A building passive energy-saving strategy generation method and system

Abstract

The application discloses a kind of building passive energy-saving strategy generation method and system, it is related to passive energy-saving strategy generation technical field, including compiling and archiving strategy generation task book, solidification target caliber source, target hierarchical rule and global partition facade roof four layers hard constraint gate and conflict record field;According to task book, it is divided into strategy unit and solidifies account snapshot according to orientation shelter, enclosure structure, opening restriction, operation constraint and construction window consistency;Evidence chain is generated in unit granularity and executes sufficiency gate and benchmark backout;Establish configuration space and eliminate configuration that triggers hard constraint;Finally generate strategy package and screen sorting, output strategy combination, implement in stages and verify read-write results.The method disclosed in the application realizes strategy explainable, implementable and verifiable by target caliber and hard constraint gate solidification, evidence chain backout, mutually exclusive dependency alternative constraint and three-stage screening, reduces conflict rework and improves energy-saving and budget controllability.

Description

Technical Field

[0001] This invention relates to the field of passive energy-saving strategy generation technology, specifically to a method and system for generating passive energy-saving strategies for buildings. Background Technology

[0002] Current building energy conservation is evolving from optimizing individual components to a systematic decision-making process encompassing the entire building lifecycle. Research and engineering practice commonly employ methods such as climate zoning, verification of building envelope thermal parameters, solar radiation and shading analysis, and assessment of natural ventilation and daylighting, supplemented by energy consumption simulation and multi-objective optimization tools to compare and iterate parameters for passive energy-saving strategies. With the development of BIM, energy consumption monitoring, and operation and maintenance data platforms, data coupling between the building design and operation phases is continuously strengthening. The generation of energy-saving strategies is shifting from experience-driven to a management decision-making model that emphasizes both data-driven approaches and rule constraints, stressing unified target definitions, traceable constraints, and feasible and verifiable solutions, thus providing support for energy-saving retrofitting and operation management in complex building scenarios.

[0003] Existing passive energy-saving strategies often rely on single simulations or localized rules, lacking a unified and standardized approach to the source of objectives and evaluation criteria. This leads to inconsistencies in the assessment of comfort, lighting, ventilation, carbon emissions, and cost among different stakeholders, making it difficult to compare and verify the results. Furthermore, constraint handling frequently relies on post-hoc verification, lacking a layered, hard-constraint gating mechanism. This results in unfeasible combinations of strategies that may exist within the boundaries of approval, aesthetics, fire safety, cleanliness, and window closures. Moreover, existing methods typically handle internal building differences at the room or floor level with coarse granularity, failing to establish modifiable units based on consistency in orientation, enclosure structure, opening restrictions, and operational constraints. This makes it difficult to implement strategy recommendations across inconsistent objects. The lack of evidence data, quality grading, and standardized records of fallback paths also hinder the tracing of the sources of diagnostic and recommendation evidence. For strategy combination generation, existing technologies generally lack structured expression and constraint propagation of mutually exclusive, dependent and substitution relationships, making it difficult to form reachable alternative paths and converge to the minimum necessary strategy set under conflict conditions. In addition, the solution output is mostly limited to one-time recommendations, lacking a closed-loop mechanism that connects with phased implementation, milestone verification and write-back updates. Therefore, it is impossible to continuously accumulate evidence reliability and rule boundaries, making it difficult to achieve interpretable, implementable and verifiable passive energy-saving strategy generation for management decision-making. Summary of the Invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problems solved by this invention are: the existing methods for generating passive energy-saving strategies for buildings have problems such as inconsistent sources of objectives and indicators that are difficult to trace, lack of gating and mutually exclusive dependency substitution constraints for conflicts between hard constraints and scheme combinations, lack of closed loop for specification rollback and implementation verification when evidence data is insufficient, and how to generate interpretable, implementable and verifiable passive energy-saving strategy combinations and phased schemes under complex constraints and multi-objective conditions.

[0006] To address the aforementioned technical problems, this invention provides the following technical solution: a method for generating passive energy-saving strategies for buildings, comprising: compiling and archiving a strategy generation task book; registering the source of target criteria and target classification rules item by item; solidifying the four-layer hard constraint gate control and conflict record fields for global partitions, facades, and roofs; dividing the building into strategy units according to the task book based on orientation obstruction, enclosure structure, opening restrictions, operational constraints, and consistency of construction windows, and solidifying the ledger snapshot; generating an evidence chain record table at the strategy unit granularity; performing sufficiency gate control on key evidence and performing rollback processing according to the benchmarking and supplementary measurement list; establishing a configuration space in the evidence chain record table and first eliminating configurations that trigger hard constraints; constructing a mutual exclusion rule table, a dependency rule table, and a substitution rule table; mapping the gap diagnosis results to dimensional branch paths and converging to the minimum trigger set to limit the candidate combination boundary; generating strategy packages according to the minimum trigger set and sequentially completing the screening and management caliber sorting output; outputting the building passive energy-saving strategy combination and the phased implementation and verification write-back results.

[0007] As a preferred embodiment of the passive energy-saving strategy generation method for buildings described in this invention, the strategy generation task book compiled and archived includes target fields, priority fields, hard constraint fields, evaluation index caliber fields, and conflict record fields; the target fields are classified and solidified into three levels: safety and operational boundaries, indoor comfort and health, and carbon and economy, and the caliber source and applicable scope are registered item by item for overheat control, heat loss control, lighting and glare, availability of natural ventilation, carbon emission reduction, budget, construction period window, and impact of shutdown; the hard constraint fields are solidified into four levels of gate control items: global, zone, facade, and roof, and the source type and number of each gate control item are registered; when a candidate strategy or strategy package triggers any hard constraint conflict, a conflict record item containing the scope of the conflict object, the trigger item number, the evidence citation field, and the timestamp is generated.

[0008] As a preferred embodiment of the passive energy-saving strategy generation method for buildings described in this invention, the strategy unit division includes merging or splitting initial candidate areas based on modifiable consistency; modifiable consistency includes consistency of building envelope, orientation and shading, consistency of opening openability and ventilation restrictions, consistency of operational constraints, and consistency of construction windows and shutdown conditions; when any consistency dimension is not satisfied, splitting is triggered and the triggering dimension, difference field, source of difference evidence, and corresponding task book constraint item number are recorded; a unique identifier is assigned to the final strategy unit and a ledger snapshot is fixed, the ledger snapshot includes sub-tables of building envelope, openings, shading and self-shading, operational constraints, and construction windows, and the source type, formation time, and verification status of each field are marked.

[0009] As a preferred embodiment of the passive energy-saving strategy generation method for buildings described in this invention, the following steps are taken: The generation of an evidence chain record table at the strategy unit granularity includes generating evidence of climate boundaries, solar exposure and shading, heat transfer and infiltration, ventilation availability, daylighting and glare, operation and approval, and construction conditions. For each evidence field, the source type, formation method, applicable scope, and verification status are registered. Evidence sufficiency gating is applied to the set of key evidence fields, classifying the strategy unit evidence status as available, revertable available, and unavailable requiring supplementary testing. When a key field is missing or untraceable, it is reverted and supplemented according to the priority of the same building history, the same benchmark library in the same climate zone, internal organizational standards, or industry guidelines, and the reverted source and applicable boundaries are registered. Fields involving hard constraints are not allowed to be directly replaced by benchmark default values.

[0010] As a preferred embodiment of the passive energy-saving strategy generation method for buildings described in this invention, the step of establishing a configuration space in the evidence chain record table and pre-eliminating configurations that trigger hard constraints includes pre-converging the configuration space based on the evidence chain record table and hard constraint gating. The configuration space is set with configuration options according to the dimensions of envelope and airtightness, solar heat gain control, heat dissipation and natural ventilation, daylighting and glare control, and roof microclimate. Applicable conditions, preconditions, constraint association items, and references are registered in each configuration option. A mutual exclusion rule table, a dependency rule table, and a substitution rule table are established. Mutual exclusion rules are used to limit configuration combinations that cannot be simultaneously established within the same object range. Dependency rules are used to limit configuration linkage or precondition satisfaction relationships. Substitution rules are used to provide alternative paths in the same dimension or across dimensions when a configuration triggers a hard constraint or the dependency is not satisfied. Configuration options that trigger hard constraints are marked as unavailable and unavailable reason records are formed.

[0011] As a preferred embodiment of the passive energy-saving strategy generation method for buildings described in this invention, the following steps are included: constructing a mutual exclusion rule table, a dependency rule table, and an alternative rule table; mapping the gap diagnosis results to dimensional branch paths and converging a minimum trigger set to limit the candidate combination boundary; generating a gap diagnosis sheet based on the task book target classification and the evidence chain record table; mapping the gap diagnosis results to gap-dimensional branch paths; the gap diagnosis includes at least overheating risk gaps, heat loss risk gaps, insufficient lighting gaps, glare risk gaps, and natural ventilation availability gaps; and referencing corresponding evidence fields and verification status for each gap; when there is a systemic ban on opening windows or the ventilation dimension is gated and closed, the heat exhaust and ventilation dimensions are not triggered, and the alternative paths of solar heat gain control dimension and envelope and airtightness dimension are switched; based on the branch mapping, a minimum trigger set is generated to cover the main gaps; the minimum trigger set limits the required dimensions, trigger configuration range, and object range; and verification records are formed for mutual exclusion pre-checks, dependency satisfiability pre-checks, and alternative accessibility pre-checks.

[0012] As a preferred embodiment of the passive energy-saving strategy generation method for buildings described in this invention, the step of generating strategy packages based on the minimum trigger set and sequentially completing screening and management caliber sorting output includes generating candidate strategy packages based on the minimum trigger set and sequentially performing feasibility screening, robustness scenario consistency screening, and management caliber screening; feasibility screening includes hard constraint conflict checking, mutual exclusion conflict checking, dependency satisfaction checking, and mandatory gap coverage checking; robustness screening is based on a scenario set including extreme high temperature and high radiation, occupancy extension, shading changes, conservative window opening, and construction window changes, and performs constraint consistency and gap coverage stability checks on the strategy packages, and solidifies the implementation scope or alternative path condition entries through the strategy packages; management caliber screening generates cost, schedule window, shutdown impact, and approval risk indicator records according to the task book caliber and sorts and outputs them; and maps the selected strategy packages to component-level lists and phased milestones, solidifies the verification caliber according to the milestones, and writes back the verification results to update the evidence chain, configuration entries, and rule table versions.

[0013] Another objective of this invention is to provide a passive energy-saving strategy generation system for buildings. This system can divide buildings into strategy units based on the consistency of orientation shading, enclosure structure, opening restrictions, operational constraints, and construction windows according to the task specification, and solidify the ledger snapshot. It generates an evidence chain record table at the granularity of the strategy unit, performs sufficiency gating on key evidence, and performs rollback processing according to the benchmarking and supplementary measurement list. This solves the problems of current passive energy-saving strategy generation methods for buildings containing hard constraints, scheme combination conflicts, lack of gating, and mutual exclusion dependency substitution constraints.

[0014] As a preferred embodiment of the passive energy-saving strategy generation system for buildings described in this invention, it includes: a task book solidification module, a unit and evidence module, a configuration and trigger module, and a strategy generation closed-loop module; the task book solidification module is used to clarify the source of the target caliber, the target classification rules, and the four-layer hard constraint gating and conflict record caliber; the unit and evidence module is used to divide the strategy units according to the reproducibility consistency and solidify the ledger snapshot, generate the strategy unit evidence chain, and complete the evidence sufficiency gating and backtracking completion; the configuration and trigger module is used to establish and converge the feasible configuration space, form mutually exclusive dependency substitution rules, map the gap diagnosis to the dimensional branch, and converge the minimum trigger set to limit the combination boundary; the strategy generation closed-loop module is used to generate strategy packages according to the minimum trigger set, complete the screening and management caliber sorting, output the phased implementation plan, and verify and update the evidence and rule versions according to the milestone.

[0015] A computer device includes a memory and a processor, the memory storing a computer program, the processor executing the computer program as a step in implementing a method for generating a passive energy-saving strategy for buildings.

[0016] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of a method for generating a passive energy-saving strategy for buildings.

