29 results about "Evaluation period" patented technology

The invention provides a performance testing method and device for a large language model and related equipment. The performance testing method comprises the steps that multiple testing platforms for testing, testing scene configuration parameters and a structured testing data set for testing are determined; on the basis of the test scene configuration parameters and the structured test data set, synchronous constraint parameters are generated and distributed to the multiple test platforms, and then a preset load generator is used for controlling the multiple test platforms to concurrently execute reasoning test tasks about the large language model; and when the test is executed, collecting multiple dimension performance indexes of each test platform in real time for data analysis, and generating a reasoning performance comparison evaluation report of the large language model in the multiple test platforms according to a data analysis result. According to the method, the cross-platform performance evaluation period is greatly shortened, and the waste of computing resources is reduced. Manual operation errors are avoided, comparability of multi-platform performance indexes is guaranteed, and more accurate and reliable multi-platform performance test results can be obtained.

The invention discloses an employee ability growth navigation and resource recommendation system and method, and relates to the technical field of employee growth navigation and resource recommendation. Comprising a learning map construction module, a multi-modal ability evaluation module, a dynamic navigation engine, a resource recommendation module, a digital twinborn training module and a user interaction module. A learning map construction module; the multi-modal fusion realizes' voice-text-image-equipment 'four-dimensional evaluation, the evaluation dimension is expanded from traditional 3 items to 8 items, and the evaluation accuracy of the cross-department cooperation capability and the fault decision-making capability is improved; the assessment weight is dynamically adjusted through reinforcement learning, the technical trend matching degree is improved, the assessment period is shortened to the hour level from the quarterly level, and the talent inventory efficiency is improved; a composite fault scene is dynamically generated, practical training pertinence is improved, and the processing success rate of employees in a multi-equipment linkage failure scene is improved.

The application provides a material arrival balance evaluation method, system, device and storage medium, the method comprises the following steps: determining a target evaluation period in each evaluation period; determining a single-species material plan exchange rate of a target evaluation material according to a single-species plan arrival amount and a single-species actual arrival amount of each material in the target evaluation period; determining a single-species material execution rate according to the single-species material plan exchange rate and a preset processing rule; and determining a total plan arrival execution rate of all target evaluation materials in the target evaluation period according to the single-species material execution rate, the single-species plan arrival amount, a non-plan variety arrival amount and a total plan arrival amount of all materials in the target evaluation period. In the scheme, the single-species material execution rate of each evaluation period can be combined with viewing to know the stability of the single-species material in each evaluation period. Through the calculation of the plan arrival execution rate, the execution situation of all varieties of materials in the target evaluation period can be determined.

The invention discloses a rental item equipment value evaluation method and a related device, and relates to the technical field of asset evaluation. The method comprises the following steps: acquiring a lease item equipment information list; inputting the plurality of characteristic parameters of each piece of lease item equipment into a lease item value evaluation model, and determining a predicted evaluation value of each piece of lease item equipment; and for each piece of lease item equipment, determining a final evaluation value of the lease item equipment based on the plurality of characteristic parameters of the lease item equipment, the predicted evaluation value of the lease item equipment and the target database. The whole process does not need a large amount of manual intervention of complex parameter comparison and value accounting procedures, so that the process is remarkably simplified, and the evaluation period is shortened; the prediction precision is improved; according to the method, the problems of complex process, long examination and approval time, complex affirmation process and relatively low working efficiency of a traditional value affirmation process of the lease item equipment are solved to at least a certain extent, and the value of the lease item can be determined more efficiently and more accurately.

The invention provides a bridge support working condition early warning method and device, a storage medium and a bridge support working condition multifunctional monitoring method, and the method comprises the steps: judging the type of a to-be-evaluated bridge support, and obtaining a disease affecting the use function of the bridge support; determining a working condition evaluation index system of the bridge support based on the diseases; obtaining the evaluation grade of each evaluation index; obtaining a representative value sequence of the monitoring value of each evaluation index in the to-be-evaluated time period, and determining the weight of each evaluation index based on the representative value sequence; calculating the overall score of the working condition of the bridge support based on the evaluation grade and the weight of each evaluation index; obtaining a dynamic early warning value corresponding to the to-be-evaluated time period based on the traffic load borne by the bridge above the bridge support in the to-be-evaluated time period and the working duration of the bridge support; when the overall score of the working condition of the bridge support is larger than the dynamic early warning value, early warning is triggered. By adopting the early warning method provided by the invention, rapid and accurate early warning of the working condition of the support can be realized.

