Multi-source new energy power station three-dimensional collaborative design and green degree evaluation method and system

Abstract

This invention discloses a method and system for three-dimensional collaborative design and greenness assessment of multi-source new energy power plants, belonging to the field of new energy power plant design technology. This invention establishes a unified cloud-based three-dimensional collaborative design platform, integrating five major professional design resources to achieve multi-professional synchronous modeling, parameter linkage updates, and cross-professional automatic conflict verification. It constructs a fully integrated three-dimensional digital model, and based on this model, builds five types of simulation sub-engines. Relying on a built-in data bus, it achieves data interoperability, linkage triggering, and collaborative iteration among multiple engines. Simultaneously, it constructs a comprehensive greenness assessment model, establishes a reverse mapping rule library between evaluation indicators and three-dimensional design parameters, and uses quantitative assessment results to reverse-correct design parameters, forming a closed-loop process encompassing three-dimensional collaborative design, simulation coupling analysis, greenness assessment, and iterative optimization. This solves the industry shortcomings of existing technologies, such as poor design collaboration, weak simulation linkage, lagging greenness assessment, and the inability to achieve closed-loop optimization.

Description

Technical Field

[0001] This invention relates to the field of new energy power plant design technology, specifically to a method and system for three-dimensional collaborative design and greenness assessment of multi-source new energy power plants. Background Technology

[0002] With the accelerated construction of new power systems based on new energy sources, multi-source new energy power plants have become an important development direction. These power plants involve numerous disciplines, have complex spatial layouts, and high design requirements. At the same time, with the popularization of BIM, GIS, and UAV laser mapping technologies, three-dimensional digital design is gradually being applied to the design of new energy power plants. There is an urgent need to build a set of three-dimensional collaborative design and greenness assessment technology solutions that are fully coupled, quantifiable, and iterative.

[0003] Existing technologies, such as the wind-solar-hydrogen multi-energy complementary system collaborative operation optimization and comprehensive benefit evaluation method disclosed in application CN120320391B, construct a hybrid power flow model coupling electricity, hydrogen, and heat, with the whole life cycle cost and new energy consumption as optimization objectives, and optimize equipment capacity configuration in conjunction with grid connection and disconnection risk constraints. Through day-ahead scheduling baseline and intraday dynamic correction strategy combined with hybrid energy storage collaborative control, it smooths out wind and solar power output fluctuations, improves grid acceptance stability, and completes the comprehensive system benefit evaluation from the dimensions of efficiency, economy, and interactivity. It relies on dynamic weights and Monte Carlo simulation to achieve iterative optimization of system parameters. This scheme mainly focuses on the operation scheduling and capacity configuration optimization of new energy power plants, emphasizing steady-state control and benefit optimization after power plant commissioning, and does not involve the three-dimensional collaborative design, multi-disciplinary modeling verification, and multi-dimensional simulation coupled optimization system in the early stages of engineering. Furthermore, its evaluation weights are only dynamically adjusted based on the system operating status, without relating to design parameters such as the three-dimensional spatial layout of the power plant, equipment layout, and pipeline arrangement. There is no mapping correlation mechanism between green evaluation indicators and three-dimensional design parameters, making it impossible to achieve accurate correction of design parameters from the green assessment results.

[0004] Existing technologies, such as the photovoltaic power plant design system and method based on ecological collaborative optimization disclosed in patent application CN119442714A, optimize the layout of photovoltaic modules based on the weights of power generation revenue and ecological benefits through initial configuration, secondary optimization, ecological assessment, and collaborative optimization modules, taking into account both power plant power generation efficiency and ecological protection effects. Although this solution involves layout optimization and ecological assessment in the power plant design stage, it is only applicable to single photovoltaic power plant scenarios and cannot adapt to the complex design scenarios of multi-source coupled power plants such as photovoltaic, wind power, biogas, landfill gas, and natural gas combined heat and power. At the same time, its evaluation and optimization mode is a post-optimization mode of design first, evaluation later, and shallow feedback. The ecological assessment is only used as a verification means after the layout is completed, and it is not deeply integrated with the multi-dimensional simulation system, lacking quantitative evaluation basis supported by simulation data.

[0005] The common shortcomings of the aforementioned existing technologies are as follows: First, most of these simulation tools operate independently for a single design stage. There is a lack of data exchange interfaces and linkage triggering mechanisms between modules such as layout simulation, lighting resource simulation, energy consumption and loss simulation, and grid connection and power flow simulation. This results in one-dimensional design changes not being automatically propagated to other simulation modules for cascading impact analysis, and simulation results not being automatically transferred and iterated collaboratively on the same platform. Furthermore, these simulation tools primarily serve the scheme comparison and verification during the design phase and have not yet established data linkage with greenness assessment. Key indicators required for assessment cannot be automatically extracted from simulation results and fed into the assessment model, creating a data barrier between assessment and simulation.

[0006] Second, existing optimization schemes are mostly aimed at the control optimization during the power plant operation phase, rather than the three-dimensional refined collaborative optimization during the engineering design phase; the few schemes involving design optimization can only achieve simple layout feedback optimization, and green assessment indicators cannot be automatically extracted from simulation results and calculated in a linked manner. There are data barriers between simulation data and green evaluation models, making it impossible to form an integrated optimization link with the same data source and synchronized process.

[0007] Third, existing technologies mostly use fixed evaluation weights or only fine-tune dynamic weights based on operational data. They have not established a reverse mapping rule library between greenness evaluation indicators and key parameters of three-dimensional design. They cannot reverse locate design defects such as spatial layout, equipment parameters, pipeline routes, and civil engineering scale based on greenness shortcomings indicators, and cannot achieve adaptive correction of three-dimensional model parameters driven by evaluation results. At the same time, the evaluation indicator system is fixed and cannot be dynamically adapted and adjusted according to the differences in energy coupling and terrain conditions of multi-source new energy power plants, which limits the accuracy and adaptability of design optimization.

[0008] Fourth, existing technologies have not established a mechanism for synchronous modeling, parameter linkage and updating, and cross-professional conflict verification among multiple disciplines such as civil engineering, electrical engineering, new energy, grid connection, and cost estimation. This makes it impossible to resolve conflicts in numerical coupling, spatial positioning, and boundary constraints of multiple disciplines in advance, resulting in many hidden dangers and low iteration efficiency during the design phase. It is difficult to meet the stringent requirements of green, intelligent, and refined design of multi-source new energy coupled power plants. Summary of the Invention

[0009] To address the aforementioned technical shortcomings, the present invention aims to provide a method and system for three-dimensional collaborative design and greenness assessment of multi-source new energy power plants.

[0010] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: The present invention provides a three-dimensional collaborative design and greenness assessment method for multi-source new energy power plants, including: Step 1, model building: build a unified cloud-based three-dimensional collaborative design platform, gather multi-disciplinary design resources, perform multi-disciplinary synchronous modeling, parameter linkage update, cross-disciplinary automatic conflict verification, and construct a fully integrated three-dimensional digital model.