[0017] The beneficial effects of this invention are as follows: The passive energy-saving strategy generation method for buildings provided by this invention, by compiling and archiving a strategy generation task book, integrates the source of the target scope, target classification rules, and four-layer hard constraint gate control and conflict record fields for global partitions, facades, and roofs, achieving a unified adjudication benchmark and traceable basis for strategy generation. This ensures consistency in evaluation criteria under conditions of multiple participants and multiple indicators, thereby avoiding the problems of incomparable indicators and difficulty in verifying disputes. By dividing strategy units according to orientation obstruction, enclosure structure, opening restrictions, operational constraints, and consistency of construction windows, and solidifying the ledger snapshot, implementable alignment at the granularity of the renovation object is achieved. This is used to merge and manage areas with similar constraints and similar construction boundaries, thereby reducing implementation deviations caused by cross-object mismatches in strategies. By generating an evidence chain record table at the granularity of strategy units and performing evidence sufficiency gating and benchmark rollback processing, quality classification and missing information completion of key fact inputs are achieved. This allows for controlled progress and a clear supplementary measurement list even when evidence is insufficient, thereby improving the completeness of the basis for diagnosis and recommendation and the interpretability of review. By establishing a configuration space and pre-eliminating configurations that trigger hard constraints, a table of mutually exclusive, dependent, and alternative rules is constructed, achieving constraint propagation and pre-convergence of the feasible solution space. This is used to automatically limit the combination boundary and provide achievable alternative paths under conflict conditions, thereby avoiding rework and unimplementable combinations. By mapping gap diagnosis to dimensional branches and converging the minimum trigger set, minimum necessary coverage of strategy combinations is achieved, reducing ineffective superposition and excessive modification, thus controlling cost and schedule risks while satisfying gaps. By generating strategy packages according to the minimum trigger set and sequentially sorting and filtering by feasibility, robustness, scenario consistency, and management caliber, a three-level convergence from usability to robustness to decision-making is achieved, outputting interpretable strategy combinations and their phased implementation arrangements. Finally, through milestone verification caliber solidification and verification write-back updates, a closed-loop iteration of strategy generation and implementation verification is achieved, used to solidify evidence reliability and rule boundaries, thereby continuously improving the stability and management adaptability of subsequent passive energy-saving strategy generation. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying 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.

[0019] Figure 1 The first embodiment of the present invention provides an overall flowchart of a method for generating a passive energy-saving strategy for buildings. Detailed Implementation

[0020] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0021] Example 1, referring to Figure 1 As an embodiment of the present invention, a method for generating a building passive energy-saving strategy is provided, comprising: S1: Compile and archive the strategy generation task book, register the source of the target scope and the target classification rules item by item, and solidify the four-layer hard constraint gating and conflict record fields for the global partition facade and roof.

[0022] Furthermore, before generating passive energy-saving strategies for buildings, a strategy generation task book is first established to constrain the processing, and this task book is then archived as the sole rule benchmark for generation. The task book is compiled based on project initiation documents, management requirements from the owner or operation and maintenance unit, existing building operation records, site survey records, design and completion data, and applicable national or industry standard provisions. Each field in the task book must specify the data source type, formation time, and applicable scope, and be bound to the current strategy generation task number to form a version snapshot, ensuring that the same set of preconditions can be traced back to any subsequent strategy selection, combination, sorting, and phased implementation. The task book includes at least five categories of content: target field, priority field, hard constraint field, evaluation indicator caliber field, and conflict record field. The target field describes the target items that the strategy generation needs to meet and their measurement caliber; the priority field describes the order and irreversible rules between target items; the hard constraint field describes inviolable boundaries and their hierarchical effect; the evaluation indicator caliber field unifies the comparability of subsequent candidate strategy packages; and the conflict record field provides verifiable conflict objects and evidence when conflicts occur.

[0023] The target fields are organized from the management perspective of passive energy-saving retrofit decision-making, and include at least the following targets: overheat control target, heat loss control target, natural lighting target, glare constraint target, natural ventilation availability target, carbon emission reduction target, investment budget target, construction period window target, and shutdown impact target. To avoid incomparability due to inconsistent target definitions, the task specification specifies the evaluation criteria and recording methods for each type of target: Overheat control targets are based on the climatic boundary conditions of the project location, the building's usage period, and the allowable indoor temperature range. The allowable indoor temperature range can reference current building thermal environment and energy-saving standards or owner-defined indoor environmental requirements. Heat loss control targets are based on current building envelope data, airtightness records, and heating or thermal supply operation conditions. Current building envelope data is derived from BIM models, as-built drawings, material lists, and necessary on-site sampling verification records. Natural lighting and glare constraint targets are based on the lighting requirements of building functional zones, sensitive area settings, descriptions of existing lighting conditions, and applicable lighting or illumination standards. The criteria for judging insufficient lighting and glare sensitivity are written into the task specification as textual rules. Natural ventilation availability targets are defined based on the availability of openings. The criteria for opening are based on conditions, permitted opening periods, and safety and noise restrictions. The permitted opening periods and restrictions are derived from operation and maintenance management systems, site survey records, or written management requirements from the owner. The definition of carbon emission reduction targets is based on energy bills or energy consumption metering system records, and the energy type, statistical period, and source of carbon emission factors (locally published factors, industry guidelines, or internal organizational standards) must be clearly stated in the task book. The definitions of investment budget targets, construction period targets, and shutdown impact targets are based on project investment control documents, construction organization constraints, shutdown approval systems, and operation calendars. The budget field must clearly define the budget ceiling, the cost composition, and whether it includes construction measures such as scaffolding, suspended platforms, garbage removal, and temporary facility restoration. The construction period field must clearly define the start and end dates or periods for permitted construction and prohibited entry windows. The shutdown impact field must clearly define the maximum area that can be shut down, the permitted shutdown period, and a list of key areas that must remain operational.

[0024] The priority field is used to fix the hierarchical order among objectives and is written into the task specification in the form of objective hierarchical rules, so that the selection of any subsequent candidate strategies or strategy packages is subject to these rules. The objective hierarchical rules contain at least three levels: Level 1 is related to safety and operational boundaries, including fire evacuation continuity requirements, structural load-bearing and roof load limits, cleanliness and safety management requirements, restrictions prohibiting changes to facade morphology or making approval infeasible, and requirements that critical areas must not be shut down; Level 2 is related to indoor comfort and health, including overheat control, heat loss control, natural lighting and glare constraints, and meeting ventilation availability under permissible conditions; Level 3 is related to carbon and economic efficiency, including carbon emission reduction, investment budget allocation, and maintenance cost burden. The objective hierarchical rules are expressed irreversibly: any subsequently generated candidate strategies or strategy packages must not violate Level 1 objectives at the expense of satisfying Level 2 or Level 3 objectives, nor may they cause Level 2 objectives to deviate significantly from the task specification at the expense of satisfying Level 3 objectives; when there are conflicts between different objectives, they are adjudicated according to the objective hierarchical order, and a conflict record entry is required explaining the basis for the adjudication.

[0025] Hard constraint fields are used to express hard constraint gating rules. Hard constraint gating is fixed in the task specification in the form of a gating table, which clearly defines the level of action, the objects affected, and the source of the judgment criteria for the hard constraints. Hard constraints are divided into at least four levels: global hard constraints, zone hard constraints, facade hard constraints, and roof hard constraints. Global-level hard constraints include, but are not limited to, the requirement that fire evacuation width and passage continuity must not be obstructed, the building structure's load-bearing safety boundary must not be breached, and prohibition requirements stipulated in major safety management regulations. These constraints are based on fire protection design documents, structural inspection or assessment conclusions, and operational safety regulations. Zonal-level hard constraints include, but are not limited to, restrictions on window openings or limited window openings in specific areas such as clean areas or wards, restrictions on external openings and construction time in noise-sensitive areas, and a list of critical business areas that must maintain continuous operation. These constraints are based on operation and maintenance regulations, written requirements from the owner, or on-site inspection confirmation records. Facade-level hard constraints include, but are not limited to, restrictions on the exterior facade not being able to project outwards, not being able to alter the facade's appearance, and not being able to add shading components that would render approval infeasible. These constraints are based on planning approval conditions, historical preservation requirements, or the owner's facade management requirements. Roof-level hard constraints include, but are not limited to, the upper limit of roof load, restrictions on the inviolability of the roof waterproofing system, and the requirement that roof equipment maintenance access must not be obstructed. These constraints are based on structural data, roof waterproofing completion data, or operation and maintenance regulations. The wording of hard constraint gating rules must be clear: when a candidate strategy or strategy package triggers a hard constraint conflict at any level, the candidate will not be included in the subsequent sorting and phasing process, and a conflict record must be generated; the conflict record must include at least the conflict strategy name, the conflict object (corresponding strategy unit, zone or facade / roof number), the triggered hard constraint item number, the source type (e.g., a system document number, a site inspection record number or an approval condition clause), and the conflict determination timestamp.

[0026] The evaluation indicator caliber field is used to ensure comparability between subsequent strategy packages. The task book uniformly defines the statistical units, statistical periods, record fields, and data sources for cost, construction period, downtime impact, approval risk, comfort and lighting indicators, ventilation availability indicators, and carbon and energy consumption indicators. The cost caliber requires specifying the pricing scope used (whether materials, labor, and contingency costs are included) and specifying the reference source (company quotas, market information prices, historical project settlement data, or inquiry records). The construction period caliber requires specifying the construction window expression method (consecutive days or a set of workable periods) and its source (operational calendar, downtime approval rules, or owner-designated window). The downtime impact caliber requires specifying the affected area, affected period, and the source of maintaining operational requirements in key areas. The approval risk caliber requires specifying the basis for risk assessment (approval conditions, appearance requirements, property or management unit regulations). All the above caliber fields are fixed in the form of a task book version snapshot and this version number is referenced in subsequent outputs to ensure consistent caliber across different batches of strategy generation for the same project.

[0027] Archive and save the task book and its version snapshot, and generate a unique identifier corresponding to this strategy generation task. The identifier must include at least the task book version number, data source set number, generation timestamp, and applicable building identifier. Subsequent strategy unit division, evidence chain generation, configuration space convergence, gap diagnosis, combination generation, three-stage screening, phased planning, and verification write-back will all be processed with this task book version as the input premise. In addition, during each gating rejection or target conflict adjudication, verifiable record entries will be output according to the conflict record field template in the task book.

[0028] It should be noted that after completing the strategy generation task book and solidifying its target priority rules, constraint hierarchical gating rules, and indicator caliber field templates, the target building is divided into reproducible consistency strategy units according to the task book. Simultaneously, a snapshot of the passive element ledger corresponding to each strategy unit is established, ensuring that subsequent evidence chain generation, configuration space convergence, gap diagnosis, combination generation, and phased planning are all processed at the same granularity. The division of reproducible consistency strategy units is based on the principle that "within the same unit, the same set of constraint gatings can be accepted, the same evaluation caliber can be used, and the same type of passive configuration can be implemented within the same construction window." This avoids mixing areas with significantly different constraint conditions or construction methods in the same unit, which would lead to subsequent recommendations that are unfeasible or incomparable. The inputs for dividing the strategy unit should include at least the hierarchical constraint items and target field definitions that have been fixed in the task book, architectural design and as-built data (such as floor plans, elevations, sections, detailed drawings, and bills of materials), BIM models or as-built models, site survey records and sampling verification records, and management requirements for windows in the operation and maintenance system regarding opening, closing, and construction prohibition. For each type of input, the source type, formation time, and applicable scope should be recorded in the strategy unit ledger so that the basis can be located during review or verification.

[0029] The strategy unit division adopts a hierarchical and segmented merging and splitting process. First, initial candidate units are generated based on the building's spatial organization. Each floor is divided into several candidate zones according to functional boundaries, with corridors, core tubes, vertical shafts, fire compartment boundaries, and structural expansion joints serving as natural boundaries for initial segmentation, ensuring that candidate units have clear physical and management boundaries. Subsequently, a modifiability consistency assessment is performed on the initial candidate units, with assessment dimensions categorized into at least five types: structural and construction consistency, orientation and external boundary consistency, opening and ventilation usability consistency, operational constraint consistency, and construction window and interference consistency. Verifiable record fields are provided for each consistency assessment category, forming a consistency assessment record table, which serves as the direct basis for subsequent merging or splitting.