The invention provides a coal supplier comprehensive evaluation method and system based on AHP and PCA, and relates to the technical field of comprehensive evaluation, and the method comprises the steps: building a comprehensive evaluation index system, and obtaining index basic data, including objective data and subjective data; based on the index basic data, determining the index weight of the evaluation index by adopting a method of combining AHP and PCA; respectively measuring and calculating according to the index weight and the corresponding organization level and evaluation period to obtain a comprehensive evaluation score; and analyzing the organization hierarchy based on the comprehensive evaluation score to obtain an analysis result. According to the method and the device, the technical problem that the management efficiency and decision scientificity of the coal supply chain in the aspects of quality control, contract execution, safety guarantee and resource allocation are further influenced due to the fact that the comprehensive evaluation result of the supplier is not accurate enough in the prior art can be solved, and the technical effect of improving the comprehensive evaluation accuracy of the coal supplier is achieved.

The application discloses a kind of volunteer service recommendation method and system, wherein method includes the following steps: based on the skill score corresponding to each skill label is quantified in the skill evaluation period of pre-set, the skill score is used to represent the influence of the exercise of corresponding skill on the personal ability improvement of volunteer;When receiving the task matching request sent by volunteer, based on the task matching request, obtain the task to be taken of matching with the volunteer, obtain the corresponding candidate task, determine the recommendation degree of the candidate task based on the skill score of each skill label corresponding to candidate task, generate the corresponding task recommendation data based on the candidate task and the recommendation degree;The application determines the recommendation degree of the candidate task by the skill score of each skill label corresponding to candidate task, carries out task recommendation to volunteer based on the recommendation degree, facilitate volunteer to take the volunteer service task that both meets task demand and is beneficial to self growth.

The invention relates to an enterprise personnel configuration and salary accounting collaborative optimization method, which comprises the following steps: in an enterprise business operation process, acquiring task execution data, identifying hidden redundant behaviors existing in posts, and generating a post adjustment instruction when the redundancy rate reaches a preset threshold value; when it is detected that task cross execution, responsibility overlapping or permission repetition exists among a plurality of posts, executing a responsibility boundary recognition process, determining post function boundary adjustment requirements by comparing task sources, execution paths and approval chains, and outputting post configuration optimization instructions; for post change implemented according to the optimization instruction, monitoring human effect change of a corresponding post in a set evaluation period and generating an adjustment measure; after the salary strategy adjustment is executed, the performance change and the cooperative behavior of the high-performance personnel are monitored, and a targeted salary correction scheme is generated, so that the dynamic cooperative optimization of personnel configuration and salary accounting is realized.

The invention provides a production and education integration quality evaluation method based on CRITIC subjective and objective combination, and belongs to the technical field of education data analysis, and the method comprises the steps: building an evaluation system of three types of first-level indexes of input, process and output and corresponding second-level indexes based on an IPO model; carrying out abnormal value processing and standardization on the original data; an objective weight is calculated by adopting a CRITIC method, a subjective weight is determined in combination with an expert consensus method, and an annual optimal combination weight is solved through a combination optimization model; generating a target evaluation period dynamic weight based on weight time sequence prediction and development stage correction; and calculating a comprehensive score of each evaluation object and generating a multi-dimensional evaluation report. Through subjective and objective combination of weighting balance data rules and professional consensus, evaluation requirements of different development stages are adapted by means of dynamic weights, evaluation accuracy and guidance are effectively improved, closed-loop management of production and teaching fusion quality evaluation, diagnosis and improvement is realized, and a scientific decision basis can be provided for education management departments, colleges and enterprises.

