Power transaction compliance method and system with multiple types of market participants

Abstract

The application is suitable for the technical field of power market transaction analysis, and provides a power transaction compliance method and system participated by multiple types of market subjects, which comprises the following steps: constructing a comprehensive relationship graph based on multi-dimensional correlation data of multiple types of market subjects; when a transaction is initiated, extracting a transaction path topology structure, calculating a node comprehensive risk value, and identifying a high-risk mode; generating a three-dimensional risk trajectory graph based on the high-risk mode, and when the risk value breaks through a threshold value, penetratingly scanning and tracing a source and dynamically adjusting a risk weight; and simulating a violation chain reaction based on the updated graph, and generating a dynamic defense strategy. Through comprehensive correlation analysis, dynamic risk assessment and accurate prevention and control, the application improves the comprehensiveness and effectiveness of the power transaction compliance management of multiple types of market subjects.

Description

Technical Field

[0001] This invention belongs to the field of power market transaction analysis technology, and in particular relates to a compliance method and system for power transactions involving multiple types of market participants. Background Technology

[0002] Against the backdrop of electricity market liberalization, electricity trading entities are becoming increasingly diversified, encompassing various market participants such as power generation companies, electricity sales companies, electricity users, and energy storage service providers. Trading models are evolving from traditional single-entity purchase and sale to multi-entity collaborative trading, with the scale and complexity of transactions continuously increasing. In this context, compliance management of electricity trading has become crucial for ensuring fair market order and preventing systemic risks, making the need for multi-entity relationship analysis, dynamic risk monitoring, and violation prevention increasingly urgent.

[0003] Currently, compliance management in the power trading sector largely relies on the qualification review of individual market participants, verification of historical performance data, and static risk scoring. Some solutions collect transaction-related data for risk assessment and implement early warning measures for high-risk participants. These solutions integrate the basic information and transaction records of market participants to form a preliminary risk assessment model, providing fundamental support for transaction compliance.

[0004] However, existing technologies are insufficient to fully capture the complex relationships among various types of market participants and cannot identify high-risk patterns hidden in the network of connections. Risk assessments are mostly static calculations, lacking dynamic analysis of risk transmission paths, resulting in inaccurate tracing of risk sources. Defense strategies are mostly generated based on fixed rules, making it difficult to cope with the chain reaction caused by violations of a single node and unable to adapt to the dynamic risk changes under the interconnectedness of multiple entities in power trading. Summary of the Invention

[0005] The purpose of this invention is to provide a compliance method and system for electricity trading involving multiple types of market participants, aiming to solve the technical problems existing in the prior art as identified in the background section.

[0006] This invention is implemented as follows: a compliance method for electricity trading involving multiple types of market participants, the method comprising:

[0007] Based on multi-dimensional correlation data of various market entities, a comprehensive relationship map covering the relationships of all market entities is constructed;

[0008] When a power transaction is initiated, the transaction path topology structure formed by all market participants involved in the power transaction is extracted from the comprehensive relationship graph, the comprehensive risk value of each node in the transaction path topology structure is calculated, and the transaction path topology structure is simulated and analyzed to identify the existing high-risk patterns.

[0009] A three-dimensional risk trajectory map is generated based on the high-risk model. When the comprehensive risk value of any market entity exceeds the preset dynamic threshold, a penetrating scan is performed based on the three-dimensional risk trajectory map to trace the source of risk. The risk weight coefficient of the risk source in the comprehensive relationship map is dynamically adjusted according to the scan results.

[0010] Based on the updated comprehensive relationship graph, the chain reaction path after a single node violates the rules is simulated, and a dynamic defense strategy is generated simultaneously.

[0011] As a further aspect of the present invention, the construction of a comprehensive relationship map covering the relationships between all market entities specifically includes:

[0012] Collect multi-dimensional correlation data of various market entities, including: basic qualifications of market entities, entity relationships, power transaction performance data, and compliance and regulatory data.

[0013] The collected multi-dimensional correlation data is preprocessed to establish a mapping relationship between each market entity and the corresponding multi-dimensional correlation data.

[0014] Based on the relationships between entities, the strength of the relationships between market entities is analyzed, and combined with multi-dimensional relationship data, a comprehensive relationship graph is constructed with each market entity as a node and the relationships between market entities as edges.

[0015] As a further aspect of the present invention, the calculation of the comprehensive risk value of each node in the transaction path topology specifically includes:

[0016] When a power transaction is initiated, the market entity identifiers of the transaction initiator and the transaction recipient are extracted from the power transaction. Based on the market entity identifiers, the corresponding nodes are located in the comprehensive relationship graph to determine the transaction initiator and the transaction recipient involved in the transaction, which are defined as core market entities.

[0017] Starting with the core market entity, we traverse the other market entities that are related to the core market entity in the comprehensive relationship graph, extract all related market entities and their relationships, and form a transaction path topology structure with transactions as the core.

[0018] Based on power transaction performance data and compliance and regulatory data, the comprehensive risk value of each node in the transaction path topology is calculated.

[0019] As a further aspect of the present invention, the identification of existing high-risk patterns specifically includes:

[0020] The transaction path topology was simulated and analyzed to simulate the transmission effect of risk changes in core nodes on non-core nodes, and the magnitude of change in the comprehensive risk value of each node in the topology under different scenarios was recorded.

[0021] A high-risk pattern library is pre-defined based on the comprehensive risk value of each node. The simulation analysis results are matched with the high-risk pattern library to identify the existing high-risk patterns.

[0022] As a further aspect of the present invention, the dynamic adjustment of the risk weight coefficient of the risk source in the comprehensive relationship graph specifically includes:

[0023] Based on the identified high-risk patterns, nodes with the top 30% comprehensive risk values ​​and ≥3 associated nodes are extracted as risk cluster nodes. For risk cluster nodes, the association paths and risk transmission probabilities between nodes are analyzed to identify the direction of risk transmission.