[0011] Step 2: Model Simulation: Based on the integrated 3D digital model, five simulation sub-engines are constructed, including layout, lighting resources, energy consumption and loss, power grid connection and power flow, and construction conditions. A data communication, linkage triggering, and collaborative iteration mechanism is established among the simulation sub-engines through the built-in data bus to optimize the design scheme through simulation.

[0012] Step 3: Model Evaluation: Construct a comprehensive greenness evaluation scoring model for multi-source renewable energy coupled power plants, conduct quantitative evaluation of greenness, establish a reverse mapping rule library between greenness evaluation indicators and three-dimensional design parameters, reverse locate design defects based on the quantitative evaluation results of greenness, adaptively correct key design parameters of the three-dimensional model, and complete the iterative optimization of the design scheme.

[0013] Secondly, this invention provides a three-dimensional collaborative design and greenness assessment system for multi-source new energy power plants, including: a model building module: used to build a unified cloud-based three-dimensional collaborative design platform, gather multi-disciplinary design resources, perform multi-disciplinary synchronous modeling, parameter linkage updates, cross-disciplinary automatic conflict verification, and construct a fully integrated three-dimensional digital model.

[0014] Model simulation module: Based on the global integrated 3D digital model, it is used to build five types of simulation sub-engines: layout, light and shadow resources, energy consumption and loss, power grid access and power flow, and construction conditions. It also establishes a data communication, linkage triggering and collaborative iteration mechanism among the simulation sub-engines through the built-in data bus to perform simulation optimization of the design scheme.

[0015] Model evaluation module: Used to construct a comprehensive greenness evaluation scoring model for multi-source renewable energy coupled power plants, conduct quantitative evaluation of greenness, establish a reverse mapping rule library between greenness evaluation indicators and three-dimensional design parameters, reverse locate design defects based on the quantitative evaluation results of greenness, adaptively correct key design parameters of the three-dimensional model, and complete the iterative optimization of the design scheme.

[0016] A historical database is designed to store all historical design schemes.

[0017] The beneficial effects of the present invention are as follows: 1. The present invention provides a multi-source new energy power station three-dimensional collaborative design and greenness assessment method and system, which builds a unified cloud-end three-dimensional collaborative design platform, gathers five major professional design resources of civil engineering, electrical engineering, new energy, power grid connection, and engineering cost, realizes multi-professional homologous synchronous modeling, parameter linkage update and cross-professional automatic conflict check, constructs a global integrated three-dimensional digital model, and unifies the multi-professional design data standards and spatial benchmarks. Based on the global integrated three-dimensional digital model, five types of simulation sub-engines for layout simulation, light and shadow resource simulation, energy consumption and loss simulation, power grid access power flow simulation, and construction condition simulation are constructed in parallel. The built-in data bus realizes data intercommunication, linkage triggering and collaborative iteration of each simulation sub-engine, breaks the disadvantages of fragmentation and data fragmentation of traditional simulation modules, and completes multi-dimensional simulation coupling analysis and multi-objective scheme collaborative optimization. On this basis, a comprehensive greenness assessment scoring model adapted to the scenario of multi-source new energy coupled power stations is constructed, and a reverse mapping rule base between greenness assessment indicators and key three-dimensional design parameters is established. Relying on the quantitative greenness assessment results, the three-dimensional design parameters are reversely corrected, so as to realize the full-process closed loop of three-dimensional collaborative design, multi-dimensional simulation coupling, quantitative greenness assessment and scheme iterative optimization, and effectively solve the technical shortcomings of poor design collaboration, weak simulation linkage, lagging green assessment and unclosed design optimization in the prior art.

[0018] 2. Relying on the unified cloud-end three-dimensional collaborative design platform, the present invention unifies the five majors of civil engineering, electrical engineering, new energy, power grid connection and engineering cost into the same coordinate system and the same data standard for synchronous modeling, supports multi-professional parameter linkage update and automatic conflict check, can identify and troubleshoot various cross-professional parameter conflicts such as numerical coupling, spatial positioning and boundary constraints in advance, and overcomes the defects of design deviation, spatial interference and later rework caused by traditional two-dimensional serial design and independent modeling of each major.

[0019] 3. The present invention builds five types of simulation sub-engines based on the unified three-dimensional digital model, and realizes data intercommunication, linkage triggering and collaborative iteration of each simulation module through the built-in data bus, so that the multi-dimensional simulation results of spatial layout, new energy resource utilization, system energy consumption and loss, power grid grid-connection power flow, and construction condition are coupled and fed back in both directions, forming a multi-constraint and multi-objective integrated simulation optimization mechanism. It solves the industry pain points in the prior art that the simulation modules operate independently, data is fragmented, and collaborative iteration cannot be performed. Compared with the traditional single power generation simulation and single economy simulation schemes, it can simultaneously take into account the safety, energy saving, construction feasibility and economic rationality of the power station, and is adapted to complex mountainous terrain and complex working conditions with multi-energy coupling.

[0020] 4. This invention automatically maps low-scoring indicators in the greenness assessment to specific design parameters and components in the 3D model through a reverse mapping rule base, and highlights the defect locations in the 3D view. Manual problem finding is transformed into automatic location and highlighting of defective components. After the correction scheme is confirmed, the model parameters are automatically modified and propagated to all related components through a linkage mechanism, automatically triggering the next round of simulation and re-evaluation verification until the greenness standard is met. This achieves a fully automated closed loop of assessment → location → correction → re-evaluation, significantly shortening the design optimization cycle.

[0021] 5. This invention is fully adaptable to the design scenarios of multi-source coupled power plants, including photovoltaic, wind power, biogas, landfill gas, and natural gas cogeneration. It breaks through the limitations of existing technologies that are only applicable to single photovoltaic or wind-solar-hydrogen single systems. It can take into account the characteristics of multi-energy coupled operation, the constraints of pipeline and electrical cross-layout, biomass resource recycling, and multi-dimensional low-carbon ecological indicators. It is adapted to the construction needs of multi-source integrated energy power plants under the new power system and has strong engineering implementation and industry promotion value. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a schematic diagram of the implementation steps of the method of the present invention.

[0024] Figure 2 This is a schematic diagram of the system structure connection of the present invention. Detailed Implementation

[0025] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] Please see Figure 1 As shown, the three-dimensional collaborative design and greenness assessment method for multi-source new energy power plants includes: Step 1, Model building: Build a unified cloud-based three-dimensional collaborative design platform, gather multi-disciplinary design resources, perform multi-disciplinary synchronous modeling, parameter linkage update, cross-disciplinary automatic conflict verification, and construct a three-dimensional digital model that integrates the entire domain.

[0027] The aforementioned multi-disciplinary design resources include design resources for five major disciplines: civil engineering, electrical engineering, new energy, power grid connection, and engineering cost.

