similarity between the current system data to be configured and the planning data in each historical configuration scheme. The higher the data similarity, the closer the data of the system to be configured is to the planning data of the historical configuration scheme, and the higher the reliability of the corresponding configuration scheme. Based on the above analysis, the data similarity between the data of the system to be configured and each historical configuration scheme in the historical configuration scheme library is given. The specific steps are as follows:

[0096] Based on the type of system data to be configured, a similarity index analysis strategy is used to analyze the local similarity with the historical data corresponding to each historical configuration scheme.

[0097] By combining the similarity weights corresponding to different types, the local similarity of each historical configuration scheme is fused to obtain the data similarity of each historical configuration scheme.

[0098] In the embodiments provided by this invention, all data are divided into continuous data and discrete data according to the value taking method of the system data to be configured. That is, the type of data to be configured includes discrete and continuous data, and the local similarity includes discrete similarity and continuous similarity. It is understood that different similarity indicators can more accurately reflect the degree of similarity between data of different types. In other embodiments, the same similarity indicator can be used for the system data to be configured, and there is no limitation on this.

[0099] Furthermore, based on the type of data in the system to be configured, a similarity index analysis strategy is used to analyze the local similarity with the historical data corresponding to each historical configuration scheme, specifically including:

[0100] The first similarity index is used to analyze the discrete similarity between discrete system data to be configured and corresponding discrete data in historical configuration schemes;

[0101] The second similarity index is used to analyze the continuous similarity between the continuous data of the system to be configured and the corresponding continuous data in the historical configuration scheme.

[0102] In the embodiments provided by this invention, the first similarity index is the Jaccard coefficient, and the second similarity index is the Euclidean distance. Other similarity indices can be used in other embodiments, such as Manhattan distance, Hamming distance, Mahalanobis distance, and Pearson correlation coefficient, etc., without limitation. The Jaccard coefficient measures the ratio of the intersection to the union of two sets. A larger Jaccard similarity coefficient indicates greater similarity between the two sets. Euclidean distance is a measure of the straight-line distance between two vectors. A smaller Euclidean distance indicates greater similarity between the two vectors. Manhattan distance is the sum of the absolute distances of two vectors on each coordinate axis. A smaller Manhattan distance indicates greater similarity between the two vectors. Hamming distance measures the number of different characters between two strings of equal length. A smaller Hamming distance indicates greater similarity between the two strings. Mahalanobis distance considers the covariance matrix of the data and can better reflect the actual distance between data points. The Pearson correlation coefficient measures the linear relationship between two variables, with a value range of [-1, 1], where 1 indicates a perfect positive correlation and -1 indicates a perfect negative correlation.

[0103] In a specific example, the continuous data in the system data i to be configured is calculated. Corresponding continuous data in historical configuration scheme j The Euclidean distance, or continuous similarity, is specifically expressed as:

[0104]

[0105] in, For continuous data in system data i to be configured Corresponding continuous data in historical configuration scheme j The Euclidean distance, where p is continuous data. Dimensions For continuous data The k-th element in For the corresponding continuous data in historical configuration scheme j The k-th element in the data, where n is the continuous data in system data i to be configured. The number of groups, where m is the number of historical configuration schemes in the historical configuration scheme.

[0106] In a specific example, calculate the discrete dataset in the system data i to be configured. The discrete dataset corresponding to the historical configuration scheme j The Jaccard coefficient, or discrete similarity, is specifically expressed as:

[0107]

[0108] in, For the discrete dataset i in the system data to be configured The discrete dataset corresponding to the historical configuration scheme j The Jaccard coefficient.

[0109] Furthermore, by combining the similarity weights corresponding to different types, the local similarities corresponding to each historical configuration scheme are fused to obtain the data similarity corresponding to each historical configuration scheme. The similarity weights include discrete similarity weights and continuous similarity weights, specifically including:

[0110] The discrete similarity and continuous similarity are normalized respectively to obtain the standard discrete similarity and standard continuous similarity;

[0111] By combining discrete similarity weights and continuous similarity weights, the standard discrete similarity and standard continuous similarity are weighted and summed to obtain the data similarity.

[0112] In one specific implementation, the continuous similarity, i.e., the Euclidean distance, is normalized to the range [0,1].

[0113]

[0114] in, For continuous data in system data i to be configured Corresponding continuous data in historical configuration scheme j Standard continuous similarity, For all The maximum value in.

[0115] Since the Jaccard coefficients range from [0,1], there is no need to normalize them.

[0116] Data similarity, specifically represented as:

[0117]

[0118] in, The similarity score between the system data i to be configured and the historical configuration scheme j in the historical configuration scheme library is calculated. For continuous similarity weights, For discrete similarity weights.

[0119] It is understandable that continuous similarity weight represents the impact of continuous data on data similarity, while discrete similarity weight represents the impact of discrete data on data similarity. The values ​​of continuous and discrete similarity weights are set according to the actual situation and are not limited thereto.

[0120] Furthermore, based on the data similarity between the system data to be configured and various historical configuration schemes in the historical configuration scheme library, the system data to be configured is analyzed, and the historical configuration schemes are adjusted to provide a target configuration scheme, which is then referenced. Figure 2 Specifically, it includes:

[0121] Based on data similarity, the historical configuration schemes in the historical configuration scheme library are sorted, and pending historical schemes are given based on the sorting results.

[0122] Based on the pre-obtained configuration mapping function, the system data to be configured is mapped to obtain the target configuration scheme. The pre-obtained configuration mapping function is obtained by analyzing and correcting the historical scheme to be configured.

[0123] Understandably, based on the sorting results, historical configuration schemes with higher data similarity are filtered out to obtain pending historical schemes. These pending historical schemes have higher reference value for the configuration system data. Using pending historical schemes to determine the configuration mapping function can ensure the configuration effect of the target configuration scheme and improve the overall energy utilization efficiency.