[0030] Regarding structural and construction consistency, the consistency of the exterior walls, roofs, floors, and key node constructions associated with candidate units is verified. For exterior walls, at least the wall system type (e.g., masonry, shear wall, curtain wall, sandwich panel, etc.), insulation system type (internal insulation, external insulation, sandwich or no insulation), main materials and thickness range, and thermal bridge sensitive node types (e.g., exposed beams and columns, window openings, balcony slabs penetrating walls, etc.) are verified. For roofs, at least the roof type (flat roof, pitched roof, green roof, or equipment roof), waterproofing system and insulation layer location, roof load-bearing capacity, and modification restrictions are verified. For floors and adjacent outdoor air floors, at least the presence of elevated structures, basement roofs, or other special boundary conditions is verified. The above construction information is sourced from as-built drawings, detailed node drawings, material lists, and on-site sampling verification. On-site sampling verification must at least record the sampling location, sampling method, and verification conclusions. If two candidate units have substantial differences in their exterior wall or roof construction systems, or if the differences in thermal bridge node types and densities lead to significantly different subsequent passive modification paths, they should not be merged into the same strategy unit. Instead, they should be split into different strategy units and each should be recorded separately.

[0031] Regarding the consistency of orientation and external boundary, orientation and external boundary type identifiers are established for each candidate unit's facade or roof segment. Orientation identifiers include at least a primary orientation and a secondary orientation, while external boundary types include at least proximity to outdoor air, internal corridors, atriums, and adjacent building shading sides. Shading level descriptions are generated based on site surveys, GIS, or surrounding building information, recording whether significant shading occurs during periods of high irradiance, whether self-shading components (eaves, balconies, etc.) exist on the facade, and the approximate window-to-wall ratio range. If two candidate units have different primary orientations or significantly different external boundary types, resulting in inconsistent solar heat gain control, shading configuration selection, and parameter ranges, they should not be merged. If they have the same primary orientation and similar shading levels and window-to-wall ratio ranges, they can proceed to the next consistency assessment category.

[0032] Regarding the consistency of openings and ventilation availability, the openability, openable proportion range, opening method, and safety restrictions of window and door openings within candidate units are verified. Simultaneously, the zoning constraints in the task specification, such as prohibited window openings, restricted window openings, noise sensitivity, and cleanliness requirements, are also checked. Recorded fields must include at least whether window opening is permitted, permitted opening times, whether convection ventilation is permitted (e.g., whether there are opposing openings or connecting paths), whether nighttime heat exhaust is restricted, and the reasons for the restrictions (safety management, noise control, cleanliness management, or other institutional requirements). If two candidate units have substantial differences in window opening permissions and times, openability, or ventilation organization possibilities, they must not be merged to avoid contradictory situations where some ventilation dimensions are available and some are unavailable within the same strategy unit. If window opening permissions and restrictions are consistent, they can be merged for the next consistency assessment.

[0033] Regarding the consistency of operational constraints, the hierarchical gating rules of the task book are used as a benchmark to verify whether candidate units belong to critical areas that must operate continuously, whether there are restrictions on entry during specific periods, whether there are restrictions on facade approval or style, and whether there are restrictions on construction impacts such as dust, noise, vibration, and light pollution. The consistency of operational constraints requires that at least the constraint items be aligned: two candidate units are only allowed to be merged if the set of constraint items is consistent or the differences do not affect the implementation of similar configurations; if there are differences, they should be separated, and the reason for separation should record which constraint triggered it and the source of the constraint (such as management system document number, approval condition clause, or written requirements from the owner).

[0034] Regarding the consistency between construction windows and disruptions, the feasible construction windows, usable shutdown areas, and shutdown periods for candidate units are checked. Records are kept of whether scaffolding or suspended platforms are permitted, whether there are restrictions on access organization or material stacking, and whether facade or roof work is permitted in the same phase. If the construction windows and shutdown conditions of two candidate units are significantly different, resulting in similar configurations being unable to be implemented in the same phase or significant differences in implementation organization, they must not be merged to avoid subsequent phase plans being unable to be compiled using strategic units as the basic unit.

[0035] After completing the above five types of consistency checks, candidate units are merged or split to form the final list of strategy units. The merging rule is: merging is allowed only if structural and construction consistency, orientation and external boundary consistency, opening and ventilation availability consistency, operational constraint consistency, and construction window and interference consistency are simultaneously satisfied. The splitting rule is: splitting is triggered if any consistency check is not met, and the specific dimension of the non-compliance, the triggering difference field, the evidence of the difference source, and the corresponding task book constraint item number must be specified in the strategy unit partitioning record. After merging or splitting, each strategy unit is assigned a unique identifier, which must at least include the building number, floor or area number, main orientation number, and version number, and this identifier will be used as the primary key in all subsequent record tables.

[0036] After the strategy unit is determined, a snapshot of the passive element ledger corresponding to each strategy unit is established for subsequent evidence chain generation and configuration space convergence. The ledger snapshot includes at least five parts: the enclosure structure sub-table, the opening sub-table, the shading and self-shading sub-table, the operational constraint sub-table, and the construction window sub-table. The enclosure structure sub-table records the type, material and thickness range, thermal bridge node type and distribution description, airtightness status record and its source for the exterior walls, roof, ground and key node structures; the opening sub-table records the orientation of windows and doors, window-to-wall ratio range, glass type or available parameter description, openability and opening restrictions and time periods; the shading and self-shading sub-table records the existence, positional relationship and modification restrictions of existing eaves, balconies, louvers and other components; the operational constraint sub-table records the constraint items such as prohibited windows, cleanliness, noise sensitivity, continuous operation, facade approval and their basis sources; the construction window sub-table records the permitted construction dates or times, prohibited windows, usable area and time of shutdown, construction access and material stacking restrictions, etc. All of the above ledger fields must specify the source type (as-built drawings, BIM, bill of materials, site survey, policy documents or approval conditions), the creation time, and the verification status. The verification status should be divided into at least four categories: from official documents, from on-site verification, from operation and maintenance policies, or from benchmarking default values, so that consistent rollback rules can be implemented when insufficient evidence is found later.

[0037] The strategy unit list, strategy unit division record table, and passive element ledger snapshot are bound and archived with the task book version number to form the input snapshot version for this strategy generation. This version number is then used as a unified input reference for subsequent evidence chain generation, configuration space convergence, gap diagnosis, and combination generation.

[0038] S2: Based on the task description, the building is divided into strategy units according to the consistency of orientation obstruction, enclosure structure, opening restrictions, operation constraints and construction windows, and the ledger snapshot is solidified. An evidence chain record table is generated at the granularity of the strategy unit, and sufficiency gating is performed on key evidence and rollback is carried out according to the benchmarking and supplementary measurement list.

[0039] Furthermore, after archiving the strategy unit list and passive element ledger snapshots, an evidence chain record table is generated for each strategy unit. During the generation process, evidence sufficiency gating and rollback are implemented to ensure that subsequent gap diagnosis, configuration space convergence, combination generation, and phased planning all rely on verifiable and traceable evidence fields as input. The evidence chain record table uses the unique identifier of the strategy unit as the primary key, establishing a binding relationship with the task book version number and the strategy unit ledger snapshot version number. Each evidence field in the record table specifies its source type, formation time, collection or derivation method, applicable scope, and verification status, thereby ensuring that the evidence fields for the same strategy unit remain consistent across different batches of strategy generation. The evidence chain field set consists of at least the following: climate boundary evidence, solar exposure and shading evidence, heat transfer and infiltration evidence, ventilation availability evidence, daylighting and glare evidence, operation and approval evidence, and construction condition evidence. The sources of each field are based on meteorological data of the project site, surrounding environment and shading information, current status data of building envelope and openings, operation and maintenance system and approval conditions, and on-site survey and sampling verification records. Default values ​​may be used when evidence is insufficient, but the source and applicable boundaries of the default values ​​must be indicated in the rollback record.

[0040] Climate boundary evidence describes the temporal distribution characteristics of the climate conditions in which the strategy unit is located. Fields must include at least the local climate zoning or meteorological station identification, statistical period, temperature and humidity characteristics for key months or key periods, solar radiation intensity distribution characteristics, prevailing wind direction and speed characteristics, and diurnal temperature range characteristics. Climate data sources can include typical meteorological year data published by national or local meteorological departments, recent observational statistics, or existing operational monitoring data from the project location. The evidence field must specify the dataset name, the range of years covered, or the statistical period, as well as the resolution caliber used for strategy generation (monthly, daily, or time period). When the project provides building management system or energy consumption monitoring platform records that allow for the extraction of outdoor environmental data, these records are permitted as a supplementary source, but the equipment source, sampling period, and data integrity check methods must be specified.

[0041] Evidence of solar exposure and shading is used to describe the exposure characteristics of the strategic unit's external boundary during periods of high irradiance. Fields must include at least the main orientation of the corresponding facade or roof of the strategic unit, the type of external boundary, the window-to-wall ratio range, the level of external shading, and a description of the existence of self-shading components and typical shading relationships. Shading information can originate from GIS data, surrounding building information, on-site survey and measurement records, or design model derivations. The evidence field must specify the source type and formation method of the shading information, such as from publicly available geographic data, on-site survey point records, or model calculation results. When shading information comes from an on-site survey, the survey date, observation point or observation range, and the type of measurement tool or recording method used should be recorded. The window-to-wall ratio range and opening distribution are derived from as-built drawings, BIM models, or on-site verification records. If the sources are inconsistent, the on-site verification conclusion takes precedence, and a description of the discrepancies should be recorded in the verification status field.

[0042] Heat transfer and permeation evidence is used to describe the thermal weaknesses and air permeation risks of the building envelope. Fields must include at least the construction type and insulation system of the exterior walls, roof, and floor; descriptions of the types and distribution of thermal bridges at key nodes; the current airtightness level or available permeation risk indicators; and descriptions of potential leakage paths. Construction information is derived from the building envelope construction sub-table in the strategy unit ledger snapshot and can be corroborated by material lists, detailed node drawings, and on-site sampling verification records. If an airtightness test report exists, the report conclusion should be used as the source, and the report number, test date, and test scope should be recorded. If no test report exists, on-site inspection records of visibility checks on door and window gaps, wall penetrations, access panels, pipe wells, etc., can be used as risk indicators, and the inspection date and scope should be recorded. For cases using default values ​​or experience-based grading, it must be indicated in the evidence field that it is a risk indicator rather than a test conclusion, and the source of the default value must be indicated as a benchmarking library or internal organizational specifications.

[0043] Evidence of ventilation availability describes whether the natural ventilation configuration meets the implementation boundaries. Fields include at least the openability of openings, permitted window opening times, the feasibility of forming convective ventilation paths, the availability of nighttime heat dissipation and the reasons for limitations, the impact of safety and noise constraints on window opening, and items prohibiting window opening due to cleanliness or management regulations. The sources of ventilation-related fields are based on maintenance regulations, written requirements from the owner, site survey records, and opening ledgers. Permitted window opening times and prohibited window opening items must reference the corresponding policy documents or management requirement item numbers. If the information can only be obtained through on-site interviews or experience-based judgment, the field should be marked as requiring review and a rollback should be triggered to avoid subsequent gap diagnosis errors triggering ventilation dimensions. The assessment of the feasibility of forming convective paths should clearly state the basis used, such as whether there are opposing openable openings in the floor plan, whether there are connectable ventilation corridors or openable interior doors. If the basis comes from drawings or BIM, the referenced drawing page number or model version number should be recorded; if it comes from on-site observation, the observation scope and time should be recorded.