[0024] The transaction path topology is mapped to a three-dimensional space, the comprehensive risk value of all nodes is mapped to the Z-axis coordinate, and the corresponding risk transmission probability is mapped to the width and color depth of the edge, generating a three-dimensional risk trajectory map containing risk clustering nodes and risk transmission directions.

[0025] Real-time monitoring of the comprehensive risk value of all nodes in the transaction path topology; when the comprehensive risk value of any node exceeds the preset dynamic threshold, the node and its associated path are marked in the three-dimensional risk trajectory map.

[0026] Starting from the node that exceeds the preset dynamic threshold, based on the risk transmission direction shown by the three-dimensional risk trajectory map, a penetrating scan is performed on all related nodes to identify the source of risk;

[0027] For the source of risk, market entities with abnormal connections to the source of risk are identified again through penetrating scanning and marked as abnormal related parties. The weight coefficient of abnormal related parties in the calculation of comprehensive risk value is dynamically adjusted according to the degree of correlation between abnormal related parties and the source of risk.

[0028] Update the dynamically adjusted weight coefficients to the comprehensive relationship graph.

[0029] Another objective of this invention is to provide a compliance system for electricity trading involving multiple types of market participants, the system comprising:

[0030] The comprehensive relationship graph construction module is used to construct a comprehensive relationship graph covering the relationships between all market entities based on multi-dimensional relationship data of various market entities;

[0031] The risk calculation module is used to extract the transaction path topology structure formed by all market participants involved in the power transaction from the comprehensive relationship graph when the power transaction is initiated, calculate the comprehensive risk value of each node in the transaction path topology structure, and perform simulation analysis on the transaction path topology structure to identify the existing high-risk patterns.

[0032] The risk weight dynamic adjustment module is used to generate a three-dimensional risk trajectory map based on the high-risk model. When the comprehensive risk value of any market entity exceeds the preset dynamic threshold, a penetrating scan is performed based on the three-dimensional risk trajectory map to trace the source of risk, and the risk weight coefficient of the risk source in the comprehensive relationship map is dynamically adjusted according to the scan results.

[0033] The defense strategy generation module is used to simulate the chain reaction path after a single node violates a rule based on the updated comprehensive relationship graph, and to generate dynamic defense strategy schemes simultaneously.

[0034] As a further embodiment of the present invention, the comprehensive relationship graph construction module includes:

[0035] The data acquisition unit is used to collect multi-dimensional related data of various market entities. The multi-dimensional related data includes: basic qualifications of market entities, entity relationships, power transaction performance data, and compliance and regulatory data.

[0036] The mapping relationship establishment unit is used to preprocess the collected multi-dimensional related data and establish the mapping relationship between each market entity and the corresponding multi-dimensional related data;

[0037] The graph construction unit is used to analyze the strength of the relationship between market entities based on the subject's relationship, and to construct a comprehensive relationship graph by combining multi-dimensional relationship data, with each market entity as a node and the relationship between market entities as edges.

[0038] As a further embodiment of the present invention, the risk calculation module includes:

[0039] The core market entity positioning unit is used to extract the market entity identifiers of the transaction initiator and the transaction recipient from the power transaction when the power transaction is initiated. Based on the market entity identifiers, the corresponding nodes are located in the comprehensive relationship graph to determine the transaction initiator and the transaction recipient involved in the transaction, and are defined as core market entities.

[0040] The transaction path topology generation unit is used to start from the core market entity, traverse other market entities that are related to the core market entity in the comprehensive relationship graph, extract all related market entities and their relationships, and form a transaction path topology with transactions as the core.

[0041] The risk value calculation unit is used to calculate the comprehensive risk value of each node in the transaction path topology based on power transaction performance data and compliance and regulatory data.

[0042] As a further embodiment of the present invention, the risk calculation module further includes:

[0043] The transaction path simulation analysis unit is used to simulate and analyze the transaction path topology, simulate the transmission impact of risk changes in core nodes on non-core nodes, and record the magnitude of changes in the comprehensive risk value of each node in the topology under different scenarios.

[0044] The high-risk pattern recognition unit is used to pre-set a high-risk pattern library based on the comprehensive risk value of each node, match the simulation analysis results with the high-risk pattern library, and identify the existing high-risk patterns.

[0045] As a further embodiment of the present invention, the risk weight dynamic adjustment module includes:

[0046] The risk cluster node identification unit is used to extract the top 30% of the comprehensive risk values ​​based on the identified high-risk patterns, and the number of associated nodes is ≥3 as risk cluster nodes. For risk cluster nodes, the association paths between nodes and the probability of risk transmission are analyzed to identify the direction of risk transmission.

[0047] The risk trajectory graph generation unit is used to map the transaction path topology to a three-dimensional space, map the comprehensive risk value of all nodes to the Z-axis coordinate, and map the corresponding risk transmission probability to the width and color depth of the edge, generating a three-dimensional risk trajectory graph that includes risk clustering nodes and risk transmission direction.

[0048] The real-time risk monitoring unit is used to monitor the comprehensive risk value of all nodes in the transaction path topology in real time. When the comprehensive risk value of any node exceeds the preset dynamic threshold, the node and its associated path are marked in the three-dimensional risk trajectory map.

[0049] The risk source identification unit is used to perform a penetrating scan of all related nodes based on the risk transmission direction displayed by the three-dimensional risk trajectory map, starting from the node that exceeds the preset dynamic threshold, and to identify the risk source.

[0050] The weighting adjustment unit is used to identify market entities with abnormal connections to the risk source again through penetrating scanning and mark them as abnormal related parties. Based on the degree of correlation between the abnormal related parties and the risk source, the weighting coefficient of the abnormal related parties in the comprehensive risk value calculation is dynamically adjusted.