[0124] The above configuration mapping function is obtained through model training. In this example, a decision tree model is used; other models can be used in other implementations, and there is no limitation on this. First, the model parameters of the decision tree model are set to complete the initialization of the decision tree model. The setting of model parameters is determined according to actual needs and is not limited in this regard. The planning data of the pending historical schemes are then... and configuration scheme data As the source domain, the system data to be configured As the target domain, it's understandable that the planning data in the source domain and the system data to be configured in the target domain are highly similar in distribution or feature form. Relatively similar distributions can be aligned through a certain mapping, thus the pattern of "planning data → configuration scheme data" in the source domain can be transferred to the target domain. The decision tree model is then applied to the source domain. Training is performed until convergence, yielding the configuration mapping function. After obtaining the configuration mapping function, the system data to be configured is substituted into the configuration mapping function to obtain the target configuration scheme.

[0125] S103: Based on the target configuration scheme, obtain the configuration structure of the integrated energy system, acquire changes in the configuration structure of the integrated energy system in real time, adjust the integrated energy system operation model, and obtain the target operation model.

[0126] Furthermore, based on the target configuration scheme, the integrated energy configuration structure is obtained, specifically including:

[0127] From the target configuration scheme, obtain the various energy devices and loads in the integrated energy system, and determine the nodes of the integrated directed graph;

[0128] Analyze the operating status of each energy device and load in the integrated energy system, and determine the attributes of each node in the integrated directed graph;

[0129] By combining the connection relationships and connection status of various energy devices and loads in the integrated energy system, the edges and edge attributes corresponding to each node in the integrated directed graph are determined.

[0130] By integrating the nodes, attributes, edges, and attributes of the integrated directed graph, the integrated directed graph is constructed, resulting in the configuration structure of the integrated energy.

[0131] In one specific implementation, a directed graph is used. To represent the integrated energy system, we obtain a comprehensive directed graph. N is the set of nodes in the comprehensive directed graph, where each node represents a type of energy device and load in the integrated energy system. For example, N1 is a photovoltaic power station, N2 is a wind turbine, N3 is a gas turbine, N4 is a kerosene generator, and N5 is a hydrogen energy storage device. This represents various types of electrical loads. Each node's attribute vector includes its power generation, load power, energy storage status, and operating status. Power generation applies only to power generation equipment in the integrated energy system, such as photovoltaic and wind turbines; load power applies only to load nodes in the integrated energy system; and energy storage status applies only to energy storage devices. In this example, the operating status value is either 1 or 0, where 1 indicates that the energy equipment or load corresponding to the node is operating normally, and 0 indicates that the energy equipment or load corresponding to the node is out of service. In a specific example, the load power and energy storage status values ​​in the attribute vector of a certain power generation device can be 0, empty, or other forms indicating that the attribute value is unavailable; there are no restrictions on this. Furthermore, for power generation devices, the corresponding attribute vector can only include power generation and operating status; similarly, for loads, the corresponding attribute vector can only include load power and operating status; and for energy storage devices, the corresponding attribute vector can only include energy storage status and operating status.

[0132] E represents the set of edges connecting nodes in a directed graph, such as power transmission lines or energy conversion channels. Each edge... This represents a connection from node i to node j. Each edge... The attribute vector includes transmission capacity, transmission loss, and operating status. The operating status takes the value 1 or 0, where 1 indicates that the corresponding edge is operating normally and 0 indicates that the corresponding edge is faulty and is not operating normally.

[0133] By constructing a comprehensive directed graph through the planning structure of the integrated energy system, the operating status and parameters of each energy device and load and their connection relationships in the integrated energy system can be obtained in a timely and clear manner. It can also capture faults or topology changes in the integrated energy system in a timely manner, providing a foundation for timely adjustment of the operation of the integrated energy system and ensuring its stable operation.

[0134] Furthermore, by acquiring real-time changes in the configuration structure of the integrated energy system, adjusting the integrated energy operation model, and obtaining the target operation model, the following steps are taken: Figure 3 Specifically, it includes:

[0135] Based on the changes in the configuration structure of integrated energy, the changes in the configuration structure are quantified, and the state changes of integrated energy are given.

[0136] Based on the pre-acquired state change threshold, the state change amount is judged, and a dynamic exploration factor is given. The dynamic exploration factor is used to influence the action selection of the comprehensive energy at the next moment.

[0137] Based on dynamic exploration factors and combined with the system loss function, the integrated energy operation model is trained until convergence, thus obtaining the target operation model.

[0138] Furthermore, based on the changes in the configuration structure of integrated energy, the changes in the configuration structure are quantified, and the state changes of integrated energy are given, specifically including:

[0139] Obtain the composite directed graph at two adjacent time points, analyze the changes in the number of nodes and edges in the composite directed graph at the corresponding two time points, and obtain the change results;

[0140] Based on the changes, we analyze the changes of each node and edge in the comprehensive directed graph at two adjacent time points to obtain the changes in nodes and edges.

[0141] By combining the changes in node values ​​and edge values, we obtain the state change values.

[0142] In one specific implementation, the combined directed graph at two adjacent time points is obtained, that is, the combined directed graph at time t is obtained. and the comprehensive directed graph at time t+1 Analyzing the changes in the number of nodes in the two composite directed graphs, we obtain... ,in, Let be the set of nodes from time t+1. Find the set of nodes at time t. Nodes not present in the text Let be the set of nodes from time t. Find the set of nodes at time t+1. Nodes that are not present in the table. If... ,Right now It is not an empty set, indicating that there are additions or deletions of nodes in the directed graph between time t and time t+1. This represents the empty set.

[0143] Continuing to analyze the changes in the number of edges in the two composite directed graphs, we obtain... ,in, Let be the set of edges from time t+1. Find the set of edges at time t. There is no edge in the middle. Let be the set of edges from time t. Find the set of edges at time t+1. There is no edge in the middle. If , which indicates the addition or deletion of edges in the composite directed graph between time t and time t+1.

[0144] if or If this is the case, then the configuration structure of the integrated energy source is determined to have changed, meaning the result of the change is a change. If and If the configuration structure of the integrated energy source remains unchanged, then the result is considered unchanged.

[0145] When the configuration structure of the integrated energy changes, the changes in nodes and edges are calculated to obtain the state changes.

[0146] In the embodiments provided by the present invention, Euclidean distance is used to quantify the impact of configuration structure changes.