[0044] Evidence regarding daylighting and glare is used to describe the risk of insufficient daylighting and glare sensitivity of the strategic unit. Fields must include at least the source of illuminance requirements corresponding to the functional type, an indication of the spatial distribution of insufficient daylighting, an indication of glare-sensitive areas, a description of the impact of window location and shading on daylighting, and constraints that shading configurations may lead to a decrease in daylighting. The source of daylighting requirements can be current daylighting or lighting standard provisions, or functional requirement documents from the owner or operator; the source of the provisions or the document basis must be specified. Historical complaints, maintenance records, or on-site feedback records regarding insufficient daylighting and glare sensitivity are allowed as supplementary evidence, but the source and date of these records must be specified. If no monitoring data is available, daylighting evidence can be generated using rule-based discrimination based on window-to-wall ratio, shading level, and functional requirements; however, the field must be marked as the rule discrimination result and the source and applicable boundaries of the rule must be recorded.

[0045] Operational and approval evidence is used to concretize the hard constraints of the task statement at the strategy unit level. Fields must include at least continuous operation area markers, shutdown restrictions and shutdown approval requirements, facade approval and style restrictions, construction noise and dust restrictions, and management items related to safety or cleanliness. The sources of these fields must be policy documents, approval conditions, or written requirements from the owner, and document numbers or clause references must be recorded. If only verbal requirements or interview information are available, the verification status should be marked as requiring review, triggering a rollback process. Subsequent strategy packages must not use this interview information as the final gate control basis until archival written evidence is formed.

[0046] Construction condition evidence is used to describe the construction organization boundaries of the strategy unit. Fields include at least the workable window, restricted window, material access and stacking restrictions, scaffolding or suspended platform availability, roof access occupancy restrictions, and maintenance requirements. The source of the workable window can be the operation calendar, shutdown approval system, or project construction organization constraint documents, and the start and end dates or time period sets should be recorded. The source of scaffolding or suspended platform availability can be site surveys and safety management systems. If safety permits are involved, the source of the permit conditions must be recorded.

[0047] After the evidence chain fields are generated, an evidence sufficiency gating is performed on each strategy unit. The evidence sufficiency gating is based on a set of key fields, which includes at least the basic fields for climate boundaries, solar exposure and shading, building envelope and airtightness risks, operational hard constraints, and fields directly related to the target gap diagnosis. The gating rules are as follows: if any key field is missing, its source is untraceable, or its verification status requires review and lacks alternative evidence, the strategy unit cannot proceed to the normal path of subsequent gap diagnosis and must first undergo rollback or be marked as requiring supplementary testing. If the key fields are complete and their sources are traceable, it is allowed to proceed to the next stage, and its evidence sufficiency status is marked as available. To avoid the overall process being stalled due to missing fields, the gating categorizes the strategy unit's evidence status into at least three types: available, available after rollback, and unavailable requiring supplementary testing. "Available" indicates that the key fields are complete and their source is traceable; "Available after rollback" indicates that the key fields are missing but have been completed through rollback and the rollback source is clear; "Unavailable after supplementary testing" indicates that the key fields are missing and cannot be completed through rollback or that rollback would cause gating distortion.

[0048] Rollback processing is used to complete fields and control risks when evidence is insufficient. Its execution order and source requirements are fixed in the evidence chain record table. Rollback source priorities include at least: historical monitoring and maintenance records of the same building or park, benchmarking data of similar buildings in the same climate zone, internal engineering experience databases or standardized parameter tables, and publicly released industry guidelines or typical value tables. Rollback processing must record the rollback field name, rollback value or rollback level, rollback source type, rollback basis number or document source, rollback applicable boundaries and validity period, and mark the field's verification status as rolled back. For fields completed by rollback, their uncertainty markers should be retained in subsequent strategy package generation and sorting for scenario consistency checks during the robustness screening phase. Simultaneously, when a rollback field involves hard constraint items (such as no-window or approval restrictions), direct replacement with benchmark default values ​​is not allowed. Archiveable policy documents or written requirements must be obtained as the source; otherwise, the field can only be marked as requiring review, and the relevant configuration dimensions should be closed or restricted to conservative configurations that do not touch the constraint to avoid outputs becoming unimplementable due to incorrect assumptions.

[0049] For strategy units that are unavailable and require supplementary testing, a supplementary testing list should be generated in the evidence chain record table. The supplementary testing list should at least include the name of the missing field, the recommended supplementary testing method, the source of responsibility for the supplementary testing (such as property management, operation and maintenance, or project site), and the record entry number that needs to be updated after the supplementary testing is completed. Supplementary testing methods may include on-site surveying of obstruction information, sampling and verification of the building envelope structure layers, obtaining supplementary energy consumption bills or monitoring data, obtaining supplementary management system texts, etc., but the supplementary testing list is only used to clarify the gap. Before the supplementary testing is completed, the strategy unit is only allowed to enter conservative paths that do not depend on the missing field in subsequent processing flows. For example, the ventilation dimension should not be triggered if the prohibition on opening windows is unclear, and the facade modification-related configuration should not be triggered if the approval restrictions are unclear. All such conservative restrictions should be written into the evidence chain record table in the form of record entries.

[0050] After completing the sufficiency of evidence gating and rollback, the evidence chain record table, gating status markers, rollback records and supplementary test list, along with the task book version number and ledger snapshot version number, will be archived together. The evidence chain record table version number will be used as a unified input reference for subsequent configuration space pre-convergence, target gap diagnosis, strategy package combination generation, and phased plan preparation.

[0051] It should be noted that after completing the evidence chain record table for each strategy unit and performing evidence sufficiency gating and rollback processing, a passive strategy configuration space is established for each strategy unit based on the hard constraint gating rules, target classification rules, and available or rollback-available evidence fields in the evidence chain record table that are already fixed in the task book. Simultaneously, mutual exclusion rule tables, dependency rule tables, and substitution rule tables are generated, ensuring that subsequent gap diagnosis, candidate generation, combination screening, and phased compilation are all conducted within a converged and traceable feasible space. The configuration space is established using the unique identifier of the strategy unit as the primary key, and is bound and archived with the task book version number, strategy unit ledger snapshot version number, and evidence chain record table version number. Each configuration option entry specifies the applicable scope, applicable conditions, preconditions, constraint associations, and necessary record fields, ensuring that any subsequent removal, substitution, or linkage can be traced back to a clear entry and source of evidence.

[0052] The configuration space is organized according to the mechanism dimension of passive modification, including at least the dimensions of envelope and airtightness, solar heat gain control, heat dissipation and natural ventilation, lighting and glare control, and site and roof microclimate. Each dimension has several configuration options, described in itemized terms based on feasible engineering practices, without using abstract concepts to replace engineering boundaries. Configuration options in the envelope and airtightness dimension should at least distinguish between types such as external wall insulation enhancement, roof insulation enhancement, joint thermal bridge treatment, and door and window airtightness improvement and gap sealing, and provide applicable condition fields for different structural systems, such as recording the modification boundaries for masonry exterior walls, shear wall exterior walls, and curtain wall systems. Configuration options in the solar heat gain control dimension should at least distinguish between external shading component types, shading adjustability categories, and window light and heat performance improvement as boundary descriptions for passive modification options, as well as enhancement methods utilizing existing self-shading components, and record in the items whether it involves facade overhang, changes to the architectural outline, and approval conditions. Configuration options in the heat dissipation and natural ventilation dimension should at least distinguish between transitional... The configuration options for seasonal convection ventilation, nighttime heat dissipation, and ventilation under limited window opening conditions should include preconditions such as permitted window opening times, prohibited window opening items, and safety and noise restrictions. For the lighting and glare control dimension, configuration options should at least differentiate between lighting enhancement, glare suppression, and synergistic lighting and glare control methods, and include entries for glare-sensitive areas, sources of illuminance requirements, and constraints on the impact of shading on lighting. For the site and roof microclimate dimension, configuration options should at least differentiate between roof greening, high-reflectivity roofs, external shading greening, and permeable evaporative cooling methods, and include preconditions such as roof load limits, waterproofing system limitations, and equipment maintenance access limitations. Each configuration option entry must include a source and basis field, which should at least include applicable standard provisions or internal organizational specifications, existing engineering practice records, and reference points associated with the strategy unit ledger or evidence chain field. Reference points can be represented by drawing page numbers, model version numbers, institutional document numbers, or site survey record numbers for verification purposes.

[0053] After the configuration space is itemized, hard constraint pre-convergence is first performed. This involves associating and matching the hierarchical hard constraint gating rules in the task book with operational and approval evidence, ventilation availability evidence, and roof load and waterproofing limitation evidence in the evidence chain. Configuration options that trigger hard constraint conflicts are directly marked as unavailable and removed from the optional set of that strategy unit. During removal, an unavailability reason record is generated, which includes at least the configuration option number, conflicting object, triggered hard constraint entry number, evidence chain reference field, decision timestamp, and decision person or decision rule identifier. After hard constraint pre-convergence, the configuration space enters a feasible state. At this point, a mutual exclusion rule table, a dependency rule table, and an alternative rule table are generated, and these three types of rules are used as mandatory constraint inputs for subsequent combination generation and filtering.

[0054] The mutual exclusion rule table describes configuration combinations that cannot be simultaneously established on the same object within the same strategy unit. The establishment of mutual exclusion relationships is based on physical space conflicts, engineering logic conflicts, and management constraint conflicts, and the scope of mutually exclusive objects is clearly defined in the rule table. Physical space conflict mutual exclusion is used to handle situations where only one type of occupier can be implemented at the same facade or roof location. For example, roof greening and high-reflective coatings are mutually exclusive options within the same roof area. Similarly, two types of shading components at corresponding locations in the same window opening are mutually exclusive if they conflict in terms of installation space or window opening function. Engineering logic conflict mutual exclusion is used to handle situations where selecting one configuration on the same target path excludes another. For example, when a window system modification option and a shading component option are incompatible in terms of process path or maintenance conditions, they must be set as mutually exclusive. Management constraint conflict mutual exclusion is used to handle situations where certain configuration combinations cannot occur simultaneously due to approval or aesthetic requirements. For example, combinations that require maintaining aesthetic integrity while introducing cantilevered components on the same facade are deemed incompatible and must be set as mutually exclusive. The record fields of the mutual exclusion rule table should include at least the mutual exclusion pair number, the set of mutual exclusion configuration option numbers, the scope of mutual exclusion objects, the mutual exclusion basis type, and the referenced evidence chain field or task constraint entry number, to ensure that the mutual exclusion relationship is not an empirical description but has a traceable basis.

[0055] The dependency rule table describes the relationships between configurations that require linkage or the fulfillment of preconditions to be enabled. Dependencies are based on passive mechanism links and management constraints, and the dependency direction and conditions must be clearly stated in the rule table. The dependency direction indicates that enabling one configuration requires the simultaneous activation of another configuration, or that a certain precondition field must be met. The dependency condition field must reference verifiable field items from the evidence chain record table. For example, when a daylight enhancement configuration may cause glare risk in a glare-sensitive area, the daylight enhancement configuration must depend on the glare suppression configuration and form a mandatory linkage; when an airtightness improvement configuration may change infiltration ventilation conditions, if the strategy unit has evidence of permitted window opening times and ventilation availability, the ventilation organization configuration must depend on the permitted window opening time field and the safety restriction field and form a time-constrained linkage; when an external shading configuration may affect fire evacuation or window opening, the external shading configuration must depend on the fulfillment of fire boundary and opening function boundary conditions, and trigger alternative rules when the conditions are not met. The record fields of the dependency rule table shall include at least the dependency pair number, the primary configuration option number, the secondary configuration option number or the precondition field number, the dependency condition description, the dependency basis type, and the referenced evidence chain field or task book entry number. The verification status of the dependency condition must be marked as available or rollback available. If it is unavailable and requires supplementary testing, the dependency link shall not be enabled.