[0051] The graph update unit is used to update the dynamically adjusted weight coefficients to the comprehensive relationship graph.

[0052] The beneficial effects of this invention are:

[0053] This invention constructs a comprehensive relationship graph encompassing all market participants, achieving a complete characterization of the complex relationships among various types of market participants. This breaks down information silos and provides a panoramic perspective for compliance analysis. Through transaction path topology extraction, dynamic risk value calculation, and high-risk pattern identification, it can accurately focus on the risk status of transaction-related entities, proactively discovering potential risk clusters, closed-loop transactions, and other high-risk patterns, thus improving the targeting and foresight of risk identification. The combination of a three-dimensional risk trajectory map and penetrating scanning enables rapid tracing of risk sources, while dynamic adjustment of risk weights allows the comprehensive relationship graph to adapt to risk changes in real time, ensuring the accuracy of risk assessment. Based on the updated graph, a chain reaction is simulated to generate dynamic defense strategies, achieving closed-loop management from risk identification to prevention and control. This effectively blocks risk transmission, avoids systemic violations, and comprehensively improves the efficiency and reliability of compliance management for power transactions involving multiple types of market participants. Attached Figure Description

[0054] Figure 1 A flowchart illustrating a compliance method for electricity trading involving multiple types of market participants, as provided in an embodiment of the present invention.

[0055] Figure 2 A flowchart for constructing a comprehensive relationship map covering the relationships between all market entities is provided for embodiments of the present invention;

[0056] Figure 3 A flowchart for calculating the comprehensive risk value of each node in the transaction path topology provided in an embodiment of the present invention;

[0057] Figure 4 A flowchart for identifying high-risk patterns provided in this embodiment of the invention;

[0058] Figure 5 A flowchart for dynamically adjusting the risk weight coefficients of risk sources in a comprehensive relationship graph provided in this embodiment of the invention;

[0059] Figure 6 A structural block diagram of a power trading compliance system involving multiple types of market participants, provided in an embodiment of the present invention;

[0060] Figure 7 This is a structural block diagram of the integrated relationship graph construction module provided in an embodiment of the present invention;

[0061] Figure 8 This is a structural block diagram of the risk calculation module provided in an embodiment of the present invention;

[0062] Figure 9 This is a structural block diagram of the risk weight dynamic adjustment module provided in an embodiment of the present invention. Detailed Implementation

[0063] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0064] Figure 1 A flowchart of a compliance method for electricity trading involving multiple types of market participants, as provided in an embodiment of the present invention, is shown below. Figure 1 As shown, the method includes:

[0065] S100 constructs a comprehensive relationship map covering the relationships between all market entities based on multi-dimensional correlation data of various market entities;

[0066] The system collects multi-dimensional data on various market participants. This data not only covers the basic qualification information of market participants, such as business licenses and industry qualifications, to ensure that the participants in the transaction have legal identities and business capabilities, but also includes data on the relationships between participants, such as equity relationships and cooperation history, to reveal potential conflicts of interest. At the same time, power transaction performance data can reflect the historical transaction performance of each participant, while compliance and regulatory data records past violations or regulatory focus areas.

[0067] After data collection, preprocessing is performed to establish a precise mapping relationship between each market entity and its corresponding data. Subsequently, based on the entity relationships, the strength of the relationships between market entities is quantitatively analyzed. Each market entity is treated as a node, and the relationships between entities are treated as edges. A comprehensive relationship graph is constructed by combining multi-dimensional data. The analysis of relationship strength can distinguish the closeness of business connections between different entities. For example, entities with long-term stable cooperation have a higher relationship strength, while entities with occasional transactions have a lower relationship strength. This quantitative processing allows the graph to not only show the surface transaction relationships but also uncover deeper potential connections.

[0068] like Figure 2 As shown, the construction of a comprehensive relationship map covering the relationships between all market entities specifically includes:

[0069] S110, collect multi-dimensional correlation data of various market entities, including: basic qualifications of market entities, entity relationships, power transaction performance data, and compliance and regulatory data;

[0070] S120, preprocess the collected multi-dimensional related data, and establish a mapping relationship between each market entity and the corresponding multi-dimensional related data;

[0071] S130 analyzes the strength of relationships between market entities based on their relationships with each entity, and combines multi-dimensional relationship data to construct a comprehensive relationship graph with each market entity as a node and the relationships between market entities as edges.

[0072] Among them, the correlation strength is used to quantify the degree of business connection between two market entities:

[0073] ;

[0074] For market entities and The correlation strength between them, with a value ranging from 0 to 1;

[0075] The transaction frequency factor is calculated as the number of transactions between the two parties in the past 12 months divided by the industry average number of transactions.

[0076] The transaction amount factor is calculated as the sum of the cumulative transaction amount over the past 12 months and the total transaction amount between both parties.

[0077] Business dependence factors refer to the degree to which one party depends on the other party's business, such as the proportion of procurement volume.

[0078] These are the weighting coefficients.

[0079] S200 extracts the transaction path topology structure formed by all market participants involved in the power transaction from the comprehensive relationship graph when the power transaction is initiated, calculates the comprehensive risk value of each node in the transaction path topology structure, and performs simulation analysis on the transaction path topology structure to identify the existing high-risk patterns.

[0080] When initiating an electricity transaction, the first step is to extract the market entity identifiers of the initiator and receiver from the transaction information. Based on these identifiers, the corresponding nodes are located in the constructed comprehensive relationship graph, and these two entities are defined as core market entities. The purpose of this step is to identify the core participants in the transaction, using them as anchors to narrow the scope of analysis, avoid interference from irrelevant information, and ensure the focus of subsequent analysis. Subsequently, starting from the core market entities, other market entities related to them in the comprehensive relationship graph are traversed, extracting all related entities and relationships to form a transaction path topology centered on this transaction. This topology construction focuses on entities and relationships that have a substantial impact on the current transaction by setting association strength thresholds and excluding closed-loop paths, forming a clear, transaction-centric star-shaped expansion structure. Essentially, it slices the complex overall relationship graph, retaining only the local networks closely related to the current transaction, making risk analysis more focused and efficient.