[0028] It should be noted that the unified cloud-based 3D collaborative design platform is deployed using a layered cloud architecture, divided from top to bottom into a business application layer, a collaborative engine layer, a data middleware layer, and an infrastructure layer. Each layer is decoupled from the others and has standardized interfaces. The basic deployment, hardware adaptation, network configuration, and system connection methods of the cloud-based layered architecture are all conventional and mature technologies in this field. This invention focuses on the innovative design of a multi-disciplinary collaborative modeling mechanism, a multi-simulation engine data bus linkage mechanism, and a reverse mapping iteration mechanism between green indicators and design parameters. Therefore, the conventional platform construction process will not be described in detail.

[0029] In a specific embodiment, the process of multi-disciplinary synchronous modeling, parameter linkage update, and cross-disciplinary automatic conflict verification is as follows: using the global 3D geographic information base as a unified data source, five major disciplines are modeled in parallel in accordance with a unified coordinate system, unified data specifications, and unified model file format; relying on the global unified coordinate system, the sub-models of each discipline are spliced ​​and integrated to generate a global integrated 3D digital model.

[0030] Enable the cloud model real-time synchronization and parameter linkage update interface. When any discipline adjusts the model structure or modifies the design parameters, the associated parameters of the overall collaborative model are automatically refreshed and updated across the entire domain. The platform calls the built-in design specification verification rule library to perform cross-disciplinary conflict verification in real time, automatically identify design defects, and push early warning information.

[0031] Preferably, the design specification verification rule base specifically includes: numerical coupling conflict rules, spatial positioning conflict rules, and boundary constraint conflict rules.

[0032] The process of constructing the numerical coupling conflict rules is as follows: extract all historical design schemes from the design history database, including design parameters, design review records and construction feedback records of each discipline; mark the design quality results for each historical design scheme, and perform working condition stratification according to the design working conditions.

[0033] The design quality results include qualified, minor defects, and serious defects.

[0034] The above describes the design operating conditions of power plants, which are categorized into resource-output conditions, energy consumption-heat conditions, grid connection and power flow conditions, construction conditions, and multi-source integrated coupling conditions. The resource-output conditions include typical annual average conditions, extreme solar and wind speed conditions, extreme solar and wind speed conditions, and low biogas / landfill gas production conditions. The energy consumption-heat conditions include summer high-temperature design conditions, winter low-temperature design conditions, and spring / autumn transitional season average conditions. The grid connection and power flow conditions include maximum output transmission conditions, minimum output conditions, fault disturbance conditions, and peak-shaving response conditions. The construction conditions include rainy season construction conditions, winter frozen soil construction conditions, conventional optimal construction conditions, and large equipment hoisting conditions. The multi-source integrated coupling conditions include multi-source full-power coupling conditions and biomass backup guarantee conditions.

[0035] All professional design parameters are uniformly coded to establish a cross-professional parameter identification system; continuous parameters are normalized.

[0036] The unified coding adopts a segmented hierarchical combination coding format. The code is composed of five fields: professional code, object type code, sub-object code, parameter type code, and serial number. The fields are separated by delimiters. The professional code is used to distinguish different design disciplines, the object type code is used to distinguish the major object category, the sub-object code is used to locate the specific equipment / structure object, the parameter type code distinguishes the physical attribute category of the parameter, and the serial number ensures that each design parameter is globally unique.

[0037] For example: code DQ-SB-GF-CS-001, DQ: professional code → electrical profession, SB: object type code → equipment class, GF: sub-object code → photovoltaic (photovoltaic module), CS: parameter type code → size parameter, 0001 is the serial number, the code means: the 0001st size parameter corresponding to the photovoltaic module under the electrical profession.

[0038] It should be noted that continuous parameters are parameters whose values ​​are continuous and can take any real number within a certain range. Examples include: component tilt angle, equipment spacing, cable cross-section, foundation thickness, light intensity, power flow, construction cost, and safety distance. These are all continuous parameters; their values ​​can change continuously, unlike discrete options such as "yes" or "no." The min-max normalization method is used to normalize the parameters to the dimensionless range of 0 to 1, eliminating differences in dimensions and numerical magnitudes between different parameters.

[0039] Within each operating condition layer, outlier detection is performed independently for each design parameter to identify whether it is a single-parameter anomaly, and a list of all single-parameter anomalies in each historical design scheme is recorded.

[0040] In the above, outlier detection: within the sample set of the same working condition layer, a one-dimensional sample subset is constructed for each design parameter, and the corresponding parameter is calculated by inputting it into the isolated forest model. When the outlier score is greater than the preset outlier score threshold, the parameter is determined to be a single parameter outlier.

[0041] Using the normalized design parameter vector of all disciplines within the same working condition layer as input, the Mahalanobis distance of this parameter vector relative to the historical normal design sample space of this working condition layer is calculated to determine whether the design scheme has multidimensional joint anomalies. When the Mahalanobis distance is greater than a preset distance threshold, the design scheme is determined to have multidimensional joint anomalies. If multidimensional joint anomalies are determined to exist, the contribution of each design parameter to the overall multidimensional joint anomaly is calculated by decomposing the SHAP value, and parameters whose contribution exceeds a preset contribution threshold are marked as anomaly source parameters.

[0042] It should be noted that the improved isolated forest algorithm, Mahalanobis distance calculation, and SHAP value decomposition are all conventional data analysis methods in this field. This invention does not improve the isolated forest algorithm itself, but only adapts the above algorithms to the power plant design parameter anomaly identification business process under the working condition layer. Therefore, the specific implementation process of the algorithm is not described in detail here.

[0043] It should be noted that historical normal design samples are qualified historical design schemes.

[0044] Meanwhile, the parameters that contribute the most to the multidimensional joint anomaly detection results are included in the candidate anomaly parameter combination pool.

[0045] For each historical design scheme, the union of all design parameters of a single-parameter anomaly with the design parameters of the candidate anomaly parameter combination pool is taken as the numerical anomaly parameter combination of that scheme.

[0046] The numerical anomaly parameters of each historical design scheme are combined to construct a transaction dataset; the Apriori algorithm, which introduces unidirectional expansion of the value range, bidirectional expansion enhancement and elbow rule dynamic partitioning strategy, is used to carry out frequent itemset mining, and all frequent itemsets with support greater than or equal to 0.05 are selected.

[0047] Among them, a support threshold of 0.05 indicates that the occurrence rate of the corresponding abnormal parameter combination in all historical design scheme samples is not less than 5%.

[0048] It should be noted that the unidirectional expansion strategy of the value range takes the historical high-frequency abnormal value range of the parameter as the benchmark and expands the range boundary in the direction of the one-sided specification boundary; the bidirectional expansion enhancement strategy expands the value range range in both directions to both sides of the specification boundary, so as to take into account the specification constraints and the distribution of historical engineering samples, and obtain a discrete value range partition that is suitable for the engineering scenario.

[0049] When implementing a unidirectional expansion strategy or a bidirectional expansion enhancement strategy for the value range, a maximum expansion coefficient threshold is preset to control the expansion range of the interval boundary; after the value range partition is expanded and adjusted, the partition boundary shall not exceed the upper and lower limits of the power design code corresponding to the parameter.

[0050] Extract numerical anomaly association rules from the frequent itemset that satisfy Confidence ≥ 0.6 and Lift > 1.