[0147] The node changes are specifically represented as follows:

[0148]

[0149] The change in edge is specifically represented as:

[0150]

[0151] Furthermore, the change in state is specifically expressed as:

[0152]

[0153] in, Let be the state change quantity corresponding to the comprehensive directed graph from time t to time t+1. This represents the change in the number of nodes in the directed graph from time t to time t+1. This represents the change in the number of edges in the directed graph from time t to time t+1. It is the second norm. Let be the attribute vector of node i at time t. For the edge The attribute vector at time t, where E is the set of edges in the composite directed graph, and n is the number of nodes in the composite directed graph.

[0154] The aforementioned state change threshold is determined based on the mean and standard deviation of the state change, which are calculated by acquiring past state changes under normal operating conditions of the integrated energy system. In a specific example, the state change threshold is obtained by summing the product of a pre-set empirical coefficient and the standard deviation with the mean. In other implementations, other methods may be used, and there are no limitations on this.

[0155] Specifically, the construction of the integrated energy operation model is determined through the following steps:

[0156] Based on the initial deep learning model, combined with the integrated directed graph and the system action space, the expected cumulative reward values ​​corresponding to different states and actions are obtained. The system action space includes various actions performed on each energy device and load in the integrated energy system.

[0157] Based on the expected cumulative reward value corresponding to different states and actions, the mean and variance of the expected cumulative reward value are given, and the state distribution of the expected cumulative reward value corresponding to different states and actions is obtained.

[0158] Based on the state distribution of the expected cumulative reward value, the model parameters of the integrated energy operation model are determined, and the construction of the integrated energy operation model is completed.

[0159] In the embodiments provided by this invention, the first step is to initialize the comprehensive directed graph, the system action space, and the model parameters. Target model parameters And the experience replay buffer. Among them, model parameters... Function for estimating Q-value Target model parameters Used for stabilizing the Q-value function During the training process, the experience replay buffer is used to store the agent's experiences interacting with the environment, including states, actions, rewards, and the next state. In this example, the Q-value function... This represents the expected cumulative reward value.

[0160] It's important to understand that in reinforcement learning, an intelligent agent is an entity capable of perceiving its environment and autonomously taking actions to achieve a specific goal. Intelligent agents learn optimal behavioral strategies through interaction with their environment. They possess autonomy, enabling them to operate without direct human intervention and make decisions based on internal logic. They have perception capabilities, sensing the state of their environment through sensors. They possess decision-making capabilities, selecting the optimal action strategy based on perceived environmental information. They are adaptive, adjusting their behavior according to changes in the environment. Finally, they are interactive, able to interact with their environment or other intelligent agents.

[0161] Then, experience collection is performed. At each time step... To obtain the current operating status of the integrated energy system Select Action And execute, observe the next state. and reward function Finally Stored in the experience replay buffer.

[0162] action The determination of is specifically expressed as:

[0163]

[0164] in, The mean of the Q values, The variance of the Q value, This is a dynamically adjusted factor.

[0165] Among them, the above reward function Specifically, it is obtained through the following steps:

[0166] Based on the current operating status and scheduling operations of the integrated energy system, the operation and carbon emissions of the integrated energy system are analyzed to determine the system power weight coefficient, system energy storage weight coefficient and system low carbon weight coefficient.

[0167] Based on the connection relationship between various energy devices and loads in the integrated energy system, analyze the operation and carbon emission of each energy device and load, and determine the equipment power weighting coefficient and the equipment energy storage weighting coefficient.

[0168] By combining the carbon emission factors of each energy device and load with the current operating status of each energy device and load in the integrated energy system, the reward function of the integrated energy system is determined.

[0169] The aforementioned system power weighting coefficient, system energy storage weighting coefficient, and system low-carbon weighting coefficient are pre-set based on the overall architecture of the integrated energy system, while the equipment power weighting coefficient and equipment energy storage weighting coefficient are pre-set based on different energy equipment and loads in the integrated energy system.

[0170] By constructing a reward function for integrated energy systems, the quality of the current state and actions of the integrated energy system is quantified. The resulting reward function value guides the integrated energy operation model in regulating the system at the next moment. In the integrated energy system, the reward function comprehensively considers factors such as power balance, energy utilization efficiency, equipment operating costs, and system stability. Based on the power generation and load power of each node, the power balance of the integrated energy system is considered to ensure a balance between power generation and load power, avoiding overload or insufficient power supply. Simultaneously, the operating costs of the integrated energy system are considered to minimize the costs of power generation and transmission. Based on the energy storage status of each energy storage device, frequent switching of device states is avoided to ensure stable system operation under various operating conditions. The carbon emission situation of different energy sources is also considered to maximize the utilization of renewable energy and reduce dependence on fossil fuels.

[0171] reward function It includes an integrated energy generation component, an integrated energy storage component, and an integrated energy low-carbon operation component. The integrated energy generation component is determined through the following steps:

[0172] Analyze the power generation and load power of various energy devices and loads, and give the power difference between power generation and load power;

[0173] By combining the power weighting coefficients of various energy devices and loads, the squares of the power differences are weighted and summed, and the system power weighting coefficients of integrated energy are integrated to obtain the integrated energy power generation part.

[0174] The integrated energy storage component is determined through the following steps:

[0175] Analyze the current energy storage status of various energy storage devices, combine the corresponding device energy storage weight coefficients, and perform a weighted summation of the absolute values ​​of the energy storage status. Integrate the system energy storage weight coefficients of comprehensive energy to obtain the comprehensive energy storage component.

[0176] The integrated energy low-carbon operation component is determined through the following steps:

[0177] By analyzing the power generation capacity of various energy sources at the current moment, combining the carbon emission factors of various energy sources, weighted summation of the power generation capacity, and integrating the system low-carbon weight coefficient of integrated energy, the low-carbon operation component of integrated energy is obtained.

[0178] Furthermore, the reward function Specifically, it is expressed as:

[0179]

[0180] in, Let be the reward function value of the comprehensive energy at time t. This is the system power weighting coefficient. Let be the equipment power weighting coefficient for the i-th power generation device. This is the system energy storage weighting coefficient. Let be the energy storage weight coefficient of the j-th energy storage device. The system's low-carbon weighting coefficient. For the carbon emission factor of energy type k, Let p be the power generation of the k-th energy type at time t, and p be the total amount of each type of energy. Let be the power generation of node i at time t. Let be the load power of node i at time t. Let represent the energy storage state of node j at time t.