[0081] Based on this, and using power trading performance data and compliance and regulatory data, the comprehensive risk value of each node in the trading path topology is calculated. The comprehensive risk value integrates the entity's qualification and compliance, historical performance rate, history of violations, and related transmission risk (i.e., the degree to which it is affected by the risks of other related entities). Through the weighted fusion of multi-dimensional data, the risk level of each market entity in the current trading scenario is quantified, taking into account both the entity's own risk and the risk transmission from related entities, thus better reflecting the actual situation of mutual influence between entities in power trading.

[0082] Next, the constructed transaction path topology is simulated and analyzed to simulate the transmission impact of risk changes in core nodes on non-core nodes. The magnitude of risk value changes for each node under different scenarios is recorded. The simulation results are then matched with a pre-set high-risk pattern library to identify potential high-risk patterns in the current transaction path. By proactively simulating the transmission chain under different risk scenarios through simulation analysis, potential risk correlation patterns that are difficult to reveal in static analysis can be discovered in advance.

[0083] like Figure 3 As shown, the calculation of the comprehensive risk value of each node in the transaction path topology specifically includes:

[0084] S210: When a power transaction is initiated, the market entity identifiers of the transaction initiator and the transaction recipient are extracted from the power transaction. Based on the market entity identifiers, the corresponding nodes are located in the comprehensive relationship graph to determine the transaction initiator and the transaction recipient involved in the transaction, and they are defined as core market entities.

[0085] S220 starts with the core market entity, traverses other market entities that are related to the core market entity in the comprehensive relationship graph, extracts all related market entities and their relationships, and forms a transaction path topology structure with transactions as the core.

[0086] The formation process of the transaction path topology is as follows:

[0087] 1. Identify the core market players in the transaction, namely, the initiator A and the receiver B;

[0088] 2. Centered on core market entities, traverse the comprehensive relationship map. All associated nodes;

[0089] 3. Extract the valid paths between related nodes, excluding closed-loop paths and weakly related paths;

[0090] 4. Construct a star topology extension structure centered on transactions, as the transaction path topology, represented as follows: adjacency matrix :

[0091] ;

[0092] The total number of market entities involved in the transaction path;

[0093] , indicating the subject and The strength of the association;

[0094] Core market entities are marked as and That is, the initiator and the recipient of the transaction;

[0095] S230 calculates the comprehensive risk value of each node in the transaction path topology based on power transaction performance data and compliance regulatory data.

[0096] The composite risk score is used to assess the compliance risk level of an individual market participant in a transaction:

[0097] ;

[0098] For market entities The overall risk value ranges from 0 to 100, with higher values ​​indicating greater risk.

[0099] For qualification compliance, the value ranges from 0 to 1, and is a score based on the compliance status of business licenses, qualification certificates, etc.

[0100] This represents the historical fulfillment rate, with a value ranging from 0 to 1, which indicates the historical transaction fulfillment rate.

[0101] This is a violation history factor, with a value ranging from 0 to 1, calculated based on the number and severity of historical violations.

[0102] This refers to the risk transmission through related entities, specifically the risk transmission value among related parties.

[0103] These are the weighting coefficients.

[0104] like Figure 4 As shown, the identification of existing high-risk patterns specifically includes:

[0105] S240 performs simulation analysis on the transaction path topology, simulates the transmission impact of risk changes in core nodes on non-core nodes, and records the variation range of the comprehensive risk value of each node in the topology under different scenarios.

[0106] Risk transmission impact indicates the degree to which changes in risk at core nodes affect non-core nodes:

[0107] ;

[0108] The change in risk value for non-core node j;

[0109] This is the risk transmission coefficient, with a value ranging from 0 to 1, set according to the business type.

[0110] core node and non-core nodes The strength of the association;

[0111] The initial risk value for core node i;

[0112] The risk isolation coefficient, ranging from 0 to 1, reflects the effectiveness of the risk prevention and control mechanisms of both parties.

[0113] Specifically: First, set the risk increase value for the core nodes, and then calculate it for each directly related node. It calculates the risk changes of indirectly related nodes layer by layer, and finally records the risk transmission results under different scenarios.

[0114] S250 uses a high-risk pattern library pre-set based on the comprehensive risk value of each node. It matches the simulation analysis results with the high-risk pattern library to identify the existing high-risk patterns.

[0115] For high-risk patterns, including:

[0116] 1. Closed-loop transaction model:

[0117] This creates a false cycle of funds or electricity, which can easily lead to idle transactions, i.e., financial settlement is completed without actual physical electricity delivery, thus circumventing electricity price supervision or artificially inflating transaction volume.

[0118] A closed-loop structure can conceal actual control relationships, facilitate the transfer of benefits to related parties, and violate the principle of fair competition in the electricity market.

[0119] A closed loop involving multiple entities can amplify the impact of violations at a single node, increasing the difficulty of regulatory traceability.

[0120] 2. Risk clustering pattern (a single node is associated with 3 or more high-risk nodes):

[0121] High-risk nodes ( These entities typically have qualification deficiencies, poor performance records, or a history of violations. The convergence of multiple high-risk nodes can create a risk resonance.

[0122] Electricity trading is highly interconnected; a default at one node can trigger a chain of defaults through related relationships, leading to regional trading disruptions.

[0123] Clustering patterns may indicate collusion to deliberately evade regulation, increasing the risk of collective violations.

[0124] 3. Hierarchical transmission model (risk is transmitted through three or more levels of related paths):

[0125] Long-path transmission can obscure the source of risk and delay intervention.