[0056] The alternative rule table describes how to select alternative paths within the same or different dimensions without violating hard constraints when the preferred configuration becomes unavailable due to hard constraint gating or unmet dependency conditions. This ensures that policy triggering after gap diagnosis can continue to converge to the implementable set. The establishment of alternative rules is constrained by the goal hierarchy rule; that is, alternatives must not compromise higher-level goals at the expense of satisfying lower-level goals, and alternatives should maintain coverage of the same gap type as much as possible. Alternative rules include at least two categories: same-dimensional alternatives and cross-dimensional alternatives. Same-dimensional substitution is used to find alternatives that do not trigger constraints within the same dimension of the configuration space. For example, when the facade cannot be extended or approval restrictions make external shading configurations unavailable, they can be replaced by shading configurations that do not change the facade outline or by modifying the window system's light and heat parameters within the solar heat gain control dimension. When roof load restrictions make roof greening unavailable, they can be replaced by high-reflectivity roofs or external shading measures within the roof microclimate dimension. When window restrictions make the ventilation dimension unavailable, all ventilation configurations can be closed within the heat dissipation and ventilation dimensions, and ventilation can be replaced by a combination of solar heat gain control and building envelope airtightness dimensions to compensate for the overheating gap. Cross-dimensional substitution is used to cover the gap when a dimension is completely closed or the main configurations are unavailable, through combinations of configurations in other dimensions. For example, after the ventilation dimension is closed, the risk of overheating can be reduced by a combination of shading and building envelope enhancement, or when the lighting target is limited, a synergistic solution of glare suppression and local lighting enhancement can be used to replace large-scale lighting enhancement. The record fields of the substitution rule table shall include at least the trigger condition number, the number of the substituted configuration option, the set of substitution configuration option numbers, the substitution path type, the scope of applicable objects for substitution, the substitution basis and reference fields, and the risk warning field after substitution. The trigger condition must be clearly defined as being triggered by a hard constraint conflict, a dependency condition not being met, or a conservative substitution triggered by evidence that is available for rollback. The substitution basis must reference the task book target level entry number and the evidence chain field.

[0057] After generating the three rule tables—mutual exclusion, dependency, and substitution—a rule consistency check is performed on each strategy unit, forming a snapshot version of the optional set together with the rule tables and configuration space entries. The rule consistency check includes at least three checks: mutual exclusion closure, dependency satisfiability, and substitution reachability. The mutual exclusion closure check prevents conflicts during combination generation due to the omission of mutually exclusive relationships within the same object scope. The dependency satisfiability check confirms that dependency condition fields are available or fallback available in the evidence chain without violating hard constraints. The substitution reachability check confirms that when a critical configuration is unavailable, there is at least one alternative path covering the same gap type, and that the alternative path does not violate the target hierarchical rules. After the check is completed, the optional set of configuration space, mutual exclusion rule tables, dependency rule tables, substitution rule tables, and rule consistency check records are bound and archived with the task book version number and evidence chain version number. This version number serves as the sole input reference for subsequent gap diagnosis triggering, minimum dimension set generation, combination generation, and three-stage filtering.

[0058] S3: Establish a configuration space in the evidence chain record table and first eliminate configurations that trigger hard constraints. Construct a mutual exclusion rule table, a dependency rule table, and an alternative rule table. Map the gap diagnosis results to dimensional branch paths and converge to the minimum trigger set to limit the candidate combination boundary.

[0059] Furthermore, after completing the feasible convergence and archiving versions of the strategy unit configuration space and its mutual exclusion rule table, dependency rule table, and substitution rule table, a target gap diagnosis is performed for each strategy unit. The gap diagnosis results are then mapped to the optional configuration space of that strategy unit, forming a gap-dimensional branch path and a minimum trigger set. This ensures that subsequent candidate strategy package generation is only performed within the dimensions and configurations necessary to cover the main gaps, avoiding unconstrained expansion that leads to strategy stacking and unimplementability. The gap diagnosis uses the target field definitions, target hierarchical rules, and priority order fixed in the task book as the primary inputs, evidence fields marked as available or fallback available in the strategy unit evidence chain record table as the factual basis, the set of optional configurations that have completed hard constraint pre-convergence in the configuration space as the feasible solution space, and mutual exclusion, dependency, and substitution rule tables as constraints. Each gap conclusion output by the gap diagnosis must reference at least one corresponding evidence field and specify the evidence source type and verification status to ensure the traceability of the gap conclusions and the ability to locate the evidence chain record table entries during review or verification.

[0060] The gap diagnosis process first establishes a diagnostic sequence based on the target hierarchy of the strategy units. Prioritizes diagnosing constraint gaps related to safety and operational boundaries, followed by gaps related to indoor comfort and health, and finally gaps related to carbon and economics. Constraint gaps include, but are not limited to, restrictions on window opening or opening restrictions, prohibition of external facade extensions or approval infeasibility, roof load and waterproofing limitations, and the prohibition of shutdown in critical areas. The diagnostic results do not trigger modification configurations but are written into the gap diagnosis form as gating conditions for subsequent dimensional opening / closing, contraction of available configuration ranges, and alternative pathways. Comfort and health-related gaps include at least overheating risk gaps, heat loss risk gaps, insufficient lighting gaps, glare risk gaps, and unusable natural ventilation gaps. Each gap has clearly defined categorization fields and criteria in the task book. Carbon and economic-related gaps include at least carbon emission reduction target gaps, budget ceiling and cost affordability gaps, construction period window gaps, and shutdown impact gaps. The diagnostic conclusions are used for subsequent strategy package prioritization and phasing, but must not exceed high-level targets and hard constraint gating.

[0061] When generating a gap diagnosis report, evidence citation and judgment rules are applied separately for each type of gap. For overheating risk gaps, at least climate boundary evidence and solar exposure evidence are cited, combined with usage period and zoning requirements to describe the gap severity. For heat loss risk gaps, at least building envelope evidence and airtightness risk evidence are cited, combined with heating or cooling operation conditions to describe the gap severity. For insufficient lighting and glare risk gaps, at least the source of functional illuminance requirements, lighting and glare evidence, and obstruction and window distribution are cited to describe the gap severity. For unusable natural ventilation gaps, at least the openability of openings, permitted window opening times, safety and noise restrictions, and cleanliness or management regulations prohibiting window opening are cited, and these are categorized into institutional unavailability and conditional unavailability. Institutional unavailability refers to situations where there is a clear legal basis for prohibiting window opening and it cannot be opened at any time, while conditional unavailability refers to situations where it is restricted only at certain times or in certain areas. For carbon and economically related gaps, at least the source of energy consumption bills or monitoring records, budget control documents, shutdown approval procedures, or operating calendars are cited to describe the gap. If a gap involves a key evidence field that is available for rollback, then the source of uncertainty is marked in the gap diagnosis form, and conservative restrictions are introduced in the subsequent trigger set, that is, configurations that do not depend on the uncertain field are preferred or the trigger range is reduced.

[0062] After the gap is determined, a gap-dimensional branch mapping is executed. Each type of gap is mapped to the mechanism dimension of the configuration space according to a preset mapping relationship. During the mapping process, mutual exclusion, dependency, and substitution rules are applied simultaneously to form executable branch paths. The branch mapping includes at least the following basic branch logic: When the overheating risk gap is significant, the solar thermal control dimension is triggered first, and within this dimension, the configuration range that can cover the high irradiance exposure period is selected first; when the evidence chain shows that the thermal resistance of the enclosure is weak or the airtightness risk is significant and overheating and heat loss exist simultaneously, the enclosure and airtightness dimensions are allowed to be triggered in conjunction to compensate for the gap from the heat transfer and permeation paths; when the heat loss risk gap is significant, the enclosure and airtightness dimensions are triggered first, and the nodal thermal bridge treatment configuration is written as an optional linkage item into the trigger range; when the insufficient lighting gap is significant, the lighting enhancement configuration range in the lighting and glare control dimensions is triggered, but if the same strategy unit If there is evidence of glare sensitivity or a glare risk gap, then the daylighting enhancement configuration must rely on the glare suppression configuration, and the dependency relationship forms a mandatory linkage branch based on the dependency rule table. When the natural ventilation unavailability gap is institutionally unavailable, the heat dissipation and natural ventilation dimensions are completely closed, and any subsequent triggering must not use the configuration of this dimension to compensate for the overheating risk gap, but should switch to the alternative path of the solar heat gain control dimension and the building envelope and airtightness dimension. When the natural ventilation unavailability gap is conditionally unavailable, the ventilation configuration is only allowed to be triggered within the permitted window opening time and safety limit range recorded in the evidence chain, and the time constraint field is fixed in the triggering range, making the ventilation configuration a restricted trigger item. The triggering of the roof microclimate dimension adopts a constraint-priority branch: when evidence of roof load or waterproofing limitations indicates that roof greening is unavailable, this configuration is replaced by an equivalent configuration of high-reflectivity roofs or external shading measures, with the replacement based on the referenced replacement rule table; when evidence of roof boundary exposure indicates high roof radiation exposure without touching hard constraints, the roof microclimate dimension is allowed to be triggered as a supplementary branch for the overheating gap, but this branch may not replace the main branch of the solar heat gain control dimension unless the main branch is closed by hard constraints and an alternative path has been recorded.

[0063] After obtaining the dimensional trigger range corresponding to each gap, a minimum trigger set is generated. The minimum trigger set aims to cover the main gaps, but adheres to the principles of minimum dimensions, minimum scope of effect, and minimum number of configurations, and strictly follows the target classification rules and hard constraint gating. The generation process of the minimum trigger set includes three stages: dimension necessity determination, configuration necessity determination, and scope of effect contraction. Dimension necessity determination is used to identify which dimensions are indispensable for gap coverage and which dimensions are only optional for improvement. Configuration necessity determination is used to select the minimum subset of configurations that meet the gap coverage requirements within a certain dimension, and prioritizes configurations that are strongly correlated with the evidence chain and whose dependency conditions can be met, avoiding the introduction of configurations that require supplementary measurement fields. Scope of effect contraction is used to limit the triggering results to the scope of objects with the clearest gap evidence. For example, only triggering shading configurations for highly exposed facades, triggering thermal insulation and airtight configurations for structurally weak enclosure objects, and triggering combined daylight enhancement and glare suppression configurations for areas with insufficient lighting and controllable glare, without generalizing the configurations to the entire strategy unit. The minimum trigger set records at least the following fields: a list of required dimensions, the trigger configuration range for each dimension, a description of the trigger object range, the corresponding gap entry number, the referenced evidence field number, and the alternative link number if an alternative path exists. When a gap is closed due to a hard constraint and an alternative path is used, the minimum trigger set must simultaneously record the reason for the closure of the main dimension, the referenced hard constraint entry number, and the adopted alternative path number to ensure that subsequent combination generation will not return to the closed dimension.

[0064] To ensure the consistency of constraints between the minimum trigger set and subsequent combinations, a trigger set consistency check is performed after the minimum trigger set is generated. The consistency check includes at least three pre-checks: mutual exclusion, dependency satisfiability, and alternative reachability. The mutual exclusion check checks whether trigger ranges in different dimensions within the minimum trigger set contain mutually exclusive pairs. If so, the trigger range needs to be narrowed or a clear alternative branch needs to be defined. The dependency satisfiability check checks whether all dependency conditions in the trigger range are available or fallback available in the evidence chain record table. If they are unavailable and require further testing, the configuration should be removed from the trigger range or downgraded to an alternative suggestion. The alternative reachability check checks whether there is an alternative path covering the same gap when a critical configuration within the trigger range is removed due to subsequent gating. If not, an alternative configuration range needs to be added during the trigger set stage, or the gap needs to be marked as requiring further testing before entering the combination. After verification, the gap diagnosis form, gap-dimensional branch record, minimum trigger set, and consistency verification record are bound and archived together with the task book version number, evidence chain version number, and configuration space version number to form a unique input constraint file for the generation of this candidate strategy package.

[0065] It should be noted that after completing the strategy unit gap diagnosis form, gap-dimensional branch record, and minimum trigger set, and completing the consistency check, candidate strategy packages are generated within the optional set of configuration space based on the minimum trigger set. A three-stage screening process is then performed on the candidate strategy packages to ensure that the output results, while not violating hard constraints and meeting the target grading rules and gap coverage requirements, further satisfy robustness consistency and comparability of management standards. The candidate strategy package generation and screening process uses the unique identifier of the strategy unit as the primary key, establishing a binding relationship with the task book version number, evidence chain version number, configuration space version number, and minimum trigger set version number. For each elimination, downgrade, replacement, or selection, a verifiable record entry is generated. Each record entry includes at least the strategy package number, the set of included configuration option numbers, the scope of the target object, the set of trigger gap entry numbers, the set of cited evidence field numbers, the set of triggered mutual exclusion / dependency / substitution rule numbers, the screening stage identifier, the judgment basis reference, and a timestamp. This ensures that the three-stage screening does not merely provide a conclusion but can trace back to specific rules and evidence.