[0051] The extracted association rules are connected according to the causal direction to form a complete numerical anomaly association chain; the parameters in the association chain are paired according to the source parameter → target parameter to establish a numerical coupling relationship mapping between parameters, and finally a structured numerical coupling conflict rule is formed, which is then encoded and stored in the rule base.

[0052] Numerical anomaly association rules are extracted from all frequent itemsets obtained from the mining. The filtering conditions are set as follows: rule confidence ≥ 0.6 and lift > 1. Here, confidence represents the conditional probability that the consequent parameter anomaly will occur simultaneously given that the antecedent parameter anomaly occurs. Lift is used to determine whether there is a real positive coupling association between parameter anomalies. Lift greater than 1 means that the association is not caused by random sample coincidence.

[0053] The numerical anomaly association rule takes the form of {anomaly source parameter combination} ⇒ {anomaly consequence parameter} (anomaly source parameter combination} ⇒ {anomaly consequence parameter}).

[0054] Spatial positioning conflict rule construction: First, read the power engineering design specifications and equipment installation technical conditions to sort out the minimum safe clearance, entity avoidance, and land occupation avoidance constraints between various components; assign a unique rule number to each spatial positioning conflict rule, configure the set of applicable objects of the rule, and bind the spatial coordinate parameters, external dimension parameters, and spatial occupancy bounding box parameters of the corresponding objects through the aforementioned globally unified parameter coding.

[0055] Based on the 3D model, extract the AABB axis-aligned bounding box or OBB oriented bounding box of the object to obtain the spatial position and shape boundary of the object in a globally unified coordinate system; configure conflict judgment logic: ① Entity interference collision judgment: when the spatial bounding boxes of two different objects intersect and overlap, it is judged that an entity spatial interference conflict has occurred; ② Insufficient safety distance judgment: calculate the shortest spatial distance between the two object model entities. If the shortest spatial distance is less than the minimum safety clearance specified in the standard, it is judged that a spatial safety clearance conflict has occurred.

[0056] Configure an output information template for each spatial positioning conflict rule. When a conflict is triggered, output the rule number, parameter code of the conflicting object, object name, actual spatial distance, standard requirement threshold, conflict location coordinates, and standard basis.

[0057] Boundary constraint conflict rule construction: Based on industry power engineering design specifications, equipment technical specifications, and boundary conditions for multi-source renewable energy power plant construction, the legal value constraint ranges of all design parameters across five major disciplines are identified, and the upper and lower limits of allowable values, engineering limit thresholds, and mandatory constraint attributes of each parameter are determined. Based on the aforementioned unified parameter coding system across all disciplines, independent boundary constraint rules are established for each code-corresponding design parameter, binding the parameter code, parameter name, standard basis, allowable value range, and engineering limit threshold one by one to form structured rule entries.

[0058] The boundary constraint conflict determination logic is as follows: read the actual value of the corresponding parameter in the current 3D design model. If the actual value of the parameter is greater than the upper limit allowed by the specification or less than the lower limit allowed by the specification, it is determined that the parameter triggers a boundary constraint conflict. For key parameters with engineering limit thresholds, further compare whether the actual value exceeds the engineering limit boundary.

[0059] At the same time, conflict output content is configured for each boundary constraint rule, including abnormal parameter code, actual parameter value, standard threshold range, out-of-limit deviation, and relevant design discipline and standard basis.

[0060] It should be noted that boundary constraint conflicts are single-parameter self-compliance anomalies, spatial positioning conflicts are multi-model spatial geometric interference anomalies, and numerical coupling conflicts are multi-parameter linkage matching anomalies. These three types of rules are independent of each other, complementary and cover each other, and together realize multi-disciplinary design full-dimensional conflict detection.

[0061] In a specific embodiment, the process of performing cross-disciplinary conflict verification in real time is as follows: when any discipline adjusts the model structure or modifies the design parameters, the change influence domain is determined, then the complete parameter set of the affected components is extracted from the change influence domain, and then the numerical coupling conflict rules, spatial positioning conflict rules and boundary constraint conflict rules in the design specification verification rule base are matched in parallel to output the conflict type.

[0062] It should be noted that the change impact domain is determined as follows: (1) Locate the source component and change parameter code of the change; read the metadata of the model association relationship, obtain the set of components that have direct topological association, parameter binding association, and spatial geometric proximity association with the source component, and use them as the first-level directly affected components; (2) Starting from the first-level directly affected components, recursively search for the second-level indirect affected components that have parameter linkage binding, spatial interference risk, and engineering constraint association; (3) Combine the spatial range of the three-dimensional model bounding box, perform spatial filtering on the above components, and remove irrelevant components whose spatial distance is greater than the preset spatial distance threshold; (4) The source component, the first-level directly affected components, and the second-level indirect affected components together constitute the change impact domain.

[0063] The term "spatial geometric proximity" refers to calculating the minimum spatial distance between the bounding boxes of two components in a globally unified three-dimensional coordinate system. If the minimum spatial distance is less than a preset spatial proximity distance threshold, the two components are determined to be spatially geometrically proximity. Components that are spatially geometrically proximity may not have parameter binding or topological logical association, but because they are close in spatial location, there is a potential risk of spatial positioning conflict after the source component is modified. Therefore, they are included in the first-level directly affected component set.

[0064] The parameter linkage binding refers to the existence of parameter-driven linkage relationships between components, where a change in the parameters of one component will drive or constrain the associated parameters of another component; the spatial interference risk refers to the fact that although a component is not directly adjacent to the spatially affected component, it is close to the spatially affected component at the first level, and the change in the source component will be transmitted through the first-level component, bringing potential spatial conflict risks; the engineering constraint association refers to the fact that there is no parameter-level spatial association between components, but according to the power engineering specifications and process requirements, the component needs to be re-verified for compliance after the source component is changed.

[0065] For multiple conflicts with causal relationships, a conflict propagation chain is constructed, and then structured conflict early warning information is generated. This information is then pushed to the relevant disciplines to modify the design parameters. The modified model is then re-executed with incremental conflict detection within the change influence domain. If conflicts still exist, the modification is repeated until all conflicts are cleared.

[0066] It should be noted that causal relationship identification: For any two conflicts output under the same change's influence domain, if any of the following conditions are met, a causal relationship is determined to exist between them: ① Parameter dependency transmission condition: There is a pre-stored parameter linkage binding driver-driven relationship between the trigger parameter codes corresponding to the two conflicts; for example: Conflict A is triggered by the bracket height parameter, and there is a parameter linkage binding between the bracket height and the component's ground clearance; Conflict B is triggered by the component's ground clearance; then A is presumed to be the cause and B to be the effect.

[0067] It should be noted that the pre-stored parameter linkage binding driver-driven relationship is a parameter logical mapping relationship pre-stored in the 3D collaborative design platform. Using the unified parameter code of each component as the association primary key, it records the driver parameter code, the driven parameter code, and the source of the linkage relationship. This indicates that a modification to a certain driver parameter will have a driving or constraining effect on the corresponding driven parameter. This binding relationship originates from equipment assembly relationships, 3D model assembly topology, power engineering design processes and specifications, and is used for recursive retrieval of change impact domains, linked updates of model parameters, and identification of causal relationships between multiple conflicts.