[0126] Each level of transmission amplifies the risk for related parties, potentially resulting in a risk that is 2-3 times greater than the initial value.

[0127] The electricity market involves multiple links such as power generation, transmission, distribution, and sales. Long-path transmission may spread risks across links, affecting the stable operation of the power system.

[0128] 4. Sudden Growth Pattern (Node Risk Value Increases by More Than 50% Within 24 Hours):

[0129] A sudden increase in risk value usually corresponds to sudden violations or abnormal transaction data.

[0130] Electricity trading is planned, and sudden changes in risk in a short period of time may trigger market panic, leading to electricity price fluctuations or chaos in the trading order.

[0131] An increase of more than 50% often means that conventional risk control mechanisms have failed, and emergency intervention is needed to prevent the risk from spreading.

[0132] 5. Abnormal Related Patterns (Entities with no historical transaction records suddenly generate large transactions):

[0133] Large transactions with no historical record may involve fictitious entities with the aim of manipulating market prices or transferring assets.

[0134] Newly associated entities lack performance credit data, and their probability of default is 3-5 times that of regular entities, which can easily lead to contract disputes.

[0135] Electricity transactions must be matched with the physical grid's transmission capacity; abnormally large transactions may lead to grid security issues such as power flow exceeding limits.

[0136] S300, a three-dimensional risk trajectory map is generated based on the high-risk model. When the comprehensive risk value of any market entity exceeds the preset dynamic threshold, a penetrating scan is performed based on the three-dimensional risk trajectory map to trace the source of risk, and the risk weight coefficient of the risk source in the comprehensive relationship map is dynamically adjusted according to the scan results.

[0137] Based on identified high-risk patterns, risk clusters with high overall risk values ​​and numerous associated nodes are extracted from the transaction path topology. This focuses on areas of high risk concentration, as these nodes are often key hubs for risk transmission. Analyzing their associated paths and transmission probabilities clarifies the flow of risk among entities, providing clues for subsequent tracing. The transaction path topology is then mapped to three-dimensional space, converting the overall risk value of each node into Z-axis coordinates. The width of the edges and color depth represent the risk transmission probability, generating a three-dimensional risk trajectory map. This visualization breaks the limitations of traditional data tables, making complex risk associations and transmission trends intuitive and perceptible. It facilitates rapid understanding of the spatial distribution and diffusion paths of risks, solving the problem of difficulty in visualizing risk relationships among multiple entities.

[0138] In the real-time monitoring phase, the system continuously tracks the comprehensive risk value of all nodes. When any node exceeds a preset dynamic threshold, the node and its associated paths are immediately marked on the 3D graph. The dynamic threshold is dynamically adjusted based on industry averages, risk volatility, and the strength of entity associations. This ensures that the threshold adapts to the risk characteristics of different types of entities while reflecting changes in the overall market risk environment, avoiding underreporting or false alarms caused by a one-size-fits-all monitoring standard. Once a threshold is exceeded, a penetrating scan is performed starting from that node, tracing the associated nodes layer by layer according to the risk transmission direction shown in the 3D graph. The source of risk is identified by calculating risk contribution ratios, gradually delving from the surface-level risk manifestation nodes to the initial risk trigger point. This solves the problem of traditional risk screening only focusing on surface phenomena and failing to address the root cause. After finding the source of risk, its abnormal related parties are further scanned. The weight coefficients of these entities in the comprehensive risk value calculation are adjusted according to the degree of association, and then updated to the comprehensive relationship graph. This step is to allow the risk assessment system to dynamically adapt to the current risk association status, because abnormal associations often hide potential risks of regulatory evasion. By increasing their weight, subsequent risk calculations can more sensitively capture such hidden dangers.

[0139] like Figure 5 As shown, the dynamic adjustment of the risk weight coefficients of risk sources in the comprehensive relationship graph specifically includes:

[0140] S310: Based on the identified high-risk patterns, extract the top 30% of nodes with the highest comprehensive risk value and at least 3 associated nodes as risk cluster nodes. For risk cluster nodes, analyze the association paths and risk transmission probabilities between nodes and identify the direction of risk transmission.

[0141] Analysis of associated paths:

[0142] Starting from the risk cluster node, traverse all associated nodes and record all valid paths with a path length ≤ 5 and an association strength ≥ 0.3;

[0143] Duplicate paths are merged, and the shortest and strongest associated paths are retained. Based on the path length and cumulative association strength, a path importance ranking is generated, and associated paths can be identified based on the path importance.

[0144] The formula for calculating the probability of risk transmission is:

[0145] ;

[0146] in, Risk from the main body Conducted to The probability of;

[0147] as the main body The risk defense capability, namely, the score based on the completeness of the risk control system;

[0148] as the main body Overall correlation strength with other entities;

[0149] For identifying the direction of risk transmission:

[0150] 1. Calculate the sum of all node pairs on the path. and ;

[0151] 2. If Then the direction of risk transmission is determined to be ;

[0152] 3. Construct a directed graph to mark all identified risk transmission directions;

[0153] 4. Nodes with similar bidirectional conduction probabilities (difference < 0.1) are marked as bidirectional conduction nodes.

[0154] S320 maps the transaction path topology to a three-dimensional space, maps the comprehensive risk value of all nodes to the Z-axis coordinate, and maps the corresponding risk transmission probability to the width and color depth of the edge, generating a three-dimensional risk trajectory map containing risk clustering nodes and risk transmission directions.

[0155] S330 monitors the comprehensive risk value of all nodes in the transaction path topology in real time. When the comprehensive risk value of any node exceeds the preset dynamic threshold, the node and its associated path are marked in the three-dimensional risk trajectory map.

