38 results about "Consistency index" patented technology

The invention discloses a standard courseware generation method for an artificial intelligence learning mode, and relates to the technical field of educational informationization, and the method comprises the steps: firstly collecting a course standard, a world language ready framework and translation memory, and generating a reference semantic vector set with a unique version fingerprint through vectorization; mapping the multilingual courseware draft on the multi-lingual courseware draft, outputting a weighted difference vector and recording a timestamp; the difference vector is decomposed into three factors of emotional tendency, cultural symbols and term consistency, and a deviation registration form is generated according to a dynamic threshold value; generating type rewriting, template replacement or manual prompt is selected according to a rewriting energy function, the high-risk chunks are locally revised, and consistency marks containing version fingerprints are written in; and finally, carrying out global residual error recheck, generating a consistency report after confirming the quality through a global consistency index, packaging the consistency report and the revised courseware into a release package for multi-end synchronization, writing fingerprints on a chain, and constructing a whole-course reliable new cross-language courseware quality closed loop.

The invention relates to the technical field of aviation life prediction, and discloses an aircraft component residual life prediction and maintenance decision-making method based on digital twinning, which comprises the following steps: step 1, establishing a mechanism model, determining a key parameter value range and an initial health state, and setting a deterministic conversion rule; 2, generating a virtual-real consistency index, executing consistency gating, and performing shrinkage updating on a key parameter value range; 3, splitting the flight task into a task spectrum scene set, and determining upper and lower limits of a load; 4, performing mechanism forward calculation based on the task spectrum scene set, and determining a residual life interval; 5, generating a candidate maintenance scheme; 6, eliminating the schemes exceeding the failure critical standard through virtual drilling; and 7, selecting a final maintenance scheme, and writing back state record data to update a health state and a key parameter value range. According to the method, prediction of the remaining life of the aircraft component and closed-loop optimization of the maintenance decision are realized.

The invention relates to the technical field of live-action three-dimensional modeling, in particular to a live-action three-dimensional model quality evaluation method, which comprises the following steps of: firstly, constructing a multi-dimensional evaluation index set covering dimensions such as geometric accuracy and texture quality; an initial weight is obtained through coupling of an analytic hierarchy process and an entropy weight method, and a dynamic weight is formed by combining scene demand coefficient adjustment; then, a subjective-objective linkage relation is established through grey correlation analysis, and index scores and weights are corrected through feedback coefficients; then calculating the total quality score of a single time node; the evaluation effectiveness in the dynamic scene is verified through a time sequence consistency index; and finally outputting the comprehensive quality grade. According to the method, a multi-dimensional index set is constructed to cover full dimensions, dynamic weights are calculated in combination with scene demand coefficients, and the problems of single dimension and poor scene adaptation are solved; building subjective-objective linkage by means of grey relational degree, correcting scores and weights, and solving disjunction; and calculating a time sequence consistency index to verify dynamic coherence, solving static limitation, and adapting to full-life-cycle management and control of the model.

The application discloses a reliability evaluation and optimization method and system for large model function code, a medium and a program product, and belongs to the technical field of artificial intelligence. A demand prompt word, a test case and expected output are acquired to form a standardized test data set; an execution consistency score, a semantic consistency index and a static structure consistency index of the code to be evaluated are calculated and fused into an intention consistency comprehensive score; a safety score is obtained by performing static safety analysis on the code to be evaluated; the intention consistency comprehensive score and the safety score are fused into a final comprehensive score, and the reliability is judged; if the reliability is not good, a defect label is generated according to the scores of the indexes and weighted fusion is performed to obtain optimization guidance information, and the code is corrected or regenerated according to the information. The application comprehensively evaluates the code quality from four dimensions of execution, semantics, structure and safety, highlights the short board effect by adopting a fusion strategy, realizes closed-loop automatic optimization, and significantly improves the reliability and safety of the function code.

The invention discloses an aero-engine measurement parameter selection method based on a clustering method, and belongs to the technical field of engine state monitoring and fault diagnosis, and the method comprises the steps: building an aero-engine model, and generating a gas path fault sample; obtaining a sample and performing normalization processing to enable the sample to only reflect the percentage of the measurement parameter change in the fault mode; hierarchical clustering is executed under different clustering numbers, and a Davies-Bouldin index is calculated; the clustering number corresponding to the minimum Davies-Bouldin and the clustering result of the minimum Davies-Bouldin are selected as the optimal division; and aligning the optimal division result with a real label, and calculating a consistency index. According to the method, on the premise that key fault information is reserved, the number of required measurement parameters is remarkably reduced, the optimized parameter combination still keeps a high fault recognition rate under the condition of limited sensors, and the efficiency and practicability of state monitoring and fault diagnosis are improved.