[0068] ② Spatial geometric transmission conditions: The triggering component of a preceding conflict is a participating component in a subsequent conflict. The preceding conflict is caused by an abnormality in the spatial state of a component, and the change in the spatial state of the component directly causes the spatial constraints of the subsequent conflict to be broken. For example: Conflict A is a spatial positioning conflict, and the source component's position is offset. The source component is a participating component in conflict B. After A occurs, the spatial state of the component changes, directly causing the spatial constraints of B to be broken. Conflict A causes the component's position to offset, which in turn triggers spatial conflict B.

[0069] ③ Engineering constraint propagation conditions: Preceding conflicts cause abnormal changes in key engineering indicators. These key engineering indicators become input conditions for subsequent conflict rules. According to power engineering design specifications or process constraints, the anomaly of the preceding conflict will directly break the verification constraints of the subsequent conflict. For example: After conflict A is triggered, a certain key indicator changes; this key indicator is the input condition for the rule corresponding to conflict B; according to power engineering processes or specifications, the anomaly of A will directly lead to the breaking of the verification conditions of B. For example: installed capacity parameter exceeding the limit conflict A → change in system short-circuit current index → ​​triggering grounding grid verification conflict B.

[0070] Based on the above determination results, a conflict causal directed graph is constructed; conflict nodes without predecessor parent conflicts are marked as root cause conflict nodes, and conflict nodes pointed to by other nodes are marked as derived conflict nodes; starting from each root cause conflict node, subsequent derived conflict nodes are sequentially connected to generate corresponding conflict propagation chains; conflicts that do not satisfy the above causal conditions are treated as independent conflict entries.

[0071] It should be noted that structured conflict early warning information is generated based on conflict detection results and conflict propagation chains; the structured conflict early warning information includes change task number, change impact domain identifier, set of independent conflict items, conflict propagation chain array, root cause components and parameter codes, relevant professional identifiers, and standard basis.

[0072] Step 2: Model Simulation: Based on the integrated 3D digital model, five simulation sub-engines are constructed, including layout, lighting resources, energy consumption and loss, power grid connection and power flow, and construction conditions. A data communication, linkage triggering, and collaborative iteration mechanism is established among the simulation sub-engines through the built-in data bus to optimize the design scheme through simulation.

[0073] In a specific embodiment, the specific process of establishing the data communication, linkage triggering, and collaborative iteration mechanism of each simulation sub-engine is as follows: a data bus is built as the data interaction hub between each simulation sub-engine to enable data communication.

[0074] Establish an event-driven linkage triggering mechanism, with the specific triggering link as follows: (1) Layout → Light and Shadow: When the layout simulation sub-engine completes and releases a new layout scheme, the data bus will automatically trigger the light and shadow simulation sub-engine, which will then recalculate the shadow distribution based on the new layout.

[0075] (2) Light and Shadow → Energy Consumption: After the light and shadow simulation is completed and the shadow data is published, this event will automatically trigger the energy consumption simulation sub-engine, which will use the shadow data as input to recalculate the power generation.

[0076] (3) Energy consumption → power flow: After the energy consumption simulation is completed and the power generation curve is published, this event will automatically trigger the power grid access power flow simulation sub-engine, and use the power curve as the boundary condition to recalculate the power flow.

[0077] (4) Construction → Layout: After the construction condition simulation is completed, the published construction condition data will automatically trigger the layout simulation sub-engine, which will use the construction constraints as new optimization conditions to re-evaluate the layout scheme.

[0078] (5) Power flow → grid connection scheme: When the power flow simulation detects problems such as voltage exceeding the limit, it will release an optimization requirement event and automatically trigger the optimization process of the grid connection scheme.

[0079] Each simulation sub-engine implements a closed-loop collaborative mechanism based on the above event-driven link, which drives subsequent simulation iterations with preceding simulation results and corrects preceding schemes with subsequent constraints, thus forming a multi-condition, multi-dimensional coupled simulation iteration optimization system.

[0080] It should be noted that the specific simulation calculation content of each simulation sub-engine can be adaptively configured according to the actual simulation design requirements of multi-source new energy power plants, and this invention does not impose specific limitations on this.

[0081] When the design scheme changes, a new round of simulation iteration is initiated. The data bus coordinates the sub-engines and completes a complete coupled simulation in sequence or in parallel according to the linkage trigger chain of layout → lighting → energy consumption → power flow → construction.

[0082] After the simulation is completed, all key indicators of this round are output and compared with all key indicators of the previous round. If the change of all key indicators is less than its preset convergence threshold, the design scheme is determined to be stable and the iteration ends. Otherwise, it is determined that the simulation has not converged. All simulation data of this round are summarized to generate a comprehensive evaluation report. The adjusted design scheme is generated through a multi-objective collaborative optimization algorithm and the model is updated.

[0083] It should be noted that the multi-objective optimization algorithm is a publicly available algorithm in this field. This invention does not improve the multi-objective optimization algorithm itself, but mainly adapts it to the business process of scheme iteration after the linkage of multiple simulation engines in new energy power plants. Therefore, the internal iterative calculation process of the algorithm will not be described in detail here.

[0084] Among the above, the key indicators are as follows: the layout simulation sub-engine outputs the site land utilization rate, equipment array spacing, equipment coverage, structure occupancy area, and layout compliance margin, etc.; the light and shadow resource simulation sub-engine outputs the time-series shadow occlusion duration, effective irradiance, and occupancy loss irradiance ratio, etc.; the energy consumption and loss simulation sub-engine outputs the power generation curve of the power plant, annual power generation, losses of various equipment and cables, and equivalent utilization hours, etc.; the grid connection power flow simulation sub-engine outputs the node voltage amplitude, line load rate, short-circuit current, power factor, and grid connection safety margin, etc.; and the construction condition simulation sub-engine outputs the available area of ​​the construction work surface, hoisting operation space margin, key process duration, and construction difficulty coefficient, etc.

[0085] The updated model triggers another round of simulation iterations until the design scheme has stabilized.

[0086] Step 3: Model Evaluation: Construct a comprehensive greenness evaluation scoring model for multi-source renewable energy coupled power plants, conduct quantitative evaluation of greenness, establish a reverse mapping rule library between greenness evaluation indicators and three-dimensional design parameters, reverse locate design defects based on the quantitative evaluation results of greenness, adaptively correct key design parameters of the three-dimensional model, and complete the iterative optimization of the design scheme.

[0087] In a specific embodiment, the construction process of the comprehensive greenness evaluation scoring model is as follows: extract the design parameters adjusted in each round of simulation and the key indicators output, and use the grey relational analysis method to calculate the grey relational degree γ between each design parameter and each key indicator. i γ i A correlation greater than 0.7 indicates a strong association, while a correlation less than 0.5 indicates a strong association. i ≤0.7 indicates moderate correlation, γ i ≤0.5 indicates a weak correlation; i represents the number of each key indicator, which takes a positive integer value.