[0181] In the embodiments provided by the present invention, the exploration rate is... Choosing random actions and exploring the unknown action space helps agents try new strategies when facing unknown or uncertain environments, avoiding reliance solely on currently known optimal strategies. (Based on the exploration rate...) Choose the optimal action based on the current Q-value and uncertainty, selecting the action most likely to yield the maximum reward. As a basis for decision-making For action a in state The estimated value below, This is the variance of the estimated value of the action. Exploration rate. It will dynamically adjust over time; in the initial stages, the agent will typically conduct more exploration (i.e., (At a relatively high level), but as learning progresses, intelligence relies more on the knowledge already acquired, that is... Increase Decrease. In this example, the initial value of the exploration rate can be set according to the actual situation, and then combined with the set decay rate, the exploration rate is adjusted to explore the unknown action space and find the optimal action.

[0182] In a specific example, in a microgrid, due to overcast weather, solar power generation is insufficient to meet demand. At this time, action... This could involve increasing the output power of the wind turbine or discharging energy from battery storage to ensure that load demands are met and that the system remains stable and reliable.

[0183] Dynamic exploration factors, specifically represented as:

[0184]

[0185] in, Let be the dynamic exploration factor at time t. As the initial exploration weight, The influence coefficient of the configuration changes of the integrated energy system on the dynamic exploration factor. This represents the state change of the integrated directed graph from time t to time t+1.

[0186] After completing experience collection, experience playback and network updates are performed. A portion of experience samples are randomly sampled from the experience playback buffer. And calculate the target Q value based on the sampling experience.

[0187]

[0188] in, Let the target Q value be the j-th experience obtained from sampling. As a discount factor, For instant rewards, This indicates that an action can be performed on all possible next states. Take the maximum value. Perform an action for the next state of the j-th experience. The target Q value. The discount factor represents the importance of future rewards and ranges from [0,1]. In this example, the discount factor ranges from [0.9,0.99].

[0189] Optimize model parameters using backpropagation and gradient descent. And update the target model parameters regularly. .

[0190]

[0191] in, The learning rate is used to control model parameters. Update step size, loss function Regarding model parameters The gradient of the target model parameters. The model is updated every fixed number of steps or fixed time steps to stabilize the training process of the integrated energy operation model.

[0192] During the training of the integrated energy operation model, a system loss function is constructed to optimize the state distribution of model parameters, enabling the integrated energy operation model to more accurately fit the data and capture uncertainties. Overfitting is avoided without sacrificing the expressive power of the integrated energy operation model.

[0193] The system loss function is obtained, specifically including:

[0194] Analyze the different model parameters in the integrated energy operation model, analyze the changes in the Q value corresponding to the model parameters, and construct the likelihood loss function.

[0195] Based on the prior and approximate posterior distributions of the model parameters in the integrated energy operation model, the KL divergence of the prior and approximate posterior distributions is given, and the KL divergence loss function is obtained.

[0196] The system loss function is obtained by combining the likelihood loss function and the KL divergence loss function.

[0197] The system loss function is specifically expressed as:

[0198]

[0199] in, Let be the system loss function. Let be the likelihood loss function. Let KL be the function value of the KL divergence loss function KL(), representing the prior distribution of model parameters in the integrated energy operation model. and approximate posterior distribution The Kullback-Leibler (KL) divergence between two probability distributions For regularization weights, Let B be the absolute value of the empirical sample set. Let Q be the target Q value corresponding to the j-th empirical sample in the empirical sample set B. In the state and actions Q value under the condition, The prior distribution of model parameters in the integrated energy operation model. This represents the approximate posterior distribution of the model parameters in the integrated energy operation model.

[0200] Likelihood loss function is used for fitting The KL divergence loss function is used to constrain the state distribution of model parameters, measure the difference between two probability distributions, and compare the similarity between two probability distributions.

[0201] In other implementations, when an error exists in the synthesized directed graph, the synthesized directed graph is first updated based on the new topology. First, the system action space A is configured to ensure that all newly added or deleted nodes and edges are correctly represented in the state vector and action set. Second, the experience replay buffer is updated. The experience replay buffer is cleared or partially cleared to prevent data from the old topology from interfering with the learning of the new strategy. New operational data of the integrated energy system is collected, and under the new topology, the system re-interacts with the environment, gradually filling the new experience replay buffer. Finally, the new experience data is used for training. The process of experience collection, experience replay, and network updates continues, using the new data to optimize the integrated energy operation model, adapting it to the new topology to achieve optimal control of the integrated energy system.

[0202] S104: Based on the target operation model, obtain real-time operation data of integrated energy.

[0203] Understandably, based on the actual needs of the integrated energy system (i.e., the data of the system to be configured), a corresponding target configuration scheme is obtained. Based on the target configuration scheme, the configuration structure of the integrated energy system is derived, completing its construction. Then, the integrated energy system operates based on its operational model. During operation, if the configuration structure changes—for example, if some lines experience problems or some equipment malfunctions—the operational model is adjusted based on this change, resulting in a target operational model. The integrated energy system continues to operate according to the operational strategy provided by the target operational model. The target operational model serves as the basis for the current operational strategy, while the previous operational model serves as the basis for the previous operational strategy. When the configuration structure of the integrated energy system remains unchanged, the operational model and the target operational model are identical.

[0204] Therefore, the integrated energy system operates according to the operational strategy given by the target operation model, and real-time operational data of the integrated energy system can be obtained. Real-time operational data includes the power generation, load power, energy storage status and operation status of various energy devices and loads in the integrated energy system, as well as energy production data (such as the output of various energy sources, operating parameters of production equipment, etc.), energy consumption data (energy consumption of different users or regions, consumption patterns, etc.), energy conversion data (such as conversion efficiency and losses of conversion equipment, etc.), storage data (storage capacity of energy storage devices, charging and discharging parameters, etc.), and environmental data (ambient temperature, humidity, etc.).

[0205] S105: Combine the pre-built evaluation index library to extract features from real-time running data to obtain target evaluation features.