The invention discloses an energy-saving and carbon-reducing evaluation method and system for a pulverized coal boiler. The method comprises the following steps: collecting operation data and basic information of the pulverized coal boiler through a sensing monitoring terminal, and uploading the operation data and the basic information to a server for initialization; receiving power plant evaluation requirements and determining priorities; screening effective historical operation data in the evaluation period; constructing a two-dimensional evaluation index system based on the effective data, optimizing index weights by using an intelligent optimization algorithm, and constructing a target evaluation function; establishing an evaluation model according to the target evaluation function, judging an energy-saving and carbon-reducing evaluation grade, and giving out early warning when the evaluation grade does not reach the standard; an evaluation result is fed back, data are counted and analyzed, and support is provided for follow-up optimization. According to the method, the problems of single dimension and low precision of traditional evaluation are solved, the evaluation accuracy is improved through comprehensive data acquisition, scientific evaluation system construction and intelligent algorithm optimization, reliable data support is provided for boiler analysis, diagnosis and optimization improvement, and reduction of carbon emission and operation cost is facilitated.

The invention discloses a cloud industrial software-oriented knowledge governance and quality evaluation method and system, and relates to the technical field of knowledge evaluation.The state information of data in a knowledge governance unit is obtained, the burst oscillation index is calculated, and whether the quality evaluation period needs to be adjusted or not is judged according to the burst oscillation index; if adjustment is needed, extracting influence degree information of data oscillation, and calculating a period change index; determining a final evaluation period according to the index and a preset gradient threshold value; and the quality evaluation unit evaluates the data according to the data and feeds back the data to the knowledge governance unit. According to the method, when the knowledge state changes suddenly, whether the evaluation period is adjusted or not can be intelligently decided according to the data change degree and type, blind adjustment is avoided, it is ensured that low-quality knowledge is recognized and processed in time, misleading reasoning and error accumulation are reduced, the accuracy of downstream industrial decisions is improved, and meanwhile resource waste and excessive cleaning are prevented.

The invention discloses an equipment operation state processing method and device, equipment, a medium and a product. The method comprises the following steps: if it is detected that current time meets an evaluation period condition preset for an equipment class, obtaining a period measured value of the equipment class under each preset evaluation index from a data table through a configured data extraction rule; the evaluation indexes cover the five fields of power generation health, equipment fault, state monitoring, reliability management and experience feedback; determining an actual score and a color grade of each evaluation index of the equipment class according to a pre-configured total score of the evaluation indexes, a scoring criterion of the evaluation indexes and a periodic measured value; determining the total score of the operation state of the equipment class according to the actual scores of all the evaluation indexes under the equipment class; and generating a structured equipment class health evaluation report based on the operation state total score of the equipment class, the actual score of each evaluation index and the color grade. The embodiment of the invention can improve the equipment operation management efficiency and the unit operation reliability.

The invention relates to the technical field of industry development evaluation, in particular to an emerging industry development evaluation method based on electric power big data and a related product, and the method comprises the steps: collecting multi-dimensional electric power data of emerging industry enterprises, and obtaining standardized electric power data; constructing a multi-dimensional evaluation index system, and determining a weight; performing quantitative scoring on each index; calculating to obtain a comprehensive evaluation score; obtaining a development prediction score; according to the method, multi-dimensional power data is used as a core data source, the Lagrange interpolation method and the normalization method are adopted for preprocessing, manual report errors and hysteresis are avoided, the evaluation period is shortened to monthly, objective weighting is achieved in combination with the entropy weight method, the influence of subjective factors of experts is eliminated, and the evaluation efficiency is improved. And the defects of unreliable data and unscientific weight of the traditional method are thoroughly overcome.

The application discloses a risk assessment method and device of equipment, which is used to improve the accuracy and comprehensiveness of the risk assessment of the equipment. The method comprises the following steps: obtaining operation data of the equipment to be evaluated in an evaluation period; determining at least one business scenario related to the equipment according to the operation data, and determining a plurality of actions performed by the equipment in the evaluation period; determining an abnormality level of each business scenario according to each action performed in each business scenario and a change degree of the execution information of each action compared with the execution information of each action in a previous period of the evaluation period; and determining a risk value of the equipment in combination with the abnormality level of the at least one business scenario and preset abnormality values of the plurality of actions.