[0088] Design parameters and key indicators are treated as two types of nodes, and a bipartite network G is constructed with correlation degree as the edge weight. bi Then, by sharing parameters, the bipartite network is projected into an index-based one-sided network G. 指标 If two key indicators a and b share at least one strongly correlated design parameter, then there is a potential correlation between a and b, with an edge weight of . Finally, hierarchical clustering or minimum spanning tree algorithms are used to analyze G. 指标 Perform tree-like processing to form a directed acyclic tree diagram.

[0089] Where q is the total number of design parameters. Let be the correlation between the a-th key indicator and the k-th design parameter. Let be the correlation between the b-th key indicator and the k-th design parameter, where k represents the design parameter number and takes a positive integer value.

[0090] Using the R-type clustering method, when the correlation coefficient between two indicators is |r ab When |>0.85, it is considered highly redundant, and one of them is retained; those with a correlation degree of γ with all design parameters are removed. i Weakly correlated indicators <0.3; after purification, obtain the remaining target indicators.

[0091] The tree-structured variable weighting method is used to allocate level coefficients according to the level l of the target indicator in the tree structure. Adjust position weights based on the number of sub-indicators. .

[0092] in, ρ is the root node level coefficient, taken as 1.0; ρ is the level attenuation coefficient, taken as 0.7~0.9; l is the level where the index is located; root node l=1. j 基准 n represents the preset baseline weight for the j-th target indicator, where j represents the target indicator number and takes the value of a positive integer. 子 For the number of sub-indicators, The coefficient of variation of the j-th target indicator, The coefficient of variation of the kth design parameter.

[0093] Greenness Comprehensive Assessment Scoring Model: Where L is the total number of layers in the tree diagram, node l is the set of indicators for the l-th layer, and s j Let G be the standardized score of the j-th target indicator, where G represents the comprehensive greenness assessment score, l represents the level number, j represents the target indicator number, and l and j are both positive integers. L represents the number of levels.

[0094] In another specific embodiment, the process of establishing the reverse mapping rule base between greenness evaluation indicators and three-dimensional design parameters is as follows: extract each target indicator and extract the edge weights between each target indicator and each design parameter from the directed acyclic tree relationship graph.

[0095] Sort the edge weights from largest to smallest, obtain the maximum edge weight corresponding to each target indicator, and determine whether the maximum edge weight corresponding to each target indicator is greater than the preset edge weight threshold.

[0096] When the maximum edge weight corresponding to the target indicator is greater than the preset edge weight threshold, the design parameter of the maximum edge weight corresponding to each target indicator is taken as the target design parameter, the three-dimensional component ID associated with the target design parameter is extracted, highlighted in the three-dimensional model, and the target design parameter and the three-dimensional component corresponding to the target design parameter are output.

[0097] When the maximum edge weight corresponding to the target indicator is less than or equal to the preset edge weight threshold, the parent node indicator of the target indicator in the tree diagram is taken as the new target indicator. The edge weights between the parent node indicator and each design parameter are extracted and sorted in descending order of edge weight. The maximum edge weight corresponding to the parent node indicator is obtained, and it is determined whether the maximum edge weight corresponding to the parent node indicator is greater than the preset edge weight threshold. The above process is repeated until an ancestor node with an edge weight greater than the preset threshold is found. The design parameter with the maximum edge weight corresponding to the ancestor node is taken as the target design parameter. According to the 3D component ID corresponding to the target design parameter, it is highlighted in the 3D model, and the target design parameter and the 3D component corresponding to the target design parameter are output.

[0098] The target design parameters are sorted from largest to smallest according to their contribution weight to obtain the inverse mapping rule corresponding to the target indicator; the inverse mapping rules corresponding to each target indicator are summarized to construct a reverse mapping rule library between greenness evaluation indicators and three-dimensional design parameters.

[0099] Preferably, the adaptive correction of key design parameters of the 3D model specifically involves the following process: when the comprehensive greenness assessment score G is lower than a preset comprehensive assessment threshold, the lowest-scoring indicator s is adjusted. min As a marker indicator that needs to be traced.

[0100] Marking index s min For indexing, query the reverse mapping rule base: if a rule with the indicator as the primary key exists in the rule base, directly extract the list of associated design parameters; if no directly matching rule exists in the rule base, trigger the step-by-step tracing process: starting from the indicator, trace the parent node indicator upwards along the tree relationship graph until an ancestor node with a strong association rule is found, then output the target design parameters corresponding to the marked indicator or ancestor node and the three-dimensional components corresponding to the target design parameters, highlight them in the three-dimensional view, and then use an improved multi-objective particle swarm optimization algorithm to generate a correction scheme and make modifications.

[0101] Please see Figure 2 As shown, the multi-source new energy power station three-dimensional collaborative design and greenness assessment system includes: Model building module: used to build a unified cloud-based three-dimensional collaborative design platform, gather multi-disciplinary design resources, perform multi-disciplinary synchronous modeling, parameter linkage update, cross-disciplinary automatic conflict verification, and construct a whole-domain integrated three-dimensional digital model.

[0102] Model simulation module: Based on the global integrated 3D digital model, it is used to build five types of simulation sub-engines: layout, light and shadow resources, energy consumption and loss, power grid access and power flow, and construction conditions. It also establishes a data communication, linkage triggering and collaborative iteration mechanism among the simulation sub-engines through the built-in data bus to perform simulation optimization of the design scheme.

[0103] Model evaluation module: Used to construct a comprehensive greenness evaluation scoring model for multi-source renewable energy coupled power plants, conduct quantitative evaluation of greenness, establish a reverse mapping rule library between greenness evaluation indicators and three-dimensional design parameters, reverse locate design defects based on the quantitative evaluation results of greenness, adaptively correct key design parameters of the three-dimensional model, and complete the iterative optimization of the design scheme.

[0104] A historical database is designed to store all historical design schemes.

[0105] The various thresholds and ranges preset in this invention are all adaptive engineering thresholds pre-configured based on multi-source new energy power plant engineering design specifications, on-site construction conditions, equipment technical parameters, historical high-quality engineering sample data, and simulation iteration accuracy requirements. These preset thresholds are not fixed values ​​and can be adaptively fine-tuned according to the power plant voltage level, site construction scale, equipment model parameters, terrain conditions, and design operating conditions. All threshold settings are constrained by meeting mandatory power engineering design specifications, ensuring simulation accuracy, avoiding design conflicts, and conforming to the design patterns of historical high-quality projects, while also considering algorithm computational efficiency and engineering practicality.

[0106] In practical implementation, various preset thresholds support manual calibration and iterative updates. The threshold value range can be continuously optimized based on historical design experience, simulation results, and engineering implementation experience, further improving the accuracy of parameter anomaly identification, conflict verification, and the rationality of scheme optimization. It should be noted that the specific values ​​of various thresholds fall within the range of conventional engineering experience in this field. This invention does not limit specific threshold values ​​but focuses on protecting the overall technical process of multi-disciplinary parameter verification, anomaly identification, association rule mining, and simulation collaborative optimization based on the collaborative implementation of various thresholds.