[0066] The generation of candidate strategy packages is based on a minimum trigger set as a boundary condition, adhering to the following restrictions: the trigger dimension list must not be exceeded, the trigger configuration range of each dimension must not be exceeded, the trigger object range must not be expanded, dependencies and time constraints must not be removed, and closed dimensions must not be enabled. The candidate strategy package is constructed based on a set of configuration option entries. During generation, the configuration options for mandatory dimensions are first determined, and then optional dimensions are incrementally combined according to task priority and gap coverage requirements. Mutual exclusion and dependency rule tables are applied in real-time during the combination process to ensure that the generated candidate strategy packages themselves satisfy basic logical consistency. For trigger ranges with two-way branches, candidate generation generates strategy package sets according to the branch paths, recording the branch identifier in the strategy package number so that subsequent screening can compare the comparability of different branches. For configuration options in the trigger range that include fallback available evidence fields, candidate generation marks the strategy package record with an uncertainty label for scenario consistency checks during the robustness screening phase. Once the candidate strategy package is generated, it enters a three-stage screening process. The three-stage screening consists of three stages: feasibility screening, robustness screening, and management screening. The three stages are executed in sequence. Strategy packages that fail the previous stage cannot enter the next stage, and the judgment of any stage cannot break through the hard constraints, gating and target classification rules fixed in the task book.

[0067] The first stage is feasibility screening, which is used to gating and eliminating candidate strategy packages at the engineering and management boundaries to ensure that strategy packages entering subsequent stages are implementable and logically consistent within the strategy unit. The criteria for feasibility screening include at least hard constraint gating rules, configuration space hard constraint pre-convergence records, mutual exclusion rule tables, dependency rule tables, and minimum trigger set consistency verification records. The feasibility screening process should perform at least the following checks and generate corresponding record entries: First, hard constraint conflict check: Check whether any configuration option in the strategy package triggers a global, partition, facade, or roof hard constraint entry. If triggered, it is deemed infeasible, and the triggered hard constraint entry number and the referenced evidence field are recorded. Second, mutual exclusion conflict check: Check whether the configuration option set in the strategy package contains mutually exclusive pairs. If contained, it is deemed infeasible, and the mutual exclusion rule number, conflict object range, and corresponding configuration number are recorded. Third, dependency satisfaction check: Check whether all dependency chains in the strategy package are satisfied, including configuration dependencies and condition dependencies. Configuration dependencies require verification from the configuration option set whether the configuration is selected simultaneously. Condition dependencies require verification from the evidence chain record table that the dependency condition field is available or rollback is available and does not trigger hard constraints. If any dependency is not satisfied, it is deemed infeasible. Fourth, trigger set coverage check: Check whether the strategy package covers the gap entries and dimensions marked as mandatory in the minimum trigger set. If the gap coverage is insufficient, it is deemed infeasible, and the gap entry number is recorded. The strategy package that passes the feasibility screening should simultaneously meet four conditions: zero conflict of hard constraints, zero conflict of mutual exclusion, satisfyable dependency links, and complete coverage of the minimum trigger set. It should also be marked as feasible in the record.

[0068] The second stage is robustness screening, used to verify the constraint consistency and gap coverage stability of the strategy package under uncertain conditions, preventing the strategy package from becoming sensitive to a single assumption and failing under actual operation or boundary changes. The inputs to robustness screening are the set of strategy packages that have passed feasibility screening, the uncertainty labels and backoff fields marked in the evidence chain record table, and the constraint items and target priorities related to operation and management in the task book. Robustness screening first establishes a scenario set, which must include at least extreme high temperature or high radiation scenarios, scenarios of extended occupancy periods or increased usage intensity, scenarios of changes in surrounding shading, scenarios of conservative or increasingly restricted window opening behavior, and scenarios of changes in the implementation window due to construction delays. The establishment of the scenario set requires recording the triggering conditions description, the involved evidence fields, and their direction of change for each scenario, and binding the scenario set version to the task book version. Subsequently, consistency checks are performed on each strategy package under various scenarios. These consistency checks include at least the following: First, constraint consistency checks, verifying that the strategy package does not trigger hard constraint entries after scenario changes, especially for configurations that rely on window opening times or approval conditions, checking whether they still hold true under scenarios with stricter window opening restrictions or tighter approvals; Second, gap coverage stability checks, verifying that the strategy package's coverage path for major gaps is not closed or weakened to an unacceptable degree after scenario changes. For example, when the ventilation dimension configuration relies on nighttime heat dissipation, under a conservative window opening scenario, it should be checked whether there is a combination of shading and enclosure supported by alternative paths to maintain coverage of overheating gaps; Third, uncertainty sensitivity checks, for configuration options supported by backoff fields, checking whether the strategy package still meets the minimum trigger set coverage requirement when the backoff field changes within a reasonable fluctuation range. If it is highly sensitive to the backoff field, it is judged as unrobust and the sensitive field number is recorded. The robustness screening output should be categorized into at least three types: robust pass, conditional pass, and fail. A robust pass indicates that constraint consistency is met and gap coverage is stable across the scenario set. A conditional pass indicates that in some scenarios, a fixed alternative path must be used or the implementation scope must be limited to maintain consistency; the condition must be recorded as an item and bound to the alternative rule number. A fail indicates that any scenario triggers a hard constraint conflict or causes the main gap coverage to fail, and there is no reachable alternative path. For policy packages that pass the conditional pass, the condition items should be written into the policy package record as mandatory preconditions for subsequent phasing and implementation lists, rather than merely as prompts.

[0069] The third stage is management screening, which, under the premise of passing feasibility and robustness gates, transforms strategy packages into comparable management indicator records and sorts them according to the fixed criteria in the task book, forming a result set that can be used for budget, schedule, downtime impact, and approval risk management. The inputs to management screening are a set of robustly or conditionally approved strategy packages, indicator criterion field templates fixed in the task book, and cost and construction condition-related fields from the evidence chain and ledger snapshots. Management screening first generates indicator record entries for each strategy package. These entries include at least the following: investment cost range and pricing scope description; maintenance cost and frequency constraints; workability window matching degree; phased implementation divisibility; downtime impact area and period; approval risk level and basis entries; and a summary of coverage mapping for target gaps such as overheating, heat loss, lighting glare, and ventilation availability. The cost range in the indicator records comes from enterprise quotas, market information prices, historical settlements, or inquiry records, with the source specified in the record; the schedule window and downtime impact come from the operation calendar and downtime approval system entries; and approval risk comes from approval conditions, appearance requirements, or management regulations entries. Subsequently, the strategy packages are sorted and prioritized according to the target hierarchy and priority rules in the task book. The sorting rules first ensure that the first-level target and hard constraint gating are not exceeded. Secondly, the third-level target is compared only if the gaps in the second-level target are covered. When a strategy package has conditions that allow it to pass, the sorting process includes its associated conditions in the impact factors to avoid misjudging strategy packages with strong conditions and numerous pre-implementation requirements as priority. The sorting results output the Top-N strategy packages, and for each selected strategy package, its screening record chain is also output, including the passed feasibility checks, robustness scenario consistency conclusions, and management indicator items, ensuring that the output can be used for subsequent phased planning and verification of the established standards.

[0070] After the three-stage screening is completed, the candidate strategy package generation record, feasibility screening record, robustness screening record, management screening indicator record, ranking results and the list of selected strategy packages are archived together, and bound to the task book version number, evidence chain version number, configuration space version number and minimum trigger set version number to form a strategy package screening version snapshot.

[0071] S4: Generate strategy packages based on the minimum trigger set and sequentially complete the filtering and management caliber sorting output, outputting the building passive energy-saving strategy combination and phased implementation and verification write-back results.

[0072] Furthermore, after completing the three-stage screening of strategy packages and obtaining a list of Top strategy packages that have passed feasibility and robustness gating and completed management indicator recording, the Top strategy packages are transformed from a set of configuration options into an executable component-level implementation list and phased implementation plan. The phased implementation plan and resource configuration fields are then archived and solidified, enabling subsequent verification criteria solidification and knowledge rewriting to be carried out with phased milestones as boundaries. Phased implementation is based on the target hierarchical rules, hard constraint gating rules, schedule windows, and shutdown impact criteria solidified in the task book. It uses the boundaries of enclosure, openings, and roof objects recorded in the strategy unit ledger snapshot as the mapping basis, and the operational and approval evidence and construction condition evidence in the evidence chain record table as constraint inputs. The strategy package indicator record items and condition pass items output by the three-stage screening serve as the direct basis for phased decisions. The phased implementation process uses the unique identifier of the strategy unit and the strategy package number as the primary key. For each step of mapping, splitting, scheduling, and resource field assignment, a verifiable record entry is formed. The record entry includes at least the configuration option number, the corresponding component object number, the scope description, the hard constraint entries and dependency condition entries involved, the planned phase number, the planned construction window, the planned decommissioning scope, the pre-approval list, and the planned acceptance milestone identifier.

[0073] First, a mapping from strategy packages to object scopes is performed. Each configuration option in the Top strategy package is mapped to a specific object list in the strategy unit ledger snapshot, based on the applicable object scope field defined in the configuration space entry. The object list must include at least exterior wall block objects, roof block objects, window and door opening objects, critical thermal bridge node objects, and facade component objects related to shading and self-shading, with each object assigned a unique object number. The mapping process must not use general descriptions to replace object numbers; a one-to-many or many-to-one relationship must be established between configuration options and object numbers, and the source of the referenced ledger field, drawing page number, or model version number must be noted in the mapping record. For strategy units involving differences in zoning constraints, the mapping record must further specify the restriction boundaries of the configuration option within the object scope, such as only applying to zoning areas where construction windows are permitted, only applying to discontinuous operating areas, or only applying to facade blocks explicitly permitted by approval conditions. This ensures that subsequent scheduling and decommissioning plans can directly reference the object scope without reinterpretation.

[0074] After completing object mapping, a component-level implementation list is generated. The component-level implementation list is indexed by object number and lists at least the configuration option types, applicable condition references, precondition references, mutual exclusion and dependency rule references for each object to be implemented, and, if the configuration is a conditional pass policy package, the condition entries need to be transcribed into executable preconditions. Precondition references include at least four categories: approval preconditions, deactivation approval preconditions, safety management preconditions, and operation and maintenance coordination preconditions. Approval preconditions refer to planning approvals, landscape reviews, or property permits that need to be met, referencing approval evidence fields in the evidence chain; deactivation approval preconditions refer to deactivation scope, deactivation period, and approval process entries, referencing deactivation guidelines and operational system entries in the task book; safety management preconditions refer to scaffolding, suspended platform, and roof operation permits and prohibited window entries, referencing construction condition evidence fields; and operation and maintenance coordination preconditions refer to window opening time management, cleanliness management, or noise control entries, referencing operational evidence fields. For configuration combinations that have mandatory linkage in the dependency rule table, such as linkage between daylight enhancement and glare suppression, or linkage between airtightness improvement and restricted ventilation periods, the implementation list must record the linkage relationship with the same object or the same phase binding method to avoid situations where individual implementation leads to unmet conditions.

[0075] Subsequently, a phased implementation plan was developed. The phased development was based on the priority of the objectives and hard constraints of the task statement. The management indicators recorded in a three-stage screening process, including items related to cost, construction period window, impact of shutdown, approval risk, and robustness conditions, were used as sorting inputs. The following scheduling principles were followed for phased decomposition: First, priority was given to ensuring that the first-level objectives and hard constraints were not violated. For facade structures involving uncertain approvals or potentially triggering style restrictions, if pre-approval could not be completed within the current construction period window, they should not be included in the first phase but should be placed in a pilot phase or subsequent phase with clearly defined pre-approval completion milestones. Second, under the premise of meeting hard constraints, priority was given to arranging structures that contributed significantly to covering the gaps in the second-level objectives and had minimal impact on construction organization into the first phase. For example, within strategy units with significant overheating gaps and clear evidence of solar exposure, priority was given to arranging structures that did not violate approval red lines and... For configurations with clearly defined solar heat gain control targets, glare suppression or functional boundary maintenance measures should be implemented simultaneously. Third, for configurations with significant impact from shutdown, the scope and timing of shutdown must be clearly defined in the phased plan, prioritizing shutdown during off-peak hours or within fixed shutdown windows. If the shutdown window is not met, the phased plan should be divided into multiple sub-phases with a narrowed target scope. Fourth, for configurations with dependency chains, scheduling must be done in dependency order; subsequent configurations cannot be initiated until preceding configurations are completed, and dependency completion checkpoints should be set in the phased milestones. Fifth, for strategy packages marked as conditionally passed in robustness screening, their condition entries must be converted into mandatory restrictions in the phased plan, such as limiting the implementation scope, limiting the implementation time, or requiring the simultaneous activation of alternative path configurations. These must be solidified as entries in the phased plan and not merely as notes.