The invention provides a performance appraisal system and method, and the method comprises the steps: firstly judging whether a current performance appraisal rule needs to be optimized and adjusted, calculating an optimized index target value under a current unit load according to the measured index data and a preset index deviation ratio when it is confirmed that optimization and adjustment need to be carried out, and carrying out the calculation in a circulation process, and obtaining optimization index target values under different unit loads, forming an optimization performance assessment rule by the unit loads and the corresponding optimization index target values, and when the initial moment of the next assessment period is reached, updating the current performance assessment rule into the optimization performance assessment rule. Therefore, automatic and intelligent optimization setting of the performance assessment rules can be realized, the influence of human factors on the performance assessment rules is reduced, and the safe, efficient and reliable operation level of a unit is improved.

The invention provides a vocational education production and education fusion quality evaluation system and an evaluation label generation method. The method comprises the following steps: constructing an evidence set for carrying out production and education fusion quality evaluation on a target evaluation subject; extracting evaluation evidence characteristics of the college side and the enterprise side from the evidence set, and constructing an evidence atlas reflecting production and teaching fusion quality based on statistical association strength of the evaluation evidence characteristics of the college side and the enterprise side; performing multi-view correlation analysis on evidence nodes in the evidence graph, and determining contribution weights of the evidence nodes in combination with credibility and timeliness of evidence sources; and detecting the characterization consistency between the last evaluation period label and the current evaluation evidence feature, when the consistency is detected to be lower than a preset threshold value, triggering a label updating mechanism, and generating an updated quality evaluation label according to the contribution weight of each evidence node and the current evaluation evidence feature. By adopting the scheme of the invention, the dynamic quantitative evaluation of the production and teaching fusion quality can be realized, so that the accuracy of the evaluation result can be improved.

A power load curve fluctuation degree evaluation method, device and equipment, the method comprises the following steps: according to the power load value of each sampling point in the power load curve in the preset evaluation period, the average load is calculated;According to the power load value and the average load, the total fluctuation energy is calculated;According to the average load, the total fluctuation energy and the total sampling point number, the text wave coefficient is calculated to evaluate the fluctuation degree of the power load curve. By introducing the evaluation period which can be flexibly set, the fluctuation details of different time scales can be captured;At the same time, the text wave coefficient is a dimensionless index, and the value directly corresponds to the strength of the fluctuation degree, which enhances the comparability of the evaluation result, effectively improves the comprehensiveness, fineness and accuracy of the power load curve fluctuation degree evaluation.

To evaluate the use form of a building by a plurality of users from the viewpoint of the degree of interaction between the users and the degree of appropriateness based on the life style of the user or the like, and to grasp the difference between appropriate use forms estimated from the respective viewpoints.SOLUTION: According to one embodiment of the present invention, the processor 21 specifies, for each of a plurality of evaluation periods, a target time period in which two or more residents are present at different locations in a predetermined time zone in the evaluation period, acquiring first input information indicating a first evaluation value related to a degree of interaction of the resident and second input information indicating a second evaluation value related to a degree of appropriateness of the target time, estimating a target time at which the first evaluation value becomes a first set value based on a first correspondence relationship between the target time and the first evaluation value, estimating a target time at which the second evaluation value becomes a second set value based on a second correspondence relationship between the target time and the second evaluation value, and calculating a ratio between estimated values of the two estimated target times; SELECTED DRAWING: Figure 2

The application provides a reinforcement learning method suitable for inhibiting performance fluctuation, a reinforcement learning intelligent agent containing an online policy network and a target policy network queue is constructed, a main experience pool and an evaluation experience pool are combined, and stable evolution and efficient convergence of a policy are realized. In an environment interaction process, the online policy network and the optimal policy in the target policy network queue are alternately sampled and obtain environment interaction information, and the environment interaction information is stored in the main experience pool. In a preset evaluation period, a parallel multi-environment rolling evaluation mechanism is used to obtain policy evaluation rewards, and the target policy network queue is dynamically updated or a policy rollback is performed based on the evaluation results. A double evaluation network is introduced, and an adversarial evaluation training mechanism based on double Q mean value and small double Q random sampling is used to inhibit value function overestimation and policy over-conservatism. Interaction data generated in the evaluation stage is stored in the evaluation experience pool, and the interaction data is used in combination with main experience pool data in the training process, so that the stability, robustness and convergence performance of the reinforcement learning training are improved.