[0107] The algorithms involved in this invention are all mature and publicly available basic algorithms in the field for conventional data analysis, geometric detection, and intelligent optimization. Specifically, they include the isolated forest outlier detection algorithm, the Mahalanobis distance multidimensional deviation calculation algorithm, the SHAP value attribution decomposition algorithm, the Apriori frequent itemset mining algorithm, the multi-objective collaborative optimization algorithm, and the 3D bounding box space collision detection algorithm. This invention does not improve or innovate the core principles, iterative logic, or calculation formulas of these basic algorithms, and therefore does not require detailed descriptions of the underlying implementation principles and basic computational processes of the algorithms.

[0108] Various basic algorithms serve only as tools to support the technical solution of this invention. By combining them with custom engineering constraints, business preprocessing strategies, scenario-based judgment rules, simulation collaboration mechanisms, and conflict iteration correction processes, they achieve the technical effects of intelligent verification of multi-disciplinary design parameters, accurate anomaly identification, coupling relationship mining, and automatic iterative optimization of design schemes. This is different from the general application mode of existing single algorithms and solves the industry pain points of traditional new energy power plant design, such as one-sided parameter verification, lagging anomaly identification, one-sided scheme optimization, and poor multi-disciplinary collaboration.

[0109] The examples described in this invention are not limited to the specific embodiments listed above. The examples are merely illustrative to facilitate understanding of the invention and do not constitute a limitation on the scope of protection of this invention. Any modifications, equivalent substitutions, etc., made within the spirit and principles of this invention should be included within the scope of protection.

[0110] The above description is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined in this specification, they should all fall within the protection scope of the present invention.

Claims

1. A three-dimensional collaborative design and greenness assessment method for multi-source renewable energy power plants, characterized in that, include: Step 1: Model Building: Build a unified cloud-based 3D collaborative design platform, gather design resources from multiple disciplines, perform multi-disciplinary synchronous modeling, parameter linkage updates, and cross-disciplinary automatic conflict verification, and construct a fully integrated 3D digital model. Step 2, Model Simulation: Based on the aforementioned integrated 3D digital model, five simulation sub-engines are constructed, including layout, lighting resources, energy consumption and loss, power grid connection and power flow, and construction conditions. A data communication, linkage triggering, and collaborative iteration mechanism for each simulation sub-engine is established through the built-in data bus to optimize the design scheme through simulation. Step 3: Model Evaluation: Construct a comprehensive greenness evaluation scoring model for multi-source renewable energy coupled power plants, conduct quantitative evaluation of greenness, establish a reverse mapping rule library between greenness evaluation indicators and three-dimensional design parameters, reverse locate design defects based on the quantitative evaluation results of greenness, adaptively correct key design parameters of the three-dimensional model, and complete the iterative optimization of the design scheme.

2. The method for three-dimensional collaborative design and greenness assessment of multi-source new energy power plants according to claim 1, characterized in that, The multi-disciplinary design resources include design resources for five major disciplines: civil engineering, electrical engineering, new energy, power grid connection, and engineering cost.

3. The method for three-dimensional collaborative design and greenness assessment of multi-source new energy power plants according to claim 1, characterized in that, The specific process of multi-disciplinary synchronous modeling, parameter linkage update, and cross-disciplinary automatic conflict verification is as follows: using the global 3D geographic information base as a unified data source, five major disciplines are modeled in parallel in accordance with a unified coordinate system, unified data specifications, and unified model file format; relying on the global unified coordinate system, the sub-models of each discipline are spliced ​​and integrated to generate a global integrated 3D digital model. Enable the cloud model real-time synchronization and parameter linkage update interface. When any discipline adjusts the model structure or modifies the design parameters, the associated parameters of the overall collaborative model are automatically refreshed and updated across the entire domain. The platform calls the built-in design specification verification rule library to perform cross-disciplinary conflict verification in real time, automatically identify design defects, and push early warning information.

4. The method for three-dimensional collaborative design and greenness assessment of multi-source new energy power plants according to claim 3, characterized in that, The design specification verification rule base specifically includes: numerical coupling conflict rules, spatial positioning conflict rules, and boundary constraint conflict rules; The process of constructing the numerical coupling conflict rule is as follows: extract all historical design schemes from the design history database, including design parameters, design review records and construction feedback records of each discipline; mark the design quality results for each historical design scheme and perform working condition stratification according to the design working conditions; All professional design parameters are uniformly coded to establish a cross-professional parameter identification system; continuous parameters are normalized. Within each working condition layer, outlier detection is performed independently for each design parameter to identify whether it is a single parameter anomaly, and a list of all single parameter anomalies in each historical design scheme is recorded. Using the normalized design parameter vector of all disciplines within the same working condition layer as input, the Mahalanobis distance of the parameter vector relative to the historical normal design sample space of this working condition layer is calculated to determine whether the design scheme has multidimensional joint anomalies. When the Mahalanobis distance is greater than a preset distance threshold, the design scheme is determined to have multidimensional joint anomalies. If multidimensional joint anomalies are determined to exist, the contribution of each design parameter to the overall multidimensional joint anomaly is calculated by decomposing the SHAP value, and parameters whose contribution exceeds the preset contribution threshold are marked as anomaly source parameters. Meanwhile, the parameters that contribute the most to the multidimensional joint anomaly detection results are included in the candidate anomaly parameter combination pool. For each historical design scheme, the union of all design parameters of a single-parameter anomaly with the design parameters of the candidate anomaly parameter combination pool is taken as the numerical anomaly parameter combination of that scheme. The numerical anomaly parameters of each historical design scheme are combined to construct a transaction dataset; the Apriori algorithm, which introduces unidirectional expansion of the value range, bidirectional expansion enhancement and elbow rule dynamic partitioning strategy, is used to carry out frequent itemset mining, and all frequent itemsets with support greater than or equal to 0.05 are selected. Extract numerical anomaly association rules from the frequent itemset that satisfy Confidence ≥ 0.6 and Lift > 1; The extracted association rules are connected in the causal direction to form a complete numerical anomaly association chain; the parameters in the association chain are paired according to the source parameter → target parameter to establish a numerical coupling relationship mapping between parameters, and finally a structured numerical coupling conflict rule is formed, which is then encoded and stored in the rule base.

5. The method for three-dimensional collaborative design and greenness assessment of multi-source new energy power plants according to claim 3, characterized in that, The process of performing real-time cross-disciplinary conflict verification is as follows: When any discipline adjusts the model structure or modifies the design parameters, the change influence domain is determined. Then, the complete parameter set of the affected components is extracted from the change influence domain. Finally, it is matched in parallel with the numerical coupling conflict rules, spatial positioning conflict rules, and boundary constraint conflict rules in the design specification verification rule base, and the conflict type is output. For multiple conflicts that have a causal relationship, construct a conflict propagation chain and then generate structured conflict early warning information; Then, the model is pushed to the relevant discipline for design parameter modification. The modified model is then re-executed for incremental conflict detection within the change influence domain. If conflicts still exist, the modification is repeated until all conflicts are cleared.

