CN-121803988-B - Resource allocation method and system based on Internet of things digital calorimeter

CN121803988BCN 121803988 BCN121803988 BCN 121803988BCN-121803988-B

Abstract

The application discloses a resource allocation method and a system based on an Internet of things digital calorimeter, belongs to the technical field of Internet of things and resource allocation management and control, and aims to solve the problems that the traditional resource allocation lacks personalized adaptation, abnormal identification is inaccurate and long-term adaptation is poor. The method comprises the steps of configuring individual resource adaptation coefficients of users in an installation and debugging stage, collecting and aligning resource allocation parameters and guide parameters through a digital calorimeter of the Internet of things in an operation stage, constructing a periodical data set and quantifying deviation parameters through correlation coefficients, screening historical data to construct a standard resource demand curve, generating a personalized demand prediction curve by combining real-time deviation and historical deviation correction after matching prediction time length, generating and dynamically correcting a resource allocation scheme based on the prediction curve and the adaptation coefficients, identifying and distinguishing sudden and persistent anomalies, and accurately correcting the adaptation coefficients for the persistent anomalies. The application realizes personalized and accurate control of resource allocation, and improves the resource utilization efficiency and the long-term stability of the system.

Inventors

WANG YUNBO
XU LIANG
CHEN Jin
LU JIAYUAN
QIN YUNZHEN
Chao Mengfan
HE HUILI

Assignees

上海熊猫机械(集团)有限公司

Dates

Publication Date: 20260512
Application Date: 20260309

Claims (4)