The application discloses an operation and maintenance data processing method and system for an AI intelligent agent, forms a heat data sequence through collecting access indexes of each shard of a storage cluster; calculates a heat value and a change degree to determine a target storage layer and compare the target storage layer with a current layer to obtain consistent and wrong layer candidates; constructs a hierarchical consistency index and a tolerance threshold to determine consistency; implements speed limiting, redirection and priority reduction isolation on the wrong layer shard, measures performance loss based on a historical baseline within an evaluation period, migrates and corrects when the performance loss is lower than a threshold, triggers an alarm and degrades when the performance loss exceeds the threshold; and outputs resource occupation and time delay after correction, gives resource configuration suggestions and hierarchical results. The method suppresses migration jitter, reduces tail time delay and cost, and improves throughput and stability.

The invention discloses a timeliness evaluation method and device for a service system, electronic equipment and a storage medium, and relates to the field of system operation and maintenance and financial science and technology, and the method comprises the steps: obtaining the completion time of each service node for processing each task in a preset evaluation period, the number ratio of the number of the tasks processed by each service node in all the tasks is determined; according to the completion time of the overtime task processed by each service node and the reference time, acquiring lagging time efficiency of each service node; and according to the lagging timeliness and the number proportion of each service node, obtaining the sensitive timeliness of each service node, and according to the sensitive timeliness of each service node, obtaining a timeliness evaluation result of the service system. According to the technical scheme, it is ensured that the timeliness evaluation result truly reflects the task processing performance of the service system in the evaluation period, and the accuracy and comprehensiveness of the timeliness evaluation result of the service system are greatly improved.

The invention discloses an operation and maintenance data processing method and system for an AI agent. The method comprises the steps that access indexes of all fragments of a storage cluster are collected to form a popularity data sequence; calculating a heat value and a change degree to determine a target storage layer, and comparing the target storage layer with the current layer to obtain consistent and staggered layer candidates; constructing a hierarchical consistency index and a tolerance threshold to judge the consistency; speed limiting, redirection and priority reduction isolation are carried out on the split-level fragments, the performance loss is measured by a historical baseline in an evaluation period, controlled migration and correction are carried out when the performance loss is lower than a threshold value, and alarm and degradation are triggered when the performance loss exceeds the threshold value; and outputting the corrected resource occupation and time delay, and giving a resource configuration suggestion and a layering result. The method inhibits migration jitter, reduces tail time delay and cost, and improves throughput and stability.

The invention relates to the technical field of information, and discloses a capability assessment method and device based on a large model, computer equipment and a medium. First, target role information is determined based on basic information of a first user. Determining first work task trace data of the first user in the target evaluation period based on the target role information and the basic information; and then, based on the first work task trace data, generating target text prompt data corresponding to the first user. And finally, inputting the target text prompt data and the work task trace data of the second user corresponding to the target role information in the target evaluation period into the role evaluation large model, and outputting a capability evaluation result corresponding to the first user. The user is helped to efficiently complete the evaluation process, the user time is saved, and the overall working efficiency is improved. In addition, in combination with analysis of a role evaluation large model, it is ensured that the ability evaluation result is more objective and accurate.

The application relates to the technical field of administration, and particularly discloses an administration method and system based on big data, which comprises the following steps: step S1: determining the post of an employee to be evaluated, and obtaining historical project data of the post; step S2: setting an evaluation period, extracting all work projects of the employee in the period, calculating the similarity by comparing historical project information, and screening similar projects; step S3: obtaining the working years and project periods of historical employees who have executed similar projects, combining the working years of the employee to be evaluated, generating a period distribution curve, and determining a maximum value point period; and step S4: quantifying the performance of the employee according to the comparison between the evaluation period length and the maximum value point period; the application is driven by data, objectively evaluates the efficiency of the employee, and is particularly suitable for performance management of project-based enterprises.