The application relates to the technical field of aviation life prediction, and discloses an aircraft component residual life prediction and maintenance decision method based on digital twinning, which comprises the following steps: step 1, a mechanism model is established, the value range of a key parameter and an initial health state are determined, and a deterministic conversion rule is set; step 2, a virtual-real consistency index is generated, and consistency gating is performed to update the value range of the key parameter; step 3, a flight task is split into a task spectrum scene set, and upper and lower limits of the load are determined; step 4, a mechanism forward calculation is performed based on the task spectrum scene set to determine a residual life interval; step 5, a candidate maintenance scheme is generated; step 6, a virtual drill is used to eliminate schemes that exceed a failure critical standard; and step 7, a final maintenance scheme is selected, and state record data is written back to update the health state and the value range of the key parameter. The application realizes closed-loop optimization of the prediction of the residual life of an aircraft component and maintenance decision.

This invention relates to the field of logistics monitoring technology, and more particularly to a cold chain logistics monitoring method and system. The invention calculates data integrity index values ​​at each moment within a historical time period, i.e., the cumulative value of the product of the effective data percentage of each sensor and its importance weight; obtains threshold exceedance index values ​​at each moment based on pre-processed sensor data and corresponding set numerical ranges; obtains multi-source consistency index values ​​at each moment based on the standard deviation of redundant sensor data, semantic labels, and data changes between adjacent moments; weights and sums the three indices to obtain the confidence weights at each moment; constructs an enhanced input vector using the confidence weights, semantic labels, and data from each sensor; and predicts the pre-processed data and semantic labels of each sensor within a preset time step in the future, based on the enhanced input vectors at each moment within the historical time period, enabling accurate monitoring of cold chain logistics and improving monitoring reliability.

The embodiment of the invention provides a sample generation method for group product fault detection and diagnosis combined verification, and the method comprises the steps: screening initial samples stored in a database according to the project demands of a target project, obtaining a first intermediate sample related to the target item; when it is determined that the sample size of the first intermediate sample is insufficient, a second intermediate sample is generated through interpolation; determining fault modes corresponding to each first intermediate sample and each second intermediate sample, and extracting the fault modes meeting a consistency index in the fault modes to obtain a to-be-verified fault mode set; determining a to-be-verified sample corresponding to each fault mode in the to-be-verified fault mode set, and performing sufficiency test on the to-be-verified samples; and according to a sufficiency test result, determining a target sample related to the target item in the to-be-verified samples.

The application discloses a kind of digital archives intelligent information mining method, it is related to file information processing technical field, the present application is evaluated data quality by a series of quantitative indicators including consistency index, integrity index, importance index and comprehensive importance index, which effectively reduces subjective judgment when manual screening, make data screening process more objective, reliable and scientific, simultaneously, by combining quantitative indicators with comprehensive weight, to automatically generate weighted threshold based on data-driven, the application of this weighted threshold not only can quickly process large-scale data set, to reduce the demand for a large number of manual annotation of manpower, and can effectively filter out high-quality data set, and be applied to subsequent deep learning model training, to significantly improve the output accuracy and stability of model, ensure that more accurate prediction results can be obtained, and based on this more accurate prediction results, more scientific and reasonable business decisions can be made.

The present application relates to the field of coating parameter self-optimization, and discloses a coating process parameter self-optimization system and method fusing model self-learning, wherein a coating process parameter self-optimization method fusing model self-learning comprises the following steps: collecting process parameters, production conditions, equipment power states and detection result data of multiple batches of coating production in real time; aligning and correlating the data according to the process sequence to form a unique mapping relationship; reversely determining corresponding process parameter nodes and working condition node sets to generate a consistency index; jointly comparing the consistency index and the corresponding data chain distribution to output corresponding working condition drift identifiers; introducing the consistency index and the working condition drift identifiers as constraint conditions into a parameter update model to output a constrained parameter update result; and writing the corresponding new round of detection results into the unique mapping relationship to generate an incremental updated mapping relationship and a reachable state space. The present application has the advantage of improving the stability of the optimization result.