[0076] The phased plan output should include at least the following fields for each phase: Construction Window, Object Scope, Deactivation Scope, Pre-Approval Checklist, Resource Consumption, and Milestone. The Construction Window field, based on the operational calendar and restricted access windows, specifies the start and end dates or set of workable periods. The Object Scope field, represented by a set of object numbers, specifies the exterior wall blocks, roof zones, opening objects, and node objects covered in this phase. The Deactivation Scope field specifies the deactivation area or boundary, deactivation period, and deactivation approval item references. The Pre-Approval Checklist field lists the approval items, permits, and corresponding item numbers that need to be completed before this phase. The Resource Consumption field should at least include the budgeted occupancy range and pricing caliber references, constraints on major construction organization methods (such as scaffolding or suspended platform availability and safety permit requirements), and operation and maintenance coordination requirements. The Milestone field should at least include pre-construction completion nodes, key object completion nodes, dependency link acceptance nodes, and this phase's verification nodes, assigning a unique identifier to each milestone.

[0077] During the phased planning process, a phased scalability record is generated concurrently to explain why a certain configuration is split or postponed. The scalability record must include at least the type of split reason, the scope of the object before and after the split, the triggered hard constraints or operational restrictions, and supporting evidence demonstrating that the gap coverage requirements are still met after the split. The types of split reasons must include at least insufficient pre-approval, insufficient shutdown windows, construction access restrictions, roof maintenance access restrictions, cleanroom or noise management restrictions, and unmet dependency links. Each split reason must reference the corresponding item number in the task book or evidence chain to avoid a lack of supporting evidence for the split decision.

[0078] After completing object mapping, generating the component-level implementation list, and compiling the phased implementation plan, the component-level implementation list, phased plan, phased splitting record, and phased milestone list are bound and archived together with the task book version number, ledger snapshot version number, evidence chain version number, configuration space version number, and screening version snapshot number to form an implementation plan version snapshot. Subsequent verification criteria solidification and knowledge write-back updates reference this implementation plan version snapshot with phased milestones as boundaries. If the construction window, approval conditions, or operational restrictions change during implementation, it is necessary to trace back to the object scope field, precondition field, and splitting reason record in the implementation plan version snapshot, re-execute the phased adjustment, and generate a new version number to maintain the continuity and traceability between the strategy generation results and the implementation plan.

[0079] It should be noted that after generating the strategy package under the minimum trigger set constraint and sequentially passing feasibility screening, robustness scenario consistency screening, and management caliber sorting output, the selected strategy package is taken as the final strategy combination result of the current strategy unit. Simultaneously, using the phased implementation arrangement and verification write-back requirements of this result as boundaries, a verification caliber is established and a result write-back update is performed. This ensures that the passive energy-saving strategy combination not only produces interpretable output, but also that its implementation and verification process can conversely constrain the version evolution of evidence, configurations, and rules. The task book version number is used as the unified benchmark for the verification caliber. The strategy package number, configuration option number set, object scope, and condition entries recorded in the strategy package screening version snapshot are used as the verification object input. The milestone markers, construction windows, decommissioning scope, and pre-approval list in the phased implementation plan version snapshot are used as the verification boundaries. Evidence fields marked as available or revertable in the evidence chain record table are used as the verification alignment basis, ensuring that the verification caliber is consistent with the preceding judgment caliber and is traceable.

[0080] First, the verification criteria package is solidified according to phased milestones. For each milestone, the verification criteria package establishes the definition of a baseline period and a control period, and links the selection of the baseline and control periods to the milestone completion node, explicitly excluding periods of construction disturbance and unstable periods due to changes in operating procedures. If changes in operating procedures are unavoidable, the source of the change must be registered in the verification criteria package as a candidate for deviation reasons. The verification criteria package must at least solidify the list of data collection items, data quality inspection rules, evaluation criteria rules, and threshold source explanations. The list of data collection items must at least include outdoor environmental data, indoor thermal environment data, energy consumption or billing data, occupancy and operating condition data, window opening management or ventilation restriction execution data, and component completion status verification data. For each type of data, the source type and number must be registered, such as the meteorological dataset name or monitoring equipment number, energy consumption metering system export number, operation and maintenance procedure or approval document number, and on-site sampling verification record number. Data quality inspection rules should include at least the following: missing data rate inspection, outlier inspection, timestamp alignment inspection, and object scope consistency inspection. The object scope consistency inspection requires that the scope of the verification data collection must be consistent with the object number set in the implementation plan. If inconsistency is found, no definitive verification conclusion should be given; instead, it should be recorded as a scope mismatch and trigger supplementary data collection or adjustment of the control period. Evaluation caliber rules must reference the fixed indicator caliber field numbers in the task book, providing consistent descriptions for overheating-related, heat loss-related, daylighting and glare-related, ventilation availability-related, energy consumption and carbon-related, and project downtime impact-related calibers. Threshold source descriptions are used to fix the source of judgment for each caliber. The source must be at least one of the following: current standard provisions, internal organizational specifications, statistical calibers from a benchmarking library of similar projects, or written requirements from the owner. The reference number or statistical scope must be specified in the verification caliber package to avoid situations where thresholds lack sources or caliber inconsistencies arise later.

[0081] Secondly, milestone verification is performed according to the verification criteria package, forming a verification record chain. The verification record chain uses the milestone identifier as the primary key and records the set of object numbers covered by that milestone, the corresponding strategy package number, the corresponding configuration option number set, baseline and control period references, completion status of the data collection items, data quality inspection conclusions, evaluation criteria calculation or judgment results, and execution verification results related to robustness screening condition items. For condition items generated during the robustness screening phase, such as implementation scope restrictions, implementation time restrictions, restrictions requiring the use of alternative paths, or restrictions requiring the satisfaction of dependency links, the verification record chain must verify each item against the implementation plan and associate the verification results with the reasons for deviations. The categories of deviation reasons should at least include changes in operating conditions, deviations in window management execution, changes in object scope due to approval conditions, construction implementation deviations, changes in external obstruction, differences in climate boundaries, and anomalies in the data collection link. Each deviation reason must be able to reference a corresponding evidence or record item number. If the verification conclusion is affected by insufficient data quality, the verification record chain should clearly indicate the reasons for insufficient data and generate a supplementary data collection list or adjustment suggestions for extending the control period, but these should not be used as the basis for rule rewrite.

[0082] Finally, a verification write-back update is performed, generating a version upgrade. Write-back updates include at least three types: evidence chain write-back, configuration entry write-back, and rule table write-back. Evidence chain write-back updates the reliability flags of evidence fields and the applicable boundaries of fallback fields. For fields previously available for fallback, if the verification conclusion matches the fallback hypothesis, their reliability is increased and their applicable scope is clarified; if they do not match, their applicable boundaries are tightened or they are upgraded to priority fields requiring supplementary testing, and the triggering conditions and object scope are recorded. Configuration entry write-back updates the applicable conditions, preconditions, and risk labels of configuration option entries. When verification shows that a configuration has stability issues under a specific object scope or a specific combination of constraints, the applicable conditions of its entry need to be converged and explicit restrictions added; when verification shows that a configuration is stably valid within a certain type of strategy unit, the risk label of its entry can be lowered and its dependency condition boundaries clarified. The rule table write-back is used to update mutual exclusion, dependency, and substitution rules. When verification shows that a certain configuration combination has incompatibility risks under engineering implementation or management constraints, it supplements the mutual exclusion relationship or limits the scope of mutually exclusive objects. When verification shows that a certain configuration must be linked to another configuration or must meet a certain precondition to be stable, the dependency relationship is changed from optional linkage to mandatory linkage and the basis is recorded. When verification shows that the original substitution path is unreachable under the actual boundary or the coverage of the main gap is unstable, the trigger conditions and substitution links of the substitution rule table are updated, and the difference entries before and after the update are recorded. After the write-back is completed, new version numbers are generated for the evidence chain record table, the configuration space entry set, and the mutual exclusion dependency substitution rule table, respectively. The new version numbers are bound and archived with the corresponding milestone verification record chains, so that when generating strategy packages in the next round according to the minimum trigger set, the latest version of the evidence reliability mark, configuration applicable boundary, and substitution link can be used first. This continuously outputs interpretable passive energy-saving strategy combinations for buildings and can support the generation of passive energy-saving strategy schemes for management decision-making with phased implementation and verification write-back results.

[0083] Example 2, one embodiment of the present invention, provides a method for generating passive energy-saving strategies for buildings. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0084] First, six existing public buildings within the same urban climate zone were selected as test subjects: office building A, teaching building B, hospital C, library D, dormitory E, and commercial building F. All buildings had as-built drawings or BIM models, energy consumption bills for the past 12 months, or data exported from metering systems, and on-site inspections and sampling verification were possible. The experiment proceeded in parallel with two processes: one a common traditional process, and the other the process of this invention. Both used the same original data as input to ensure comparability. The process of this invention first generated a strategy task book. The task book itemized the sources and scope of application for objectives such as overheat control, heat loss control, daylighting and glare, availability of natural ventilation, carbon and cost, construction period window, and impact of shutdown. It also solidified four layers of hard constraint gate control items: global, zoned, facade, and roof. For example, hospital C had institutional restrictions on window opening in certain areas; commercial building F had restrictions on facade approval and prohibitions on cantilevered structures; and library D had explicit constraints on both daylighting and glare in glare-sensitive areas. Subsequently, each building was divided into strategy units based on orientation shading, building envelope, opening restrictions, operational constraints, and consistency of construction windows, and a ledger snapshot was created. The ledger snapshot included sub-tables for building envelope, opening openability, existing shading and self-shading, operational constraints, and construction windows, and recorded the source and verification status of each field. An evidence chain record table was generated at the strategy unit level. This evidence chain at least covered evidence fields such as climate boundaries, solar exposure and shading, heat transfer and infiltration, ventilation availability, daylighting and glare, operational approvals, and construction conditions. Sufficiency gating was applied to key evidence, and missing fields were rolled back using a benchmark library or supplementary measurement list. For fields involving hard constraints, written documentation was prioritized. Based on the evidence chain, a configuration space was established, and configurations triggering hard constraints were eliminated during the generation phase. For example, for commercial building F, shading configurations that would cause the facade to overhang were eliminated, and for hospital building C, the natural ventilation dimension dependent on window opening times was closed. Simultaneously, mutual exclusion rule tables, dependency rule tables, and alternative rule tables were constructed to constrain propagation and convergence of reachable alternative paths. Next, gap diagnosis is formed based on the evidence chain. The gap is mapped to dimensional branch paths and converges to the minimum trigger set to limit the candidate combination boundary and avoid invalid superposition. Finally, a strategy package is generated according to the minimum trigger set, and feasibility screening, robustness scenario consistency screening, and management caliber sorting output are completed in sequence to generate the Top strategy combination, which is mapped to a component-level list and phased milestones. The verification caliber is solidified according to the milestones, and the source of verification data and quality inspection conclusions are recorded. The verification deviation is written back to update the reliability of the evidence field and the version of the mutual exclusion dependency substitution rule for the second round of deviation convergence verification. The traditional process control group uses an experience-driven strategy list and one-time combination sorting. Constraints are mostly discovered in the later review or implementation stage, and there are no evidence sufficiency gates, mutual exclusion dependency substitution propagation, and write-back update links.