[0206] Reference Figure 4 Specifically, it includes:

[0207] Based on a pre-built comprehensive energy index library, corresponding initial assessment features are selected from real-time operation data;

[0208] Principal component analysis was used to analyze the initial evaluation features and identify the core features.

[0209] The target evaluation features are determined based on the similarity between each initial evaluation feature and the core feature.

[0210] The aforementioned integrated energy index database collects historical operating data of integrated energy systems and conducts in-depth analysis using data mining techniques and statistical analysis methods. It analyzes the fluctuation patterns of energy supply and consumption under different seasons and operating conditions to extract key parameters closely related to system performance as indicators. It uses methods such as cluster analysis to identify high-incidence patterns of energy equipment failures and incorporates relevant characteristics into the equipment operation index system. It also extensively studies successful evaluation cases of similar integrated energy systems at home and abroad to select indicators.

[0211] By constructing a comprehensive energy index database, assessment characteristics relevant to the operation of the comprehensive energy system are initially screened from a large amount of operational data. This initial screening reduces the computational intensity of subsequent core characteristic analysis, thereby improving the efficiency of the assessment process while ensuring the accuracy of the comprehensive energy system assessment.

[0212] Furthermore, principal component analysis is used to analyze the initial evaluation features and identify the core features, which specifically include:

[0213] Analyze the degree of linear correlation between the initial evaluation features and construct the covariance matrix of the initial evaluation features;

[0214] The covariance matrix is ​​subjected to eigenvalue decomposition, which yields multiple eigenvalues.

[0215] Based on the variance explanation ratio of each feature value and combined with a preset explanation threshold, the core features are obtained by filtering among multiple feature values.

[0216] Furthermore, the linear correlation between the initial evaluation features is analyzed, and the covariance matrix of the initial evaluation features is constructed, specifically including:

[0217] Based on the elements in each initial evaluation feature and the feature mean of the corresponding initial evaluation feature, the deviation feature of each initial evaluation feature is given.

[0218] Based on the deviation features of any two initial evaluation features, the mean of the product of the two deviation features is given, and the degree of linear correlation between the initial evaluation features is obtained.

[0219] By integrating the degree of linear correlation among the various initial evaluation features, a covariance matrix of the initial evaluation features is constructed.

[0220] Specifically, the degree of linear correlation between the various initial evaluation features is expressed as follows:

[0221]

[0222] in, Let m represent the degree of linear correlation between the u-th and v-th initial evaluation features, and m be the sample size. Let k be the feature value in the u-th initial evaluation feature. Let be the feature mean of the u-th initial evaluation feature. For the v-th initial evaluation feature, the k-th feature value is... Let be the feature mean of the v-th initial evaluation feature.

[0223] Bias features refer to a matrix composed of the differences between each feature value and the feature mean in the initial evaluation features.

[0224] The covariance matrix is ​​specifically represented as:

[0225]

[0226] Where C is the covariance matrix, is a matrix composed of standardized initial evaluation features, where m is the sample size. Transpose matrix.

[0227] Furthermore, multiple eigenvalues ​​are specifically represented as follows:

[0228]

[0229] Where C is the covariance matrix of the initial evaluation features. Let be a diagonal matrix, where each element on the diagonal is an eigenvalue. It is the identity matrix. Let V be the transpose of V, where V is the eigenvector matrix, and the elements in the eigenvector matrix are the eigenvectors corresponding to the eigenvalues.

[0230] In one specific implementation, the covariance matrix C is decomposed into eigenvalues. and the corresponding feature vector The eigenvalue decomposition is specifically represented as... ,in, It is a diagonal matrix, and the elements on the diagonal are eigenvalues. And they are arranged in descending order. V is composed of eigenvectors. The matrix formed satisfies .

[0231] Furthermore, based on the variance explanation ratio of each feature value and a preset explanation threshold, core features are obtained by filtering among multiple feature values, specifically including:

[0232] Based on the variance explained by each eigenvalue, the cumulative variance explained by each eigenvalue is given.

[0233] The cumulative variance explained by each feature value is compared with the preset explanation threshold to obtain the comparison results;

[0234] Based on the comparison results, the feature values ​​are filtered, and the feature vectors corresponding to the filtered feature values ​​are given.

[0235] The core features are obtained by projecting the standard data matrix onto a vector matrix composed of the feature vectors corresponding to the filtered feature values.

[0236] In one specific implementation, the feature values ​​are arranged in descending order, and the core features are selected based on the cumulative variance interpretation ratio.

[0237] Calculate the proportion of cumulative variance explained for each eigenvalue:

[0238]

[0239] in, The proportion of cumulative variance explained by the k-th eigenvalue. Let be the proportion of variance explained by the i-th eigenvalue. Let be the i-th eigenvalue, and m be the number of eigenvalues.

[0240] In a specific example, the preset interpretation threshold is: Select the one that satisfies The minimum value k is used as the number of principal components. After selecting k principal components, the standardized data matrix is... Projected onto a matrix consisting of the first k eigenvectors The core features are obtained from the above. Specifically, it is expressed as:

[0241]

[0242] in, It consists of the first k eigenvectors Matrix, core features for The matrix, core features elements in This represents the score of the i-th sample on the j-th principal component.

[0243] Specifically, based on the similarity between the initial evaluation features and the core features, and combined with a preset similarity range, the initial evaluation features are screened to provide target evaluation features;

[0244] Based on the information content of the target evaluation features, the corresponding evaluation weights are given.

[0245] In the embodiments provided by the present invention, the similarity between each initial evaluation feature and the core feature is represented by the Pearson correlation coefficient. In other embodiments, other methods may be used to determine the similarity between each evaluation feature and the core feature, and there is no limitation on this.

[0246] In the embodiments provided by this invention, a similarity range is set, and it is determined whether the similarity is within the preset similarity range. If it is, the initial evaluation feature is used as the target evaluation feature; otherwise, if the similarity is not within the preset similarity range, the initial evaluation feature is removed and not used as the target evaluation feature for evaluating the integrated energy system. The target evaluation features obtained after screening are closely related to the core features, can deeply reflect the key features and operational essence of the integrated energy system, effectively avoid interference from irrelevant or weakly related indicators, significantly improve the targeting and effectiveness of the evaluation, and ensure that the evaluation focuses on core elements and accurately understands system performance.