The invention belongs to the technical field of machine learning, and particularly relates to an emulsifier production process temperature control method and system based on machine learning, and the method comprises the steps: obtaining the operation data of an emulsification kettle, and removing the no-load loss to construct a shearing power density sequence; obtaining an embedded vector representing viscosity and heat retardation through the rheological state mapping network; the embedded vector is injected into a gate control circulation unit updating gate, and a temperature prediction sequence is output; calculating a consistency index based on the thermodynamic entropy yield, and combining sensible heat temperature rise power residual error to construct a correction term to rectify the prediction sequence; and inputting the consistency temperature into a model prediction controller to generate a valve instruction. According to the method, deep learning is constrained through a physical mechanism, and the problems of temperature control lag and overshoot of the non-Newtonian fluid caused by shear heat are solved.

The application relates to the technical field of migration verification of databases, in particular to a Xinchuang database migration verification method and system, which comprises the following steps: determining a migration consistency index based on constraint matching reliability and syntax difference at each collection time; counting the cost in the migration process of the execution plan data set of a source database to the execution calculation data set of a target database at each collection time, and analyzing the comprehensive difference between the migration consistency index and the path difference at all collection times before each collection time to determine the migration semantic fidelity of the Xinchuang database migration process at each collection time, so as to optimize the number of decision trees in a random forest algorithm and verify the Xinchuang database migration state. The application solves the problem that a traditional migration tool cannot accurately verify the business logic breakage and data semantic distortion caused by the difference between database kernels, and improves the success rate and continuity of Xinchuang database migration.

The invention provides a parameterized characterization method for quantifying building energy consumption curve prediction precision and difficulty. The method comprises the following steps: dividing a building group according to types; building a building type consistency index, wherein the building type consistency index is used for measuring the similarity of the same type of buildings in the aspect of power consumption behaviors; obtaining building type consistency indexes of building groups of the same type; constructing an electrical load prediction model of the building group of the same type; inputting the historical power utilization data set of the building groups of the same type into the power utilization load prediction model of the building groups of the same type to obtain the prediction precision of the power utilization load prediction model of the building groups of the same type; and establishing a first mapping relationship between the building type consistency index of the building groups of the same type and the prediction precision of the electrical load prediction model of the building groups of the same type. The method can assist in completing data screening of tasks in energy consumption prediction of large-scale building groups.

The invention provides a steel rail three-dimensional profile consistency polishing effect evaluation method and device, and relates to the technical field of railway steel rail maintaining.The method comprises the steps that first continuous three-dimensional profile data of a target steel rail section before polishing and second continuous three-dimensional profile data of the target steel rail section after polishing are obtained; for each of the first continuous three-dimensional profile data and the second continuous three-dimensional profile data, extracting position coordinate data along a plurality of preset longitudinal continuous lines; the position coordinate data comprises a vertical coordinate value, a transverse coordinate value and a longitudinal coordinate value; according to the position coordinate data, longitudinal deviation consistency indexes are calculated, and the longitudinal deviation consistency indexes comprise a high-low standard deviation index, an area deviation index average value, a standard deviation and a steel rail longitudinal consistency evaluation index; and according to the change of the longitudinal deviation consistency index before and after polishing, the steel rail polishing effect is comprehensively evaluated. According to the invention, the effect of steel rail polishing operation can be accurately and comprehensively evaluated.

The application provides a kind of liquid food straw sucking difficulty prediction method based on rheological test, the characteristic flow curve of liquid food sample is measured by steady flow test, respectively Carreau-Yasuda model or Casson model (if it is yield type fluid) fitting is carried out, and the predicted flow Q of straw is calculated c , then Herschel-Bulkley model fitting is carried out, and the consistency index K h , power law index n, Herschel-Bulkley yield stress τ h (if it is yield type fluid) is obtained.According to the equation of pipe flow flow, τ R Is calculated and substituted into Herschel-Bulkley model, and the apparent viscosity corresponding to shear rate is obtained, then the prediction model is constructed to predict the liquid food straw sucking difficulty, replace the traditional sensory evaluation method, and provide new ideas for food enterprise research and development decision.