The invention relates to the technical field of logistics engineering, and discloses a logistics park carbon emission and intelligent operation evaluation method comprising the following steps: 1, logistics park basic data statistical survey: statistical survey is carried out on the intelligent operation condition of each operation link in the park since the logistics park is built over the years, and 2, according to the logistics park carbon emission and intelligent operation evaluation method, mathematical modeling analysis is carried out from three aspects of the logistics park throughput carbon emission intensity, the investment return rate and the overall intelligent operation proportion, and comprehensive evaluation of the logistics park is realized. And finally determining high-to-low logistics park carbon emission and intelligent operation evaluation standards of A, B, C, C and the like (corresponding to excellent, good and unqualified), wherein the evaluation results are scientific, comprehensive and objective. According to the method, a three-stage method operation process is provided, the operability is high, the popularization value is high, the evaluation period is shortened while the objectivity of the evaluation result is guaranteed, and the efficiency of carbon emission accounting and evaluation work is improved.

This invention relates to the field of automated dispensing equipment management technology, specifically to an automated dispensing management system and method based on artificial intelligence. Upon detecting a new fault report, the invention triggers an assessment of the dispensing control platform within the corresponding work evaluation period. It determines whether this is due to performance degradation. The analysis involves analyzing the continuous operational distribution of the dispensing robot within the evaluation period and the frequency distribution of challenging task change nodes. Ultimately, this allows for the evaluation of the dispensing control platform's operational performance index within the evaluation period. This invention effectively monitors and addresses aging or damage caused by prolonged operation of the dispensing control platform, promptly identifying and resolving performance degradation issues.

The invention provides a cross-team evaluation method and system based on adversarial learning and an optimal transmission theory, and the method is applied to an evaluation system which comprises a data processing module, a capability representation extraction module, a template gap recognition module and an evaluation module. The method specifically comprises the following steps: constructing a corresponding multi-dimensional capability feature vector based on multi-dimensional working data of each team employee in a predetermined assessment period; generating corresponding capability representation based on adversarial learning; in combination with an optimal transmission theory, obtaining a template difference between the template distribution and preset template distribution; and calculating a performance alignment score of each employee based on the template gap, and mapping the performance alignment score to a preset score interval to obtain a fair performance index. According to the method, adversarial learning is utilized to extract team and group irrelevant ability characterization, the unified value scale is established in combination with the optimal transmission theory, and then team and group systematic deviation is avoided through accurate alignment of the ability characterization and the unified value scale, and the accuracy and reliability of a final evaluation result are improved.

The invention discloses a safety production standardized score management method, and the method comprises the steps: obtaining a corresponding element rating according to an actual score corresponding to each first-level element of each evaluation in an evaluation period when a distribution interval of a total score of evaluation of each expert in the evaluation period is between a critical point of a rating threshold value; calculating the weight of each first-level element and the connection degree corresponding to each evaluation, obtaining a transfer matrix through the conversion of element rating in two adjacent evaluation processes, obtaining an evaluation connection degree according to the average value of the connection degrees and the average value of the transfer matrix, and finally selecting a final score according to the evaluation connection degree and carrying out grade judgment. Meanwhile, the invention also discloses a safety production standardization scoring management device for realizing the safety production standardization scoring management method. The method can reflect the safety management level trend of a certain enterprise in a certain time period, is a powerful reference basis for grade judgment, and can reduce the subjectivity of a traditional scoring mode.

This invention discloses a human resources performance evaluation system based on big data analysis, belonging to the field of big data technology. It organizes relevant business data of target employees within a preset evaluation period into a performance event set (Evt), a target record set (Tar), and a responsibility record set (Rol). Then, it identifies change node results (Nod) and generates interval division results (Seg). This accurately separates evaluation stages within the same preset evaluation period that change due to target or responsibility adjustments, and then generates segmented performance results (Prf), percentage results (Wgt), final performance results (Fin), and explanatory results (Exp) for each stage. This avoids the mixing of work before and after target changes, ensuring that the results of the previous stage are not diluted by the evaluation conditions of the later stage, and that the high-pressure handling work of the later stage is not averaged over by the stable tasks of the previous stage. The interval division result (Seg) clearly separates the two stages.