[0247] S106: Integrate the evaluation weights corresponding to the evaluation characteristics of the target, and give the comprehensive energy operation evaluation results.

[0248] Among them, reference Figure 5 The evaluation weights corresponding to the target evaluation features are obtained through the following steps:

[0249] Based on the values ​​of each element in the target evaluation feature, the information content of the target evaluation feature is given;

[0250] By combining the information content of the target evaluation features, the degree of variation of the target evaluation features is analyzed;

[0251] Based on the degree of variation of all target evaluation features, and in combination with the degree of variation of target evaluation features, the evaluation weights corresponding to the target evaluation features are given.

[0252] Furthermore, the evaluation weights corresponding to the target evaluation features are specifically expressed as follows:

[0253]

[0254] in, Let the evaluation weight be the evaluation feature corresponding to the j-th target. Let m be the information content of the j-th target feature weight, and m be the number of target evaluation features. To assess the degree of variability of the target features.

[0255] In the embodiments provided by the present invention, the entropy weight method is used to set the evaluation weights. The information entropy of each target evaluation feature is calculated to reflect the information content of each target evaluation feature. The weights are allocated entirely based on the inherent information features of the data, ensuring the objectivity and scientific nature of the weight setting.

[0256] Information entropy reflects the amount of information contained in the target evaluation features. The greater the information entropy of the target evaluation features, the smaller the degree of variation of the target evaluation features, the less information they provide, and the smaller their role in the comprehensive evaluation. Conversely, the smaller the information entropy of the target evaluation features, the greater the degree of variation of the target evaluation features, the more information they provide, and the greater their weight should be.

[0257] By determining the evaluation weights through the information entropy of the target evaluation features, the weight allocation is based entirely on the information features of the data itself, avoiding interference from human factors, ensuring the objectivity and scientific nature of the weight setting, and making the evaluation results more convincing and reliable.

[0258] After determining the evaluation weights, the target evaluation characteristics and their corresponding evaluation weights are weighted and summed to obtain the comprehensive energy operation evaluation results.

[0259] S107: Based on the operational assessment results, adjust the target operational model and complete the configuration of integrated energy.

[0260] Furthermore, referring to Figure 6 The target operation model includes a reward function, which consists of a comprehensive energy generation component, a comprehensive energy storage component, and a comprehensive energy low-carbon operation component.

[0261] Based on the operational assessment results, the target operational model is adjusted to complete the configuration of integrated energy resources, specifically including:

[0262] If the operational evaluation result is unsatisfactory, then analyze the sub-scores of each part of the operational evaluation result;

[0263] Analyze the achievement status of each sub-score, and adjust the coefficients of the corresponding part of the reward function in the target running model according to the achievement status;

[0264] The target running model is trained based on the adjusted reward function until convergence;

[0265] Based on the target operating model after training, run the integrated energy system and complete the configuration of the integrated energy system.

[0266] It can be understood that the aforementioned target operation model is obtained through adjustment and training of the integrated energy operation model. Since the integrated energy operation model uses a reward function during its training and adjustment process, the adjustment of the target operation model will also utilize the reward function. Based on this, and according to the judgment of the operation evaluation results, the various parts of the reward function are adjusted. According to the adjusted reward function, the target operation model is adjusted and trained again to obtain a new target operation model. The integrated energy system will operate according to the operation strategy given by the new target operation model, providing new real-time operation data and new operation evaluation results. This process is repeated until the operation evaluation results meet the needs of the integrated energy system, i.e., the operation evaluation results are deemed satisfactory.

[0267] In one specific implementation, the operational evaluation result ranges from 0 to 1, and is graded according to the performance of the integrated energy system. For example, five grades can be set: Excellent (operational evaluation result between 0.8 and 1), Good (operational evaluation result between 0.6 and 0.8), Average (operational evaluation result between 0.4 and 0.6), Poor (operational evaluation result between 0.2 and 0.4), and Very Poor (operational evaluation result between 0 and 0.2). This grading helps to intuitively understand the relative position and performance level of the integrated energy system within similar systems. The grading criteria can refer to industry averages, performance data of advanced benchmark systems, and relevant policies, standards, and specifications. In the example provided in this application, Excellent and Good are considered compliant, while Average, Poor, and Very Poor are considered non-compliant. The operational evaluation result is a result of evaluating the overall performance of the integrated energy system. If the operational evaluation result is compliant, but the parameter settings of some equipment in the integrated energy system are non-compliant, then the overall result is the guiding principle, and adjustments are not made to the small non-compliant parts. If the operational evaluation result is unsatisfactory, it indicates that there is a problem with the overall operational configuration of the integrated energy system. At this time, it is necessary to analyze which part is unsatisfactory, adjust the reward function according to the unsatisfactory situation, and give a new target operational model to complete the configuration of the integrated energy system.

[0268] As the aforementioned analysis shows, the operational data of integrated energy systems can be divided into three parts: data related to power generation equipment, data related to energy storage equipment, and data reflecting the carbon emissions of power generation equipment, energy storage equipment, and various energy sources. That is, each target assessment characteristic used to determine the operational assessment result can also be categorized into at least one of these three parts. In a specific example, when the operational assessment result is non-compliant, the compliance status of each target assessment characteristic is analyzed. Based on the corresponding compliance analysis table for each target assessment characteristic, if the value corresponding to the target assessment characteristic falls within the non-compliant range, then that target assessment characteristic is non-compliant; conversely, if the value corresponding to the target assessment characteristic falls within the compliant range, then that target assessment characteristic is compliant. The compliance analysis table corresponding to each target assessment characteristic is pre-set, and the settings may vary in different integrated energy systems; no specific limitation is imposed. The analysis determines whether each non-compliant target assessment characteristic belongs to the power generation equipment, energy storage equipment, or carbon emission component. Based on the number or proportion of each non-compliant target assessment characteristic in the corresponding component, the adjustment parameters corresponding to the reward function are determined. In different implementations, the aforementioned sub-scores can be the feature values ​​of each target evaluation feature, the number of each non-compliant target evaluation feature in the corresponding part, or its proportion, without limitation. For example, if only 5 non-compliant target evaluation features belong to power generation equipment, the corresponding system power weight coefficient in the reward function will be increased. As another example, if there are 10 target evaluation features belonging to power generation equipment, and 5 are non-compliant, the proportion of non-compliant features is 0.5, the system power weight coefficient in the reward function will be adjusted according to the adjustment parameter corresponding to 0.5. If the non-compliant target evaluation features involve multiple parts, the coefficients corresponding to different parts will be adjusted separately. The correspondence between the number or proportion of different target evaluation features and the corresponding adjustment parameters is preset according to the actual scenario.