6. The method for three-dimensional collaborative design and greenness assessment of multi-source new energy power plants according to claim 1, characterized in that, The specific process for establishing the data communication, linkage triggering, and collaborative iteration mechanism among the various simulation sub-engines is as follows: A data bus is built as the central hub for data interaction between various simulation sub-engines to enable data communication. Establish an event-driven linkage triggering mechanism; when the design scheme changes, start a new round of simulation iteration, coordinate each sub-engine through the data bus, and complete a complete coupled simulation in sequence or in parallel according to the linkage triggering chain of layout → lighting → energy consumption → power flow → construction. After the simulation is completed, all key indicators of this round are output and compared with all key indicators of the previous round. If the change of all key indicators is less than its preset convergence threshold, the design scheme is determined to be stable and the iteration ends. Otherwise, it is determined that the simulation has not converged. All simulation data of this round are summarized to generate a comprehensive evaluation report. The adjusted design scheme is generated through a multi-objective collaborative optimization algorithm and the model is updated. The updated model triggers another round of simulation iterations until the design scheme has stabilized.

7. The method for three-dimensional collaborative design and greenness assessment of multi-source new energy power plants according to claim 1, characterized in that, The specific construction process of the comprehensive greenness assessment scoring model is as follows: The design parameters adjusted and the key output indicators in each simulation round are extracted, and the grey relational analysis method is used to calculate the grey relational degree γ between each design parameter and each key indicator. i γ i A value >0.7 indicates a strong correlation, and 0.5 < γ indicates a strong correlation. i ≤0.7 indicates moderate correlation, γ i ≤0.5 indicates a weak correlation; i represents the number of each key indicator, which takes a positive integer value; Design parameters and key indicators are treated as two types of nodes, and a bipartite network G is constructed with correlation degree as the edge weight. bi Then, by sharing parameters, the bipartite network is projected into an index-based one-sided network G. 指标 If two key indicators, a and b, share at least one strongly correlated design parameter, then there is a potential correlation between a and b, with an edge weight of . Finally, hierarchical clustering or minimum spanning tree algorithms are used to analyze G. 指标 Perform tree-like processing to form a directed acyclic tree diagram; Using the R-type clustering method, when the correlation coefficient between two indicators is |r ab When |>0.85, it is considered highly redundant, and one of them is retained; those with a correlation degree of γ with all design parameters are removed. i Weak correlation indicators <0.3; obtain the remaining target indicators after purification; The tree-structured variable weighting method is used to allocate level coefficients according to the level l of the target indicator in the tree structure. Adjust position weights based on the number of sub-indicators. ; Greenness Comprehensive Assessment Scoring Model: Where L is the total number of layers in the tree diagram, node l is the set of indicators for the l-th layer, and s j Let G be the standardized score of the j-th target indicator, where G represents the comprehensive greenness assessment score, l represents the level number, and j represents the target indicator number. Both l and j are positive integers.

8. The method for three-dimensional collaborative design and greenness assessment of multi-source new energy power plants according to claim 7, characterized in that, The specific process for establishing the inverse mapping rule base between greenness evaluation indicators and three-dimensional design parameters is as follows: Extract each target index and extract the edge weights between each target index and each design parameter from the directed acyclic tree diagram; Sort the edge weights from largest to smallest, obtain the maximum edge weight corresponding to each target indicator, and determine whether the maximum edge weight corresponding to each target indicator is greater than the preset edge weight threshold. When the maximum edge weight corresponding to the target indicator is greater than the preset edge weight threshold, the design parameter of the maximum edge weight corresponding to each target indicator is taken as the target design parameter, the three-dimensional component ID associated with the target design parameter is extracted, highlighted in the three-dimensional model, and the target design parameter and the three-dimensional component corresponding to the target design parameter are output. When the maximum edge weight corresponding to the target indicator is less than or equal to the preset edge weight threshold, the parent node indicator of the target indicator in the tree diagram is taken as the new target indicator. The edge weights between the parent node indicator and each design parameter are extracted and sorted in descending order of edge weight. The maximum edge weight corresponding to the parent node indicator is obtained, and it is determined whether the maximum edge weight corresponding to the parent node indicator is greater than the preset edge weight threshold. The above process is repeated until an ancestor node with an edge weight greater than the preset threshold is found. The design parameter with the maximum edge weight corresponding to the ancestor node is taken as the target design parameter. According to the 3D component ID corresponding to the target design parameter, it is highlighted in the 3D model, and the target design parameter and the 3D component corresponding to the target design parameter are output. The target design parameters are sorted from largest to smallest according to their contribution weight to obtain the inverse mapping rule corresponding to the target indicator; the inverse mapping rules corresponding to each target indicator are summarized to construct a reverse mapping rule library between greenness evaluation indicators and three-dimensional design parameters.

9. The method for three-dimensional collaborative design and greenness assessment of multi-source new energy power plants according to claim 8, characterized in that, The specific process for adaptively correcting the key design parameters of the 3D model is as follows: When the overall greenness assessment score G is lower than the preset overall assessment threshold, the lowest-scoring indicator s will be... min As a marker indicator that needs to be traced; Marking index s min For indexing, query the reverse mapping rule base: if a rule with this indicator as the primary key exists in the rule base, directly extract the list of associated design parameters; If there is no directly matching rule in the rule base, the hierarchical tracing process is triggered: starting from the indicator, trace the parent node indicator upwards along the tree relationship graph until the ancestor node with a strong association rule is found. Then, output the target design parameters corresponding to the marked indicator or ancestor node and the three-dimensional components corresponding to the target design parameters, highlight them in the three-dimensional view, and then use the improved multi-objective particle swarm optimization algorithm to generate a correction scheme and make modifications.

10. A system implemented using the three-dimensional collaborative design and greenness assessment method for multi-source new energy power plants according to any one of claims 1-9, characterized in that, include: Model building module: Used to build a unified cloud-based 3D collaborative design platform, gather design resources from multiple disciplines, perform multi-disciplinary synchronous modeling, parameter linkage updates, cross-disciplinary automatic conflict verification, and construct a fully integrated 3D digital model. Model simulation module: Based on the global integrated 3D digital model, it is used to build five types of simulation sub-engines: layout, light and shadow resources, energy consumption and loss, power grid access and power flow, and construction conditions. It also establishes a data communication, linkage triggering and collaborative iteration mechanism between the simulation sub-engines through the built-in data bus to perform simulation optimization of the design scheme. Model evaluation module: Used to construct a comprehensive greenness evaluation scoring model for multi-source renewable energy coupled power plants, conduct quantitative evaluation of greenness, establish a reverse mapping rule library between greenness evaluation indicators and three-dimensional design parameters, reverse locate design defects based on the quantitative evaluation results of greenness, adaptively correct key design parameters of the three-dimensional model, and complete the iterative optimization of the design scheme.

Images