[0156] The dynamic threshold is adjusted in real time according to the following rules:

[0157] ;

[0158] as the main body Dynamic risk threshold; This represents the industry average risk value. The industry risk standard deviation; For safety, the default value is 1.5; The correlation adjustment coefficient is set to 0.1 by default. The more entities associated, the lower the dynamic threshold. as the main body The overall correlation strength;

[0159] S340, starting from the node that exceeds the preset dynamic threshold, performs a penetrating scan on all related nodes based on the risk transmission direction displayed by the three-dimensional risk trajectory map to identify the source of risk;

[0160] Specifically:

[0161] 1. When node of At that time, initiate a penetration scan;

[0162] 2. Inspection and Nodes All directly related nodes ,calculate From node contribution ratio ;

[0163] 3. Contribution ratio >20% of nodes Repeat step 2 to perform the previous level scan;

[0164] 4. A source of risk is identified when the following conditions are met:

[0165] Scan path length ≥ 2 levels;

[0166] Risk value of this node ;

[0167] The cumulative contribution to downstream nodes is >50%;

[0168] 5. Examine the historical violation records and real-time transaction data of the source node to confirm the reasonableness of the risk;

[0169] 6. Generate a risk source map, marking key transmission paths and their impact weights.

[0170] S350, for the source of risk, further penetration scanning is used to identify market entities with abnormal connections to the source of risk and mark them as abnormal related parties. Based on the degree of correlation between the abnormal related parties and the source of risk, the weighting coefficient of the abnormal related parties in the calculation of the comprehensive risk value is dynamically adjusted.

[0171] ;

[0172] in, These are the adjusted weighting coefficients, corresponding to the adjusted... ;

[0173] The original weight coefficients, i.e., the corresponding ;

[0174] For adjustment factors;

[0175] The risk contribution ratio of risk source s to abnormal related party j;

[0176] The excess risk value at the source of risk;

[0177] As the source of risk The dynamic threshold.

[0178] S360 updates the dynamically adjusted weight coefficients to the comprehensive relationship graph.

[0179] S400, based on the updated integrated relationship graph, simulates the chain reaction path after a single node violates a rule, and simultaneously generates a dynamic defense strategy.

[0180] After the S300 steps of dynamic adjustment of risk weights, the comprehensive relationship graph can more accurately reflect the actual risk association status of various market entities, including the influence of the risk source and the potential risks of abnormal related parties. At this point, simulating the chain reaction path after a single node's violation is essentially a stress test of the risk transmission chain among multiple types of market entities. It observes how the risk spreads to upstream and downstream or related parties through associations when a certain entity violates the rules. For example, a power generation company's default may affect electricity sales companies that rely on its power supply, thereby affecting industrial users, or be transmitted to other business segments through equity relationships. This simulation needs to combine the association strength of each entity, the probability of risk transmission, and the adjusted weight coefficients to ensure the authenticity of the path extrapolation and avoid prediction bias caused by static data.

[0181] While simulating the chain reaction path, the system will simultaneously generate dynamic defense strategy plans. If the simulation shows that the risk of violation is mainly transmitted rapidly through a few strongly related paths, the strategy may include temporarily blocking the transaction permissions of these paths, requiring relevant entities to supplement performance guarantees, or activating backup transaction channels to isolate the risk; if the risk spread involves multiple types of entities across different stages, the strategy may link the regulatory mechanisms of different stages and coordinate the resources of all parties for collaborative prevention and control.

[0182] Figure 6This is a structural block diagram of a power trading compliance system with multiple types of market participants provided in an embodiment of the present invention, such as... Figure 6 As shown, the system includes:

[0183] The comprehensive relationship graph construction module 100 is used to construct a comprehensive relationship graph covering the relationships of all market entities based on multi-dimensional relationship data of various market entities;

[0184] The risk calculation module 200 is used to extract the transaction path topology structure formed by all market participants involved in the power transaction from the comprehensive relationship graph when the power transaction is initiated, calculate the comprehensive risk value of each node in the transaction path topology structure, and perform simulation analysis on the transaction path topology structure to identify the existing high-risk patterns.

[0185] The risk weight dynamic adjustment module 300 is used to generate a three-dimensional risk trajectory map based on the high-risk mode. When the comprehensive risk value of any market entity exceeds the preset dynamic threshold, a penetrating scan is performed based on the three-dimensional risk trajectory map to trace the source of risk, and the risk weight coefficient of the risk source in the comprehensive relationship map is dynamically adjusted according to the scan results.

[0186] The defense strategy generation module 400 is used to simulate the chain reaction path after a single node violates a rule based on the updated comprehensive relationship graph, and to generate a dynamic defense strategy scheme simultaneously.

[0187] like Figure 7 As shown, the comprehensive relationship graph construction module 100 includes:

[0188] Data acquisition unit 110 is used to collect multi-dimensional related data of various market entities. The multi-dimensional related data includes: basic qualifications of market entities, entity relationships, power transaction performance data, and compliance and supervision data.

[0189] The mapping relationship establishment unit 120 is used to preprocess the collected multi-dimensional related data and establish the mapping relationship between each market entity and the corresponding multi-dimensional related data;

[0190] The graph construction unit 130 is used to analyze the correlation strength between market entities based on the subject correlation relationship, and to construct a comprehensive relationship graph by combining multi-dimensional correlation data, with each market entity as a node and the correlation relationship between market entities as edges.

[0191] like Figure 8 As shown, the risk calculation module 200 includes:

[0192] The core market entity positioning unit 210 is used to extract the market entity identifiers of the transaction initiator and the transaction recipient from the power transaction when the power transaction is initiated, locate the corresponding nodes in the comprehensive relationship graph based on the market entity identifiers, determine the transaction initiator and the transaction recipient involved in the transaction, and define them as core market entities.