[0269] Based on the operational assessment results, specific decision-making recommendations are provided for the optimization and development of the integrated energy system. Regarding energy supply, if supply stability is insufficient, consideration can be given to increasing backup energy supply equipment, optimizing energy dispatch strategies, or strengthening interconnection with surrounding energy systems to improve supply reliability. In the energy consumption phase, if the energy consumption structure is unreasonable, with excessive reliance on a single high-pollution or high-cost energy source, the transition to clean energy through energy-saving equipment can be promoted. For equipment operation, if the equipment failure rate is high, it is recommended to increase investment in equipment maintenance, establish a full life-cycle management system for equipment, and plan equipment upgrades in advance. Regarding environmental indicators, if carbon emissions exceed standards, measures such as promoting the construction of renewable energy projects and implementing carbon capture and storage (CCS) technology pilot projects can be implemented to reduce the system's environmental impact and achieve sustainable development of the integrated energy system.

[0270] Reference Figure 7This invention provides an integrated energy configuration device, employing an integrated energy configuration method as described above, comprising:

[0271] The sensing module is used to acquire data about the system to be configured.

[0272] The storage module is used to store a library of historical configuration schemes.

[0273] The calculation module analyzes the system data to be configured based on the data similarity between the system data to be configured and the historical configuration schemes in the historical configuration scheme library, adjusts the historical configuration schemes, and provides a target configuration scheme. Based on the target configuration scheme, it obtains the configuration structure of the integrated energy system and acquires real-time changes in the configuration structure, adjusts the integrated energy system operation model, and obtains the target operation model. Based on the target operation model, it obtains real-time operation data of the integrated energy system. Combining a pre-built evaluation index library, it extracts features from the real-time operation data to obtain target evaluation features. It integrates the evaluation weights corresponding to the target evaluation features to provide the operation evaluation results of the integrated energy system. Based on the operation evaluation results, it adjusts the target operation model to complete the configuration of the integrated energy system.

[0274] In addition, the present invention also provides an integrated energy configuration system, including: a network device, an execution device, and an integrated energy configuration device as described above.

[0275] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process of the described module can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0276] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0277] The units described in the embodiments of this disclosure can be implemented in software or hardware. The names of the units are not, in some cases, intended to limit the specific unit.

[0278] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention. Clearly, those skilled in the art can make various alterations and modifications to the invention without departing from its spirit and scope. Thus, if these modifications and variations of the invention fall within the scope of the claims and their equivalents, the invention is also intended to include these modifications and variations.

Claims

1. A method for configuring integrated energy, characterized in that, include: Obtain the system data to be configured; Based on the data similarity between the system data to be configured and the historical configuration schemes in the historical configuration scheme library, the system data to be configured is analyzed, the historical configuration schemes are adjusted, and the target configuration scheme is given. Based on the target configuration scheme, the configuration structure of the integrated energy system is obtained. The changes in the configuration structure are quantified, and the state change of the integrated energy system is given. Based on a pre-obtained state change threshold, the state change is judged, and a dynamic exploration factor is given. This dynamic exploration factor influences the action selection of the integrated energy system at the next moment. Based on the dynamic exploration factor and combined with the system loss function, the integrated energy system operation model is trained until convergence, yielding the target operation model. The construction of the integrated energy system operation model is determined through the following steps: Based on the initial deep learning model, combined with the integrated directed graph and the system action space, the expected cumulative reward values ​​corresponding to different states and actions are obtained. The system action space includes various actions performed on each energy device and load in the integrated energy system. Based on the expected cumulative reward values ​​corresponding to different states and actions, the mean and variance of the expected cumulative reward values ​​are given, resulting in a state distribution of the expected cumulative reward values ​​corresponding to different states and actions. Based on the state distribution of the expected cumulative reward values, the model parameters of the integrated energy system operation model are determined, completing the construction of the integrated energy system operation model. Based on the target operation model, real-time operation data of integrated energy is obtained; By combining a pre-built evaluation index library, feature extraction is performed on real-time operational data to obtain target evaluation features; By integrating the assessment weights corresponding to the assessment characteristics of the target, a comprehensive energy operation assessment result is given; Based on the operational assessment results, the target operational model is adjusted to complete the configuration of integrated energy resources.

2. The method for configuring integrated energy as described in claim 1, characterized in that, The similarity between the system data to be configured and the data of each historical configuration scheme in the historical configuration scheme library is obtained through the following steps: Based on the type of system data to be configured, a similarity index analysis strategy is used to analyze the local similarity with the historical data corresponding to each historical configuration scheme. By combining the similarity weights corresponding to different types, the local similarity of each historical configuration scheme is fused to obtain the data similarity of each historical configuration scheme.

3. The method for configuring integrated energy as described in claim 2, characterized in that, The types of data to be configured include discrete and continuous types, and the local similarity includes discrete similarity and continuous similarity. Based on the type of system data to be configured, a similarity index analysis strategy is used, and the local similarity with historical data corresponding to each historical configuration scheme is analyzed, specifically including: The first similarity index is used to analyze the discrete similarity between discrete system data to be configured and corresponding discrete data in historical configuration schemes; The second similarity index is used to analyze the continuous similarity between the continuous data of the system to be configured and the corresponding continuous data in the historical configuration scheme.

4. The method for configuring integrated energy as described in claim 3, characterized in that, Similarity weights include discrete similarity weights and continuous similarity weights; By combining the similarity weights corresponding to different types, the local similarity scores of each historical configuration scheme are fused to obtain the data similarity scores for each historical configuration scheme, specifically including: The discrete similarity and continuous similarity are normalized respectively to obtain the standard discrete similarity and standard continuous similarity; By combining discrete similarity weights and continuous similarity weights, the standard discrete similarity and standard continuous similarity are weighted and summed to obtain the data similarity.