The invention discloses a multi-microgrid cooperative active support control method, and relates to the technical field of energy cooperative control, and the method monitors the operation state information in the continuous operation period of a power distribution system, and switches to a cooperative active support mode when the operation state continuously deviates from a preset allowed operation interval. A power withdrawal analysis time window is determined by constructing an operation quantity time sequence and combining stable interval judgment and change direction consistency analysis. Constructing a power retracement variable quantity delta P in a retracement stage, and forming a retracement consistency index Cret through difference integral analysis to perform retracement state evaluation; and constructing a retracement stability index Rsta in a consistent structure state to carry out rhythm stability evaluation. And when the withdrawal rhythm of the micro-grid is stable, constructing a collaborative support constraint quantity Uad, mapping participation depth limitation information into collaborative active support control parameters, and enabling a multi-micro-grid collaborative active support process to form a closed-loop control structure based on withdrawal behavior analysis.

The invention discloses a multi-source heterogeneous environmental protection monitoring data analysis method and system, and belongs to the technical field of environmental monitoring. The method comprises the following steps: acquiring multi-source heterogeneous original observation data and carrying out standardized preprocessing to form a data stream with a unified space-time reference; performing rationality check on the data based on an environmental physical prior knowledge base; constructing a space-time neighborhood window and calculating a multi-scale consistency index; generating a dynamic credibility score in combination with historical statistical characteristics; determining the validity weight factor of each data according to the test result and the credibility score; and finally, performing adaptive weighted fusion based on the weight factor to generate a high-confidence environment state estimation result. The system correspondingly comprises a multi-source data access module, a standardized preprocessing module, a physical rationality test module, a dynamic credibility evaluation module, a self-adaptive fusion module and the like. According to the method, the accuracy and decision support capability of multi-source environment data fusion are effectively improved by fusing dynamic evaluation of physical rule constraint and data driving.

The application discloses a standard courseware generation method for an artificial intelligence learning mode, and relates to the technical field of educational informatization. First, course standards, world language readiness frameworks and translation memories are collected, and a reference semantic vector set with a unique version fingerprint is generated by vectorization. Then, a multi-language courseware draft is mapped thereon, a weighted difference vector is output, and a time stamp is recorded. Subsequently, the difference vector is decomposed into three factors of emotional tendency, cultural symbol and term consistency, a deviation registration table is generated according to a dynamic threshold, and then, according to a rewriting energy function, a generated rewriting, template replacement or manual prompt is selected, a high-risk language block is locally revised, and a consistency mark containing a version fingerprint is written. Finally, the global residual error is rechecked, the quality is confirmed through a global consistency index, a consistency report is generated, the revised courseware is packaged as a release package for multi-end synchronization, a fingerprint is written on the chain, and a new cross-language courseware quality closed loop is constructed.

The embodiment of the application provides a kind of reliability analysis method and system of embodied model, medium, equipment and product, the method comprises: receiving the target video of target agent execution operation task generated by the embodied model;One or more of motion consistency analysis, visual consistency analysis and logic rationality analysis are carried out on the target video, to obtain one or more of motion consistency index value, visual consistency index value and logic rationality index value;Based on one or more of the motion consistency index value, the visual consistency index value and the logic rationality index value, determine the reliability analysis result of the embodied model.The embodiment of the application can effectively evaluate the reliability of embodied model in actual task scene by analyzing the target video of target agent execution operation task generated by embodied model, to obtain key index.

The present application belongs to the technical field of geophysical exploration, and provides a tunnel advanced geological prediction automation evaluation and feedback correction method and system, which performs weighted summation on interval coincidence degree indexes, geological risk consistency indexes, geological type consistency indexes, hydrogeological consistency indexes and surrounding rock grade consistency indexes to determine a comprehensive score; adjusts the index weight according to a preset adjustment instruction, and corrects the evaluation benchmark; determines a unified standardized evaluation index, establishes an automatic feedback link from exposure to evaluation to correction, changes the traditional post-event static analysis mode, enables the prediction system to quickly learn from historical errors, forms self-adaptive correction capability, effectively prevents the repeated occurrence of similar errors, and significantly improves the timeliness of construction safety guarantee and the continuous optimization capability of the prediction model.