1. The resource allocation method based on the digital calorimeter of the Internet of things is characterized by comprising the following steps of: In the installation and debugging stage, configuring individual resource adaptation coefficients of users; The individual resource adaptation coefficients are used for mapping transmission distance differences, terminal characteristic differences and equipment use state differences of different users and comprise resource transmission adaptation coefficients, terminal efficiency coefficients and equipment operation compensation coefficients, wherein the resource transmission adaptation coefficients are calculated based on material heat conduction coefficients, lengths and pipe network design pressures of transmission pipelines and combined with loss characteristics in a resource transmission process; in the resource allocation operation stage, acquiring resource allocation parameters and resource allocation guide parameters of a user through an Internet of things digital calorimeter, aligning data, associating individual resource adaptation coefficients, constructing a periodical allocation data set, and quantifying resource allocation deviation parameters; The resource allocation parameters are parameter sets which directly reflect the real-time running state of a user side resource allocation system and are directly related to indoor resource allocation effects, and have the core functions of providing basic data support for calculating actual resource consumption of a user, monitoring the real-time state of indoor resource allocation and subsequent resource allocation decision feedback, wherein the basic data support comprises resource supply temperature, backwater temperature, hot water flow and actual resource supply temperature; The resource allocation guiding parameters are parameter sets for providing decision basis for formulating personalized resource allocation control strategies and predicting user resource demands, and have the core functions of reflecting heat source supply capacity, external environment influence and user resource allocation demand targets, providing reference standards for resource demand prediction curve construction and resource allocation decision strategy formulation, wherein the resource end input temperatures comprise resource end input temperatures, environment temperatures and resource demand target temperatures, the resource end input temperatures are hot water initial temperatures output to a resource allocation pipe network by a heat source, the hot water initial temperatures are acquired in real time through a water inlet side temperature sensing component integrated with an Internet of things digital heat meter or acquired by a linkage resource allocation supervision platform, the environment temperatures are acquired in real time through an outdoor temperature sensing component built in the Internet of things digital heat meter or acquired by a docking area meteorological data platform, and the resource demand target temperatures are acquired by autonomous setting of a mobile phone APP (application) associated with the user logging in an Internet of things or acquired based on environment temperature matching generation; the step of quantifying the resource allocation deviation parameter comprises: Carrying out outlier identification and elimination on the resource allocation parameters and the resource allocation guide parameters acquired by the digital calorimeter of the Internet of things, carrying out fitting filling on missing data of the real-time resource allocation guide parameters based on the historical resource allocation guide parameters, and marking an acquisition time stamp; According to the marked acquisition time stamp, performing data alignment on the resource allocation parameters and the resource allocation guide parameters through time stamp synchronous matching and acquisition frequency adaptation; The resource allocation period is configured, the resource allocation parameters after data alignment and the resource allocation guiding parameters are subjected to period division according to the resource allocation period, a periodical allocation data set is generated, and the periodical allocation data set is associated with an individual resource adaptation coefficient of the current resource allocation period; Based on the periodical allocation data set, implementing resource allocation deviation parameter quantification by combining the individual resource adaptation coefficients; the implementation of the resource allocation deviation parameter quantization by combining the individual resource adaptation coefficients comprises the following steps: Extracting a statistical average value of resource loss differences of the input temperature of the resource end and the supply temperature of the resource in each resource allocation period, and introducing a resource transmission adaptation coefficient associated with the corresponding resource allocation period to correct the resource loss differences to serve as a resource end allocation deviation parameter; Extracting a statistical average value of terminal loss differences of actual resource supply temperature and resource demand target temperature in each resource allocation period, and introducing terminal efficiency coefficients associated with corresponding resource allocation periods to correct the terminal loss differences to serve as terminal allocation deviation parameters; The method comprises the steps of periodically distributing data in a data set, screening a preset number of historical resource distribution period data from the data set, constructing a historical resource distribution set, constructing a reference resource demand curve by extracting resource consumption statistical characteristics of each historical period, matching a predicted time length based on fluctuation amplitude of resource distribution guiding parameters of a current resource distribution period, and carrying out layered progressive correction on the reference resource demand curve by combining real-time resource consumption deviation and resource distribution deviation parameters of each historical period to generate a personalized resource demand prediction curve adapting to the predicted time length; the step of constructing a reference resource demand curve comprises the following steps: Invoking the environment temperature and the resource demand target temperature of each historical period in the periodic allocation data set, performing similarity matching with corresponding parameters of the current resource allocation period, screening historical resource allocation parameter data of the number of preset resource allocation periods through parameter similarity matching, and constructing a historical resource allocation set; dividing the resource allocation period according to the set time granularity for each history resource allocation period after screening, and calculating the history resource consumption of each resource allocation period; summarizing resource consumption data according to the same resource allocation period for all the historical resource allocation periods of the historical resource allocation set, and performing statistical analysis to extract resource consumption statistical characteristics; arranging the time as a horizontal axis according to the sequence of the resource allocation time periods, taking the resource consumption as a vertical axis, taking the average value of the resource consumption of each resource allocation time period as a core data point, and smoothly connecting each core data point according to the sequence of time to form a reference resource demand curve; The specific steps of generating the personalized resource demand prediction curve adapting to the predicted time length are as follows: Acquiring resource allocation guide parameters in a current resource allocation period, calculating fluctuation amplitude of each resource allocation guide parameter, respectively carrying out fluctuation grading on the fluctuation amplitude of the resource allocation guide parameters, and matching the prediction duration of the personalized resource demand prediction curve according to a preset association matching rule; the real-time resource allocation parameters of the resource allocation period which is generated in the current resource allocation period are called, and the actual resource consumption of each resource allocation period is calculated; According to the predicted time length, a reference resource demand curve segment of the current resource allocation period is intercepted, and the reference resource demand curve segment is subjected to hierarchical progressive correction based on the actual resource consumption deviation and the resource allocation deviation parameter of the historical resource allocation set, so that a personalized resource demand prediction curve adapting to the predicted time length is generated; the step of carrying out hierarchical progressive correction on the reference resource demand curve segment by using the actual resource consumption deviation and the resource allocation deviation parameter of the historical resource allocation set comprises the following steps: according to the amplitude and the change trend of the actual resource consumption deviation, the resource consumption predicted value of each resource allocation period in the standard resource demand curve segment is mapped and adjusted through the deviation proportion; Extracting resource allocation deviation parameters of each historical period in a