5. The method for configuring integrated energy as described in claim 1, characterized in that, Based on the data similarity between the system data to be configured and various historical configuration schemes in the historical configuration scheme library, the system data to be configured is analyzed, and the historical configuration schemes are adjusted to provide a target configuration scheme, specifically including: Based on data similarity, the historical configuration schemes in the historical configuration scheme library are sorted, and pending historical schemes are given based on the sorting results. Based on the pre-obtained configuration mapping function, the system data to be configured is mapped to obtain the target configuration scheme. The pre-obtained configuration mapping function is obtained by analyzing and correcting the historical scheme to be configured.

6. The method for configuring integrated energy as described in claim 1, characterized in that, Based on the target configuration scheme, the integrated energy configuration structure is obtained, specifically including: From the target configuration scheme, obtain the various energy devices and loads in the integrated energy system, and determine the nodes of the integrated directed graph; Analyze the operating status of each energy device and load in the integrated energy system, and determine the attributes of each node in the integrated directed graph; By combining the connection relationships and connection status of various energy devices and loads in the integrated energy system, the edges and edge attributes corresponding to each node in the integrated directed graph are determined. By integrating the nodes, attributes, edges, and attributes of the integrated directed graph, the integrated directed graph is constructed, resulting in the configuration structure of the integrated energy.

7. The method for configuring integrated energy as described in claim 1, characterized in that, Based on changes in the configuration structure of integrated energy, the changes in configuration structure are quantified, and the state changes of integrated energy are given, specifically including: Obtain the composite directed graph at two adjacent time points, analyze the changes in the number of nodes and edges in the composite directed graph at the corresponding two time points, and obtain the change results; Based on the changes, we analyze the changes of each node and edge in the integrated directed graph at two adjacent time points to obtain the changes in nodes and edges. By combining the changes in node values ​​and edge values, we obtain the state change values.

8. The method for configuring integrated energy as described in claim 1, characterized in that, By combining a pre-built evaluation index library, feature extraction is performed on real-time operational data to obtain target evaluation features, specifically including: Based on a pre-built comprehensive energy index library, corresponding initial assessment features are selected from real-time operation data; Principal component analysis was used to analyze the initial evaluation features and identify the core features. The target evaluation features are determined based on the similarity between each initial evaluation feature and the core feature.

9. The method for configuring integrated energy as described in claim 8, characterized in that, Principal component analysis was used to analyze the initial evaluation features and identify the core features, which include: Analyze the degree of linear correlation between the initial evaluation features and construct the covariance matrix of the initial evaluation features; The covariance matrix is ​​subjected to eigenvalue decomposition, which yields multiple eigenvalues. Based on the variance explanation ratio of each feature value and combined with a preset explanation threshold, the core features are obtained by filtering among multiple feature values.

10. The method for configuring integrated energy as described in claim 9, characterized in that, Analyze the degree of linear correlation among the initial evaluation features and construct the covariance matrix of the initial evaluation features, specifically including: Based on the elements in each initial evaluation feature and the feature mean of the corresponding initial evaluation feature, the deviation feature of each initial evaluation feature is given. Based on the deviation features of any two initial evaluation features, the mean of the product of the two deviation features is given, and the degree of linear correlation between the initial evaluation features is obtained. By integrating the degree of linear correlation among the various initial evaluation features, a covariance matrix of the initial evaluation features is constructed.

11. The method for configuring integrated energy as described in claim 9, characterized in that, Based on the variance explanation ratio of each feature value and a preset explanation threshold, core features are obtained by filtering from multiple feature values. These core features include: Based on the variance explained by each eigenvalue, the cumulative variance explained by each eigenvalue is given. The cumulative variance explained by each feature value is compared with the preset explanation threshold to obtain the comparison results; Based on the comparison results, the feature values ​​are filtered, and the feature vectors corresponding to the filtered feature values ​​are given. The core features are obtained by projecting the standard data matrix onto a vector matrix composed of the feature vectors corresponding to the filtered feature values.

12. A device for configuring integrated energy, characterized in that, The method for configuring integrated energy as described in any one of claims 1-11 includes: The sensing module is used to acquire data about the system to be configured. The storage module is used to store a library of historical configuration schemes. The calculation module analyzes the system data to be configured based on the data similarity between the system data to be configured and the historical configuration schemes in the historical configuration scheme library, adjusts the historical configuration schemes, and provides the target configuration scheme. Based on the target configuration scheme, it obtains the configuration structure of the integrated energy system, quantifies the changes in the configuration structure based on these changes, and provides the state change amount of the integrated energy system. Based on a pre-acquired state change threshold, it judges the state change amount and provides a dynamic exploration factor, which influences the action selection of the integrated energy system at the next moment. Based on the dynamic exploration factor and the system loss function, it trains the integrated energy system operation model until convergence, obtaining the target operation model. The construction of the integrated energy system operation model is determined through the following steps: based on the initial deep learning model, combined with the integrated directed graph and system dynamic... The system action space is used to obtain the expected cumulative reward values ​​corresponding to different states and actions. The system action space includes various actions performed on each energy device and load in the integrated energy system. Based on the expected cumulative reward values ​​corresponding to different states and actions, the mean and variance of the expected cumulative reward values ​​are given, resulting in a state distribution of the expected cumulative reward values ​​for different states and actions. According to the state distribution of the expected cumulative reward values, the model parameters of the integrated energy operation model are determined, completing the construction of the integrated energy operation model. Based on the target operation model, real-time operation data of the integrated energy system is obtained. Combined with a pre-built evaluation index library, features are extracted from the real-time operation data to obtain target evaluation features. The evaluation weights corresponding to the target evaluation features are integrated to give the operation evaluation results of the integrated energy system. Based on the operation evaluation results, the target operation model is adjusted to complete the configuration of the integrated energy system.

13. A comprehensive energy configuration system, characterized in that, include: Network device, execution device, and integrated energy configuration device as described in claim 12.

Images