[0193] The transaction path topology generation unit 220 is used to start from the core market entity, traverse other market entities that are related to the core market entity in the comprehensive relationship graph, extract all related market entities and their relationships, and form a transaction path topology with transactions as the core.

[0194] The risk value calculation unit 230 is used to calculate the comprehensive risk value of each node in the transaction path topology based on power transaction performance data and compliance and regulatory data.

[0195] like Figure 8 As shown, the risk calculation module 200 further includes:

[0196] The transaction path simulation analysis unit 240 is used to simulate and analyze the transaction path topology, simulate the transmission effect of risk changes in core nodes on non-core nodes, and record the change range of the comprehensive risk value of each node in the topology under different scenarios.

[0197] The high-risk pattern recognition unit 250 is used to pre-set a high-risk pattern library based on the comprehensive risk value of each node, match the simulation analysis results with the high-risk pattern library, and identify the existing high-risk patterns.

[0198] like Figure 9 As shown, the risk weight dynamic adjustment module 300 includes:

[0199] The risk cluster node identification unit 310 is used to extract the top 30% of the comprehensive risk values ​​based on the identified high-risk patterns, and the number of associated nodes is ≥3 as risk cluster nodes. For risk cluster nodes, the association path and risk transmission probability between nodes are analyzed, and the direction of risk transmission is identified.

[0200] The risk trajectory graph generation unit 320 is used to map the transaction path topology to a three-dimensional space, map the comprehensive risk value of all nodes to the Z-axis coordinate, and map the corresponding risk transmission probability to the width and color depth of the edge, thereby generating a three-dimensional risk trajectory graph containing risk cluster nodes and risk transmission direction.

[0201] The real-time risk monitoring unit 330 is used to monitor the comprehensive risk value of all nodes in the transaction path topology in real time. When the comprehensive risk value of any node exceeds the preset dynamic threshold, the node and its associated path are marked in the three-dimensional risk trajectory map.

[0202] The risk source identification unit 340 is used to perform a penetrating scan on all related nodes based on the risk transmission direction displayed by the three-dimensional risk trajectory map, starting from the node that exceeds the preset dynamic threshold, and to identify the risk source.

[0203] The weight adjustment unit 350 is used to identify market entities with abnormal connections to the risk source again through penetrating scanning and mark them as abnormal related parties. Based on the degree of correlation between the abnormal related parties and the risk source, the weight coefficient of the abnormal related parties in the comprehensive risk value calculation is dynamically adjusted.

[0204] The graph update unit 360 is used to update the dynamically adjusted weight coefficients to the comprehensive relationship graph.

[0205] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0206] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

[0207] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A compliance approach for electricity trading involving multiple types of market participants, characterized in that: The method includes: Based on multi-dimensional correlation data of various market entities, a comprehensive relationship map covering the relationships of all market entities is constructed; When a power transaction is initiated, the transaction path topology structure formed by all market participants involved in the power transaction is extracted from the comprehensive relationship graph, the comprehensive risk value of each node in the transaction path topology structure is calculated, and the transaction path topology structure is simulated and analyzed to identify the existing high-risk patterns. The composite risk score is used to assess the compliance risk level of an individual market participant in a transaction: ； For market entities The overall risk value ranges from 0 to 100, with higher values ​​indicating greater risk. For qualification compliance, the value ranges from 0 to 1, and is a score based on the compliance status of business licenses, qualification certificates, etc. This represents the historical fulfillment rate, with a value ranging from 0 to 1, which indicates the historical transaction fulfillment rate. This is a violation history factor, with a value ranging from 0 to 1, calculated based on the number and severity of historical violations. This refers to the risk transmission through related entities, specifically the risk transmission value among related parties. These are the weighting coefficients; A three-dimensional risk trajectory map is generated based on the high-risk model. When the comprehensive risk value of any market entity exceeds the preset dynamic threshold, a penetrating scan is performed based on the three-dimensional risk trajectory map to trace the source of risk. The risk weight coefficient of the risk source in the comprehensive relationship map is dynamically adjusted according to the scan results. Based on the updated comprehensive relationship graph, the chain reaction path after a single node violates the rules is simulated, and a dynamic defense strategy is generated simultaneously. The calculation of the comprehensive risk value of each node in the transaction path topology specifically includes: When a power transaction is initiated, the market entity identifiers of the transaction initiator and the transaction recipient are extracted from the power transaction. Based on the market entity identifiers, the corresponding nodes are located in the comprehensive relationship graph to determine the transaction initiator and the transaction recipient involved in the transaction, which are defined as core market entities. Starting with the core market entity, we traverse other market entities that are related to the core market entity in the comprehensive relationship graph, extract all related market entities and their relationships, and form a transaction path topology structure with transactions as the core. Based on power transaction performance data and compliance regulatory data, the comprehensive risk value of each node in the transaction path topology is calculated. The identified high-risk patterns specifically include: The transaction path topology was simulated and analyzed to simulate the transmission effect of risk changes in core nodes on non-core nodes, and the magnitude of change in the comprehensive risk value of each node in the topology under different scenarios was recorded. A high-risk pattern library is pre-set based on the comprehensive risk value of each node. The simulation analysis results are matched with the high-risk pattern library to identify the existing high-risk patterns. The dynamic adjustment of the risk weight coefficients for risk sources in the comprehensive relationship graph specifically includes: Based on the identified high-risk patterns, nodes with the top 30% comprehensive risk values ​​and ≥3 associated nodes are extracted as risk cluster nodes. For risk cluster nodes, the association paths and risk transmission probabilities between nodes are analyzed to identify the direction of risk transmission. The transaction path topology is mapped to a three-dimensional space, the comprehensive risk value of all nodes is mapped to the Z-axis coordinate, and the corresponding risk transmission probability is mapped to the width and color depth of the edge, generating a three-dimensional risk trajectory map containing risk clustering nodes and risk transmission directions. Real-time monitoring of the comprehensive risk value of all nodes in the transaction path topology; when the comprehensive risk value of any node exceeds the preset dynamic threshold, the node and its associated path are marked in the three-dimensional risk trajectory map. Starting from the node that exceeds the preset dynamic threshold, based on the risk transmission direction shown by the three-dimensional risk trajectory map, a penetrating scan is performed on all related nodes to identify the source of risk; For the source of risk, market entities with abnormal connections to the source of risk are identified again through penetrating scanning and marked as abnormal related parties. The weight coefficient of abnormal related parties in the calculation of comprehensive risk value is dynamically adjusted according to the degree of correlation between abnormal related parties and the source of risk. Update the dynamically adjusted weight coefficients to the comprehensive relationship graph.