historical resource allocation set, including resource end allocation deviation parameters and terminal allocation deviation parameters, allocating weighted weights of the resource allocation deviation parameters of each historical period according to the parameter similarity of the resource allocation guiding parameters of each historical period and the current resource allocation period, and calculating comprehensive resource allocation deviation parameters; Acquiring real-time resource allocation deviation parameters of a resource allocation period which occurs in a current resource allocation period, including real-time resource end allocation deviation parameters and real-time terminal allocation deviation parameters, calculating the deviation degree of each real-time resource allocation deviation parameter and a corresponding comprehensive resource allocation deviation parameter, and obtaining the comprehensive deviation degree through weighted summation; Performing ratio operation on the comprehensive deviation degree and the current resource demand target temperature to obtain an initial resource consumption adjustment coefficient, and correcting the initial personalized resource demand prediction curve again based on the resource consumption adjustment coefficient; Based on the personalized resource demand prediction curve and related individual resource adaptation coefficients, generating and executing a personalized resource allocation scheme, monitoring dynamic changes of resource allocation parameters and resource allocation guiding parameters in real time so as to carry out feedback correction on the personalized resource allocation scheme, and acquiring environment adaptation type terminal allocation deviation parameters of each resource allocation period, wherein the environment adaptation type terminal allocation deviation parameters are used for judging whether resource allocation abnormality exists or not and distinguishing sudden abnormality from persistent abnormality; the step of generating the personalized resource allocation scheme comprises the following steps: Based on a personalized resource demand prediction curve adapting to the predicted time length, extracting a resource consumption predicted value of each resource allocation period in the predicted time length to acquire an initial theoretical resource flow of each resource allocation period, and constructing an initial resource allocation scheme; according to the equipment operation compensation coefficient in the individual resource adaptation coefficient, mapping equipment operation resource flow error, correcting the initial theoretical resource flow of each resource allocation period of the initial resource allocation scheme, and generating a personalized resource allocation scheme; in the execution process of the personalized resource allocation scheme, aiming at the current resource allocation period, monitoring the dynamic changes of the resource allocation parameters and the resource allocation guide parameters in real time, and judging whether to trigger secondary correction; when the triggering of the secondary correction is judged, the secondary correction is carried out on the resource flow of the current time period of the personalized resource allocation scheme based on the real-time resource end allocation deviation parameter and the real-time terminal allocation deviation parameter of the current resource allocation time period, so as to obtain the final target resource flow of the current resource allocation time period, and the personalized resource allocation scheme is corrected; And aiming at persistent abnormality, identifying the abnormality type of the individual resource adaptation coefficient, and carrying out feedback correction on the individual resource adaptation coefficient.
2. The method for allocating resources based on the internet of things digital calorimeter as set forth in claim 1, wherein the step of determining whether there is a resource allocation abnormality and distinguishing between bursty abnormality and persistent abnormality comprises: in the execution process of the personalized resource allocation scheme, acquiring real-time terminal allocation deviation parameters and real-time environment temperature of each resource allocation period, and acquiring environment adaptive terminal allocation deviation parameters through the deviation calibration processing of the environment temperature interval adaptation; configuring an allocation deviation threshold value, and identifying an abnormal resource allocation period when the allocation deviation parameter of the real-time terminal is larger than the allocation deviation threshold value; counting the duration of an abnormal resource allocation period, if the duration is longer than a preset deviation duration threshold, judging that the resource allocation is abnormal, and starting the distinction between bursty abnormality and persistent abnormality based on abnormal stability analysis; acquiring real-time terminal allocation deviation parameters corresponding to abnormal resource allocation periods of the historical resource allocation set according to the period identification of the abnormal resource allocation periods, and constructing a real-time terminal allocation deviation parameter sequence of each abnormal resource allocation period; and calculating the abnormal stability of each abnormal resource allocation period through the fluctuation amplitude of the real-time terminal allocation deviation parameter sequence, and distinguishing sudden abnormality from persistent abnormality.
3. The resource allocation method based on the internet of things digital calorimeter as set forth in claim 1, wherein the step of performing feedback correction on the individual resource adaptation coefficients includes: When the continuous abnormality is judged, extracting abnormal allocation characteristics including resource supply deviation characteristics, heat exchange efficiency deviation characteristics and execution flow deviation characteristics when a personalized resource allocation scheme of an abnormal resource allocation period in the current resource allocation period is executed; comparing the abnormal allocation characteristics through a preset reference deviation threshold value to identify obvious abnormal types and locate individual resource adaptation coefficients to be corrected; Aiming at the individual resource adaptation coefficient to be corrected, calculating a correction index according to the deviation degree of the abnormal allocation characteristic and the corresponding reference deviation threshold value, and performing weighted correction on the individual resource adaptation coefficient to be corrected; Based on the corrected individual resource adaptation coefficients, synchronously correcting the personalized resource allocation scheme of the resource allocation period which is not executed in the current resource allocation period, updating the abnormal allocation characteristics of the corrected resource allocation period, and recalculating the deviation degree of the abnormal allocation characteristics and the corresponding reference deviation threshold value, so as to judge whether the abnormality is relieved or not, and executing differentiated treatment.
4. The resource allocation system based on the Internet of things digital calorimeter is used for realizing the resource allocation method based on the Internet of things digital calorimeter as set forth in any one of claims 1-3, and is characterized by comprising an individual adaptation module, a deviation quantification module, a demand prediction module, a dynamic optimization module, an abnormality identification module and a feedback correction module; The system comprises an individual adaptation module, a deviation quantification module, a dynamic optimization module, an environment adaptation type terminal, a feedback correction module and an individual resource adaptation module, wherein the individual adaptation module is used for configuring individual resource adaptation coefficients of a user, the deviation quantification module is used for collecting resource allocation parameters and resource allocation guiding parameters of the user through an Internet of things digital heat meter and conducting data alignment, associating the individual resource adaptation coefficients to construct a periodical allocation data set and quantifying the resource allocation deviation parameters, the demand prediction module is used for screening historical resource allocation period data of a preset quantity to construct a standard resource demand curve, matching a prediction duration based on fluctuation amplitude of the resource allocation guiding parameters of a current resource allocation period to generate an individual resource demand prediction curve adapting to the prediction duration, the dynamic optimization module is used for generating and executing an individual resource allocation scheme, monitoring dynamic changes of the resource allocation parameters and the resource allocation guiding parameters in real time to conduct feedback correction on the individual resource allocation scheme, the abnormality identification module is used for obtaining environment adaptation type terminal allocation deviation parameters of each resource allocation period and judging whether resource allocation abnormality exists, and distinguishing bursty abnormality and persistence abnormality exists, and the feedback correction module is used for identifying the individual resource adaptation coefficients.