The invention relates to an automatic detection method and system for air duct design of an electric leaf blower in the technical field of intelligent manufacturing, and the method comprises the steps: obtaining initial air flow distribution data from air duct design parameters through simulation software, calculating an energy loss value through a finite volume method, and obtaining an air flow characteristic index; if the noise level meets the standard, the optimized airflow distribution model is imported into a production database, an automatic detection sequence is obtained, and an internal flow field scanning path is determined; according to the obtained average wind speed distribution diagram, if the local wind speed deviation is larger than a threshold value, a calibration mechanism is triggered, and a corrected product consistency index is obtained; key features are extracted from the corrected product consistency indexes, noise source types are classified by adopting a support vector machine, and whether the overall performance improves the market competitiveness or not is judged; simulating software parameters are updated through the judged overall performance data, next round of air duct design iteration input is obtained, and a final automatic detection report is obtained.

This invention discloses a deep learning-based intelligent anomaly safety detection system for industrial dehumidifiers, comprising the following modules: a behavior pair construction module, used to establish behavior parameter and inverse behavior parameter pairs from multi-source parameters and set the response direction, coupling strength, and allowable delay; a consistency index calculation module, used to calculate instantaneous consistency, trend consistency, and rate consistency and generate a consistency score; a response relationship determination module, used to determine the response relationship of the pairs and generate a mismatch marker; an anomaly event generation module, used to construct anomaly events and determine the safety level; a linkage control module, used to invoke control strategies and output control commands to the execution components; and a status output module, used to generate anomaly types and safety statuses and output them to the monitoring terminal. This invention, by constructing behavior pairs and quantifying multi-dimensional consistency indices, achieves highly reliable anomaly detection, safety level determination, and linkage control output under complex operating conditions.

The invention discloses a PET facing artificial board fitting analysis method and system based on multi-point sensing fusion, and relates to the technical field of data analysis. Collaborative consistency among modals is evaluated through a cross mapping residual consistency index, stability of signals of the modals is measured in combination with a local variation spectrum jump index, and fitting analysis of the PET facing artificial board is realized. Dual-angle modal credibility judgment is realized, and whether the modal credibility is suitable for participating in fusion analysis or not is comprehensively expressed by a unified reliability judgment index; when a certain mode drifts, is shielded or fails, the system can automatically identify the mode as a sensing failure mode and rejects or downloads the weight, so that the abnormal data is prevented from interfering with the fusion result, the stability and accuracy of the fitting analysis system in a complex field environment are remarkably improved, the quality judgment is ensured to be based on high-reliability data, and the reliability of the fitting analysis system is improved. And reliable support is provided for subsequent defect identification and control decision.

The application discloses a metamaterial design method based on density constraint and explainable AI, and relates to the technical field of additive manufacturing; geometric data of the metamaterial is acquired to generate a three-dimensional model, a topographic parameter extraction script is run to obtain a topographic parameter vector, and a mechanical property vector of the metamaterial is acquired, a density constraint condition is set according to the metamaterial, a candidate enhancement strategy is generated for a training set composed of the topographic parameter vector and the mechanical property vector, a consistency index is calculated according to the topographic parameter vector, and an optimal enhancement strategy is selected, an optimal model is obtained through model selection based on an integrated learning model family, global and individual explanations are performed through SHAP, and key topographic parameter features are identified, and the action trend and sensitive interval of the key topographic parameter features on the target mechanical property vector are identified; the metamaterial design method based on the density constraint and the explainable AI is used, the reliability of model prediction is enhanced, and the cost of traditional test iteration and parameter trial and error is significantly reduced.

The invention provides a steady-state operation point drift sensitive index construction method based on alignment data, and belongs to the technical field of metallurgical process control, and the method comprises the steps: firstly, calculating the total causal time difference based on the strip steel operation speed and the furnace area length, carrying out the space-time alignment of the plate temperature and each furnace area variable, and screening a steady-state reference area to construct a baseline statistic; then, on the basis of the aligned data and baseline statistics, a standardized residual error, a drift rate and a temperature-energy consistency index are calculated on line; and finally, carrying out smoothing processing on the indexes, carrying out fusion calculation on the comprehensive drift intensity of each furnace area and the plate temperature, when the comprehensive drift intensity exceeds a preset threshold value, judging that steady-state operation point drift occurs, and identifying a leading furnace area and a drift direction. Through multi-dimensional index fusion, the problems that in the prior art, slow drifting is not sensitive, and a drifting source cannot be accurately positioned are effectively solved, early and accurate detection of steady-state operation point drifting is achieved, and reliable guarantee is provided for stable operation of the production process.