[0085] Table 1 Experimental Data

[0086] Table 1 illustrates the differences in the entire process for the same group of building objects under the two types of workflows. Firstly, the total time spent generating strategies shows that this invention is significantly shorter than the traditional method for all six objects. For example, office building A's time decreased from 16.5 hours to 6.2 hours, and hospital C's time decreased from 19.5 hours to 7.0 hours. This difference is not simply due to compressed calculations, but rather to the process convergence resulting from the solidification of the task specification and the pre-implementation of hard constraint gating, as well as the reduction of rework through strategy units and evidence chain gating. Traditional workflows often only discover boundary conflicts such as approvals, window restrictions, and window shutdowns later in the process, leading to repeated modifications and secondary comparisons, thus lengthening the cycle. Secondly, the number of candidate strategy packages is reduced from the traditional 88-150 to 24-40, reflecting the constraint effect of gap diagnosis and minimum trigger set mechanism on the combinatorial boundary. This mechanism does not reduce exploration at the expense of quality; on the contrary, it eliminates infeasible combinations in advance and provides achievable alternative paths through mutually exclusive dependency substitution rules, allowing exploration to focus on feasible subspaces that cover gaps. Therefore, while having fewer candidates, the feasibility pass rate is significantly improved: traditional methods only achieve 35%-50%, while the three-stage screening of this invention reaches 72%-85%. This result shows that this invention not only reduces invalid solutions but also significantly increases the proportion of solutions that can be implemented, especially in the complex constraints of hospital C and commercial F. The feasibility pass rates of traditional processes are 35% and 38%, respectively, while this invention still maintains 72% and 74%, demonstrating the adaptability of hard constraint hierarchical gating and configuration pre-convergence to complex scenarios.

[0087] Secondly, the hard constraint conflict index exhibits typical forward control characteristics: traditional methods add 3 to 8 new hard constraint conflicts during the implementation phase, meaning that a large number of strategy combinations are not identified in the generation phase and the risks are only exposed in the approval, construction, or operation coordination phases; this invention intercepts 7 to 12 hard constraint conflicts in the generation phase, indicating that conflicts are identified and blocked from entering subsequent phases at an early stage, reducing the uncertain costs of later shutdowns and rework from a management perspective. More importantly, the actual energy saving rate after verification reflects the contribution of feasibility and robustness screening to the real effect: the energy saving rate of this invention is 15.4% to 19.6% on various objects, significantly higher than the traditional 8.1% to 11.0%. In particular, Library D requires a linkage between daylight and glare due to its sensitivity to daylight. Traditional processes often adopt conservative shading under empirical trade-offs, resulting in limited energy saving improvement. However, this invention enforces linkage through dependency rules and controls glare risk under robust scenarios, thus maintaining a high energy saving level of 19.6% in the verification. Budget deviations also demonstrate the advantages of standardized terminology and phased milestone mapping: traditional budget deviations range from 18% to 30%, while this invention reduces them to 6% to 11%. This indicates that the strategy package incorporates management fields such as construction window, decommissioning impact, and pre-approval requirements during the generation phase, reducing settlement deviations caused by later additions to measure costs and decommissioning adjustments. Finally, the second-round prediction error after write-back is controlled at 3.5% to 4.9%, demonstrating that after updating the reliability of evidence, the applicable boundaries of the configuration, and the alternative links through verification write-back, subsequent generation can achieve deviation convergence. Traditional processes lack a write-back closed loop, making it difficult to transform verification deviations into reusable boundaries for rules and evidence, thus failing to consistently improve prediction consistency in the next project or the next iteration. In summary, the table data, from the dimensions of time, feasibility, forward movement of constraint conflicts, actual energy saving improvement, budget controllability, and closed-loop convergence, collectively support the creative improvement path of this invention in the generation of passive energy-saving strategies.

[0088] Example 3, an embodiment of the present invention, provides a building passive energy-saving strategy generation system, including a task specification module, a unit and evidence module, a configuration and trigger module, and a strategy generation closed-loop module.

[0089] The task book solidification module is used to clarify the source of the target caliber, the target classification rules, and the four-layer hard constraint gating and conflict record caliber; the unit and evidence module is used to divide the strategy units according to modifiable consistency and solidify the ledger snapshot, generate the strategy unit evidence chain and complete the evidence sufficiency gating and backtracking; the configuration and trigger module is used to establish and converge the feasible configuration space, form mutually exclusive dependency substitution rules, map the gap diagnosis to the dimensional branch and converge the minimum trigger set to limit the combination boundary; the strategy generation closed-loop module is used to generate strategy packages according to the minimum trigger set and complete the screening and management caliber sorting, output the phased implementation plan and verify and update the evidence and rule versions according to the milestone.

[0090] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0091] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0092] More specific examples (a non-exhaustive list) of computer-readable media include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which programs can be printed, because programs can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0093] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc. It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

[0094] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for generating passive energy-saving strategies for buildings, characterized in that, include: Compile and archive the strategy generation task book, register the source of the target scope and the target classification rules item by item, and solidify the global partition facade and roof four-story hard constraint gating and conflict record fields; Based on the task description, the building is divided into strategic units according to the consistency of orientation obstruction, enclosure structure, opening restrictions, operational constraints and construction windows, and the ledger snapshot is solidified. An evidence chain record table is generated at the strategic unit granularity, and sufficiency gating is performed on key evidence and rollback is carried out according to the benchmarking and supplementary measurement list. Establish a configuration space in the evidence chain record table and first eliminate configurations that trigger hard constraints. Construct a mutual exclusion rule table, a dependency rule table, and an alternative rule table. Map the gap diagnosis results to dimensional branch paths and converge to the minimum trigger set to limit the candidate combination boundary. Generate strategy packages based on the minimum trigger set, and sequentially complete the filtering and management classification sorting output. Output the building passive energy-saving strategy combination and the phased implementation and verification write-back results.

2. The method for generating passive energy-saving strategies for buildings as described in claim 1, characterized in that: The task book for compiling and archiving the strategy includes target fields, priority fields, hard constraint fields, evaluation indicator caliber fields, and conflict record fields. The target fields are classified and solidified into three levels: safety and operational boundaries, indoor comfort and health, and carbon and economy. The source and scope of the data are registered for each item, including overheat control, heat loss control, lighting and glare, availability of natural ventilation, carbon emission reduction, budget, construction period window and impact of shutdown. Hard constraint fields are fixed into four levels of gated entries: global, zone, facade, and roof, and each gated entry is registered based on its source type and number. When a candidate strategy or strategy package triggers any hard constraint conflict, a conflict record entry is generated, which includes the scope of the conflicting object, the triggering entry number, the evidence reference field, and the timestamp.

3. The method for generating passive energy-saving strategies for buildings as described in claim 2, characterized in that: The partitioning strategy unit includes merging or splitting the initial candidate regions based on modifiable consistency. The modifiable consistency includes consistency in enclosure structure, orientation and obstruction, opening openability and ventilation restrictions, operational constraints, and construction windows and shutdown conditions. When any consistency dimension is not satisfied, a split is triggered and the triggering dimension, the difference field, the source of the difference evidence, and the corresponding task book constraint item number are recorded. A unique identifier is assigned to the final strategy unit and a ledger snapshot is fixed. The ledger snapshot includes sub-tables for enclosure structure, openings, shading and self-shading, operation constraints and construction windows, and each field is labeled with source type, formation time and verification status.

4. The method for generating passive energy-saving strategies for buildings as described in claim 3, characterized in that: The evidence chain record table generated at the strategy unit granularity includes generating evidence of climate boundaries, solar exposure and shading, heat transfer and infiltration, ventilation availability, lighting and glare, operation and approval, and construction conditions. For each evidence field, the source type, formation method, applicable scope, and verification status are recorded. Perform evidence sufficiency gating on the key evidence field set and classify the evidence status of the strategy unit into available, rollback available, and unavailable that needs to be retested; When key fields are missing or untraceable, they should be rolled back and supplemented according to the priority of the same building history, the same benchmark library in the same climate zone, the organization's internal specifications or industry guidelines, and the rollback source and applicable boundaries should be registered. Fields involving hard constraints are not allowed to be directly replaced by benchmark default values.

5. The method for generating passive energy-saving strategies for buildings as described in claim 4, characterized in that: The process of establishing a configuration space in the evidence chain record table and first eliminating configurations that trigger hard constraints includes pre-converging the configuration space based on the evidence chain record table and hard constraint gating. The configuration space is set with configuration options according to the dimensions of enclosure and airtightness, solar heat gain control, heat dissipation and natural ventilation, lighting and glare control, and roof microclimate. Applicable conditions, prerequisites, constraint association items and references are registered in each configuration option. Establish a mutual exclusion rule table, a dependency rule table, and an alternative rule table. Mutual exclusion rules are used to limit the combination of configurations that cannot be simultaneously true within the same object scope. Dependency rules are used to limit configuration linkage or precondition satisfaction relationships. Alternative rules are used to provide alternative paths in the same dimension or across dimensions when configurations trigger hard constraints or dependencies are not satisfied. The configuration options that trigger hard constraints are marked as unavailable, and an entry is created to record the reason why they are unavailable.

6. The method for generating passive energy-saving strategies for buildings as described in claim 5, characterized in that: The construction of mutual exclusion rule tables, dependency rule tables, and substitution rule tables, mapping gap diagnosis results to dimensional branch paths, and converging to the minimum trigger set to limit the candidate combination boundary includes generating a gap diagnosis form based on the task book target classification and the evidence chain record table, and mapping the gap diagnosis results to gap-dimensional branch paths; The gap diagnosis includes at least the overheating risk gap, heat loss risk gap, insufficient lighting gap, glare risk gap, and natural ventilation availability gap, and references the corresponding evidence field and verification status for each gap; When there is an institutional ban on opening windows or the ventilation dimension is closed by gate control, the alternative path of solar heat gain control and building envelope and airtightness is switched instead of the heat dissipation and ventilation dimension. Based on the branch mapping, a minimum trigger set is generated to cover the main gaps. The minimum trigger set limits the required dimensions, trigger configuration range, and object range, and a check record is formed for mutual exclusion preflight, dependency satisfiability preflight, and alternative reachability preflight.

7. The method for generating passive energy-saving strategies for buildings as described in claim 6, characterized in that: The process of generating a strategy package based on the minimum trigger set and sequentially completing the screening and management caliber sorting output includes generating a candidate strategy package based on the minimum trigger set and sequentially performing feasibility screening, robustness scenario consistency screening, and management caliber screening. Feasibility screening includes hard constraint conflict checking, mutual exclusion conflict checking, dependency satisfaction checking, and mandatory gap coverage checking. Robustness screening is based on a set of scenarios including extreme high temperature and high radiation, extended occupancy, changes in shading, conservative window opening, and changes in construction window. The strategy package is checked for constraint consistency and gap coverage stability, and the implementation scope or alternative path condition items are solidified through the strategy package. The management scope is filtered to generate and sort records of cost, schedule window, impact of shutdown and approval risk indicators according to the task book scope, and output them in order. The selected strategy packages are mapped to component-level lists and phased milestones. The verification criteria are solidified according to the milestones, and the verification results are written back to update the evidence chain, configuration entries, and rule table versions.

8. A system employing the building passive energy-saving strategy generation method as described in any one of claims 1 to 7, characterized in that: It includes a task specification module, a unit and evidence module, a configuration and trigger module, and a strategy generation closed-loop module; The task book solidification module is used to clarify the source of the target caliber, the target classification rules, and the four-layer hard constraint gating and conflict record caliber. The unit and evidence module are used to divide the strategy units according to modifiable consistency and solidify the ledger snapshot, generate the strategy unit evidence chain and complete the evidence sufficiency gating and rollback completion; The configuration and triggering module is used to establish and converge the feasible configuration space, form mutually exclusive dependency substitution rules, map the gap diagnosis to dimensional branches and converge the minimum trigger set to limit the combination boundary. The strategy generation closed-loop module is used to generate strategy packages based on the minimum trigger set, complete the screening and management caliber sorting, output phased implementation plans, and verify and update evidence and rule versions according to milestones.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method for generating a passive energy-saving strategy for buildings as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the method for generating a passive energy-saving strategy for buildings as described in any one of claims 1 to 7.

Images