Description

Resource allocation method and system based on Internet of things digital calorimeter Technical Field The invention relates to the technical field of Internet of things and resource distribution management, in particular to a resource distribution method and system based on an Internet of things digital calorimeter. Background Under the background of enabling resource distribution control by the internet of things technology, accurate and personalized resource supply becomes a core requirement for improving user experience and resource utilization efficiency, and a resource distribution method based on various metering devices is widely applied in actual scenes, but still faces a plurality of technical bottlenecks for restricting industry development. The existing resource allocation scheme mostly adopts unified configuration logic, objective differences of different users in transmission distance, terminal equipment characteristics, equipment use states and the like are not fully considered, so that the problem that resource allocation strategies lack pertinence and are difficult to match individual demands of the users and excessive or insufficient resource supply is easy to occur is caused, meanwhile, a resource demand prediction link often depends on a fixed historical data model, prediction duration is stiff, flexible adjustment cannot be realized according to real-time fluctuation of parameters such as environment temperature, user demand targets and the like, deviation information of all links of resource allocation cannot be effectively fused, a prediction result is disjointed with real-time demands of the users, and allocation accuracy is further lowered. In the aspect of abnormality identification and treatment, the prior art can only preliminarily judge whether abnormality exists, is difficult to strip the influence of environmental interference factors on abnormality judgment, and cannot effectively distinguish sudden interference from persistent system abnormality, so that abnormality treatment lacks pertinence, is difficult to fundamentally solve the problem, and restricts the intelligent level of resource allocation and the improvement of user experience. Therefore, in order to overcome the limitations, the invention provides a resource allocation method and a resource allocation system based on the Internet of things digital calorimeter. Disclosure of Invention Aiming at the defects existing in the prior art, the invention aims to provide a resource allocation method and a system based on an Internet of things digital calorimeter, which solve the problems of how to adapt to objective differences of different users in a resource allocation process, accurately predict the resource demands of the users and attach real-time changes, effectively distinguish different types of abnormal resource allocation, and dynamically adjust an adaptation mechanism to maintain a long-term accurate resource allocation effect. In order to achieve the above purpose, the present invention provides the following technical solutions: the resource allocation method based on the Internet of things digital calorimeter comprises the following steps: In the installation and debugging stage, configuring individual resource adaptation coefficients of users; in the resource allocation operation stage, acquiring resource allocation parameters and resource allocation guide parameters of a user through an Internet of things digital calorimeter, aligning data, associating individual resource adaptation coefficients, constructing a periodical allocation data set, and quantifying resource allocation deviation parameters; The method comprises the steps of periodically distributing data in a data set, screening a preset number of historical resource distribution period data from the data set, constructing a historical resource distribution set, constructing a reference resource demand curve by extracting resource consumption statistical characteristics of each historical period, matching a predicted time length based on fluctuation amplitude of resource distribution guiding parameters of a current resource distribution period, and carrying out layered progressive correction on the reference resource demand curve by combining actual resource consumption deviation and resource distribution deviation parameters of each historical period to generate a personalized resource demand prediction curve adapting to the predicted time length; Based on the personalized resource demand prediction curve and related individual resource adaptation coefficients, generating and executing a personalized resource allocation scheme, monitoring dynamic changes of resource allocation parameters and resource allocation guiding parameters in real time so as to carry out feedback correction on the personalized resource allocation scheme, and acquiring environment adaptation type terminal allocation deviation parameters of each resource allocation period, whe