The invention discloses a metamaterial design method based on density constraint and interpretable AI, and relates to the technical field of additive manufacturing. The method comprises the following steps: acquiring geometric data of a metamaterial to generate a three-dimensional model, operating a morphology parameter extraction script to obtain a morphology parameter vector, acquiring a mechanical property vector of the metamaterial, setting a density constraint condition according to the metamaterial, generating a candidate enhancement strategy for a training set composed of the morphology parameter vector and the mechanical property vector, according to the morphology parameter vector, calculating a consistency index and selecting an optimal enhancement strategy, carrying out model sea selection based on an integrated learning model family to obtain an optimal model, explaining the overall situation and individuals through SHAP, and identifying key morphology parameter feature morphology parameters and an action trend and a sensitive interval of the key morphology parameter feature morphology parameters on a target mechanical property vector; according to the metamaterial design method based on the density constraint and the interpretable AI, the reliability of model prediction is enhanced, and the traditional test iteration and parameter trial and error cost is remarkably reduced.

The invention discloses a two-stage investment portfolio optimization method achieved based on fuzzy preference and group consensus, which is more practical by providing an acceptable additive consistency framework of FPR, defining a consistency index and setting a threshold value and establishing a consistency evaluation standard conforming to an actual decision-making scene, and has the advantages that the method is more practical by constructing a two-stage FPR adjustment optimization framework, and the optimization efficiency is improved. In the first stage, the minimum adjustment distance is determined, and in the second stage, investment portfolio selection is optimized on the premise that the minimum information loss is guaranteed; according to the method, original preference information of experts is reserved to the maximum extent while mathematical preciseness is kept, and when the individual consensus level, the group consensus level and an acceptable group consensus threshold value are lower than the threshold value, a system automatically identifies the lowest expert and performs adjustment by applying a two-stage optimization model; according to the iteration process, individual consistency and investment portfolio performance constraints are ensured to be met in each iteration, and individual consistency and group consensus are effectively coordinated.

The present application provides a full life cycle demand intelligent analysis system and method, relates to the product life cycle management technical field, and the present application constructs a unified data set and extracts standardized demand characteristic parameters by acquiring multi-source demand data of each stage of the product life cycle; a demand evolution index is constructed to realize dynamic quantitative evaluation of the demand state, and a threshold value is determined to trigger a graded demand adjustment strategy; a design consistency index and a demand stability index are constructed to realize two-dimensional analysis of system consistency and stability, and a differentiated collaborative optimization strategy is triggered; a demand deviation propagation index is constructed to quantitatively analyze the deviation source and propagation path, and a closed-loop optimization control is realized in combination with a convergence determination mechanism. The present application can realize accurate identification and dynamic regulation and control of the demand evolution process, and improve the system operation stability and the intelligent level of demand management.

The application discloses a river ecological abnormal change identification method based on machine learning, relates to the technical field of machine learning, and comprises the following steps: collecting multi-section ecological monitoring data, preprocessing, and constructing a time-space sequence dataset; a river directed topological structure is constructed, a topological adjacency matrix is generated, and the association mapping of time-space data and nodes is completed; the time-space characteristics and the adjacency matrix are inputted, the diffusion law is learned by improving DCRNN, and a dynamic ecological baseline sequence is outputted; a stable ecological sample set is constructed, a stable subspace is generated by mapping, the baseline is corrected, and the structural deviation is calculated; an ecological energy transmission interval matrix is constructed, bidirectional propagation consistency discrimination is performed, and a propagation consistency index is obtained; the baseline residual error is calculated, the structural deviation is combined, and an abnormal identification result is outputted. The application realizes accurate identification and early warning of gradual ecological abnormal changes in the river by constructing a topological constraint dynamic ecological baseline and fusing stable ecological structure and propagation consistency discrimination.

The application provides a parameterized representation method for quantifying building energy consumption curve prediction accuracy and difficulty. The method comprises: dividing building groups according to types; constructing a building type consistency index, which is used to measure the similarity of the same type of buildings in electricity consumption behavior; obtaining the building type consistency index of the same type of building group; constructing an electricity load prediction model of the same type of building group; inputting the historical electricity consumption data set of the same type of building group into the electricity load prediction model of the same type of building group to obtain the prediction accuracy of the electricity load prediction model of the same type of building group; and establishing a first mapping relationship between the building type consistency index of the same type of building group and the prediction accuracy of the electricity load prediction model of the same type of building group. The application can assist in completing the data screening of large-scale building group energy consumption prediction tasks.