2. The method according to claim 1, characterized in that, The construction of a comprehensive relationship map covering the connections between all market entities specifically includes: Collect multi-dimensional correlation data of various market entities, including: basic qualifications of market entities, entity relationships, power transaction performance data, and compliance and regulatory data. The collected multi-dimensional correlation data is preprocessed to establish a mapping relationship between each market entity and the corresponding multi-dimensional correlation data. Based on the relationships between entities, the strength of the relationships between market entities is analyzed, and combined with multi-dimensional relationship data, a comprehensive relationship graph is constructed with each market entity as a node and the relationships between market entities as edges.

3. A power trading compliance system involving multiple types of market participants, used to execute the power trading compliance method involving multiple types of market participants as described in claim 1, characterized in that, The system includes: The comprehensive relationship graph construction module is used to construct a comprehensive relationship graph covering the relationships between all market entities based on multi-dimensional relationship data of various market entities; The risk calculation module is used to extract the transaction path topology structure formed by all market participants involved in the power transaction from the comprehensive relationship graph when the power transaction is initiated, calculate the comprehensive risk value of each node in the transaction path topology structure, and perform simulation analysis on the transaction path topology structure to identify the existing high-risk patterns. The risk weight dynamic adjustment module is used to generate a three-dimensional risk trajectory map based on the high-risk model. When the comprehensive risk value of any market entity exceeds the preset dynamic threshold, a penetrating scan is performed based on the three-dimensional risk trajectory map to trace the source of risk, and the risk weight coefficient of the risk source in the comprehensive relationship map is dynamically adjusted according to the scan results. The defense strategy generation module is used to simulate the chain reaction path after a single node violates a rule based on the updated comprehensive relationship graph, and to generate dynamic defense strategy schemes simultaneously. The risk calculation module includes: The core market entity positioning unit is used to extract the market entity identifiers of the transaction initiator and the transaction recipient from the power transaction when the power transaction is initiated. Based on the market entity identifiers, the corresponding nodes are located in the comprehensive relationship graph to determine the transaction initiator and the transaction recipient involved in the transaction, and are defined as core market entities. The transaction path topology generation unit is used to start from the core market entity, traverse other market entities that are related to the core market entity in the comprehensive relationship graph, extract all related market entities and their relationships, and form a transaction path topology with transactions as the core. The risk value calculation unit is used to calculate the comprehensive risk value of each node in the transaction path topology based on power transaction performance data and compliance regulatory data. The risk calculation module also includes: The transaction path simulation analysis unit is used to simulate and analyze the transaction path topology, simulate the transmission impact of risk changes in core nodes on non-core nodes, and record the change range of the comprehensive risk value of each node in the topology under different scenarios. The high-risk pattern recognition unit is used to pre-set a high-risk pattern library based on the comprehensive risk value of each node, match the simulation analysis results with the high-risk pattern library, and identify the existing high-risk patterns.

4. The system according to claim 3, characterized in that, The comprehensive relationship graph construction module includes: The data acquisition unit is used to collect multi-dimensional related data of various market entities. The multi-dimensional related data includes: basic qualifications of market entities, entity relationships, power transaction performance data, and compliance and regulatory data. The mapping relationship establishment unit is used to preprocess the collected multi-dimensional related data and establish the mapping relationship between each market entity and the corresponding multi-dimensional related data; The graph construction unit is used to analyze the strength of the relationship between market entities based on the subject's relationship, and to construct a comprehensive relationship graph by combining multi-dimensional relationship data, with each market entity as a node and the relationship between market entities as edges.

5. The system according to claim 3, characterized in that, The risk weight dynamic adjustment module includes: The risk cluster node identification unit is used to extract the top 30% of the comprehensive risk values ​​based on the identified high-risk patterns, and the number of associated nodes is ≥3 as risk cluster nodes. For risk cluster nodes, the association paths between nodes and the probability of risk transmission are analyzed to identify the direction of risk transmission. The risk trajectory graph generation unit is used to map the transaction path topology to a three-dimensional space, map the comprehensive risk value of all nodes to the Z-axis coordinate, and map the corresponding risk transmission probability to the width and color depth of the edge, generating a three-dimensional risk trajectory graph that includes risk clustering nodes and risk transmission direction. The real-time risk monitoring unit is used to monitor the comprehensive risk value of all nodes in the transaction path topology in real time. When the comprehensive risk value of any node exceeds the preset dynamic threshold, the node and its associated path are marked in the three-dimensional risk trajectory map. The risk source identification unit is used to perform a penetrating scan of all related nodes based on the risk transmission direction displayed by the three-dimensional risk trajectory map, starting from the node that exceeds the preset dynamic threshold, and to identify the risk source. The weighting adjustment unit is used to identify market entities with abnormal connections to the risk source through a penetrating scan and mark them as abnormal related parties. Based on the degree of correlation between the abnormal related parties and the risk source, the weighting coefficient of the abnormal related parties in the comprehensive risk value calculation is dynamically adjusted. The graph update unit is used to update the dynamically adjusted weight coefficients to the comprehensive relationship graph.

Images