CN-121977725-A - Heat supply working condition multisource parameter real-time acquisition heat meter performance intelligent analysis system

CN121977725ACN 121977725 ACN121977725 ACN 121977725ACN-121977725-A

Abstract

The invention relates to the technical field of heat meter performance test and operation and maintenance management, in particular to an intelligent analysis system for heat meter performance collection in real time of heat supply working condition multi-source parameters. The multi-dimensional variable working condition is built through the working condition simulation subsystem, the high-precision synchronous acquisition is realized by the acquisition module, the intelligent analysis engine integrates various algorithms to complete data fusion, performance prediction, health assessment and fault early warning, and the full-flow closed-loop management is realized by the data management module. The invention improves the accuracy and the working condition coverage of the heat meter test, realizes intelligent analysis and accurate calibration, reduces the operation and maintenance cost and ensures the stable and efficient operation of the heating system.

Inventors

SU YOULIANG
YANG XIAOXIN
DING MEISHUANG
WANG WENJUN
ZHANG DEYU
CHEN BIN
FANG JUNJIE
LIU ZHEYUAN
WANG JUNQIAN

Assignees

青岛能源热电集团有限公司

Dates

Publication Date: 20260505
Application Date: 20260129

Claims (10)

1. Heat supply operating mode multisource parameter real-time acquisition heat table performance intelligent analysis system, its characterized in that includes: the working condition simulation subsystem (100) is used for simulating the real heating pipe network running environment with high precision and comprises heat source equipment (110), a circulating water pump (120), a temperature regulating device (130) and a water quality simulation and regulation unit (140), wherein the water quality simulation and regulation unit (140) can simulate different water quality conditions by injecting calcium ions, magnesium ions, suspended sediment and rust impurities with specific concentrations; The multi-source parameter real-time acquisition module (200) is connected with the working condition simulation subsystem (100) and the tested heat meter (10) and comprises an environment parameter sensor array (210) for measuring pipeline pressure, temperature and flow rate and a working parameter acquisition unit (220) for reading instantaneous flow, accumulated heat and supply and return water temperature difference of the tested heat meter (10), and the module is also integrated with a high-precision standard ultrasonic heat meter (230) serving as a measurement reference; An intelligent analysis engine (300) configured to receive a real-time data stream from the multi-source parameter real-time acquisition module (200) and perform the following core processes: Fusion processing is carried out on the multi-sensor data based on a Kalman filtering algorithm so as to improve the signal-to-noise ratio and the reliability of the measured data; Predicting the performance degradation trend of the heat meter by using a trained long-short-term memory network (LSTM) model; Adopting a fuzzy logic algorithm, integrating multidimensional indexes of water hardness, temperature fluctuation and flow speed stability, and calculating the real-time health degree score of the heat meter; real-time data flow is monitored based on an isolated forest algorithm, so that intelligent recognition and early warning of early faults of the heat meter are realized; The data management and visualization interaction interface (400) is used for storing history and real-time data, displaying a thermal meter performance curve, a health degree evaluation result, fault early warning information and generating a comprehensive performance analysis report in a graphical mode.
2. The intelligent analysis system for heat supply condition multi-source parameter real-time collection heat meter performance according to claim 1, wherein the performance degradation prediction model in the intelligent analysis engine (300) adopts an LSTM network with enhanced attention mechanism, and the Performance Degradation Index (PDI) calculation formula is: ; Wherein, the As an index of performance degradation at time t, For the weight coefficient of each parameter, In order for the parameters to be a rate of fluctuation, For adjusting the coefficients.
3. The intelligent analysis system for real-time acquisition of heat meter performance of heat supply working condition multisource parameters according to claim 2, wherein the performance degradation prediction model adopts forgetting gate of long-term memory network LSTM model Input door Cell status The calculation mode of (2) is as follows: ; ; ; ; Wherein, the , , As a matrix of weights, the weight matrix, , , As a result of the bias term, Is that The function is activated and the function is activated, Representing the hadamard product.
4. The heating condition multisource parameter real-time acquisition heat meter performance intelligent analysis system according to claim 1, wherein the real-time health score (HIS) of the heat meter is calculated by the following fuzzy logic formula: ; Wherein HIS is the health comprehensive score, Membership functions for the jth evaluation index, including metering error, battery voltage, data communication stability, Is the weight of the corresponding index.
5. The heating condition multisource parameter real-time collection heat meter performance intelligent analysis system according to claim 1, wherein the water quality influence factor of the water quality simulation and regulation unit (140) Calculated by the following quantization model: ; Wherein, the 、 Respectively the concentration of calcium and magnesium ions, For the turbidity of the water, Is an index of the corrosiveness of the water, To the point of The influence coefficients obtained by fitting experimental data are used.
6. The intelligent analysis system for real-time acquisition of heat meter performance by heating condition multisource parameters according to claim 1, wherein in the improved isolated forest algorithm adopted by the fault early warning module, the anomaly score s (x, n) of the sample point x is determined by the following formula: ; Where h (x) is the path length of the sample point in the isolated tree, c (n) is the average of the path lengths in the forest at the time of the given sample number n space time, for normalization processing, and E (h (x)) is the expected value of the path length.
7. The heating condition multisource parameter real-time acquisition heat meter performance intelligent analysis system according to claim 1, wherein the system further comprises a digital twin virtual calibration module that obtains optimal calibration parameters of the heat meter by solving the following optimization problem: ; wherein J (theta) is a loss function, theta is a heat meter calibration parameter set to be optimized, As a result of the actual observation value, R (theta) is a regularization term for a predicted value of the digital twin model under a parameter theta, and lambda is a regularization coefficient.
8. The intelligent analysis system for real-time collection of heat meter performance by heat supply condition multi-source parameters according to claim 1, wherein the multi-source parameter real-time collection module (200) adopts a data packing protocol based on time stamp synchronization, so as to ensure that environmental parameters and heat meter working parameters are strictly aligned in a time dimension.
9. The heating condition multisource parameter real-time acquisition heat meter performance intelligent analysis system according to claim 1, wherein the data management and visualization interaction interface (400) comprises the following functional modules: The data storage unit (410) adopts a distributed time sequence database to store full data, the data format accords with the requirements of JJF1033-2023 metering standard assessment Specification, the data storage unit comprises a time stamp, a sensor number, a working condition number, a parameter name, a parameter value, a data quality mark and an analysis result key field, the data storage and the traceability inquiry of history data of more than or equal to 10 calendar are supported, and the data reading and writing speed is more than or equal to 1000 pieces/second; the visual display unit (420) provides a multi-dimensional data visual function, including a real-time parameter monitoring curve, a heat meter performance trend graph, a health degree scoring radar graph, a working condition parameter thermodynamic diagram and a multi-brand heat meter performance comparison histogram, supports user-defined time interval data query and chart derivation, and comprises a format of PNG, PDF, excel; The report generating unit (430) supports automatic generation of three types of standardized reports, namely a thermal form item performance test report, a multi-working condition adaptability evaluation report and a heat exchange station instrument calibration analysis report, wherein the report comprises test conditions, a data form, a trend curve and a conclusion suggestion module, and can configure a report template according to user requirements to support batch export and printing; The early warning pushing unit (440) synchronously pushes fault early warning information through four modes of short message, APP pushing, mail and on-site audible and visual alarm, wherein the early warning information comprises fault hotlist numbers, installation positions, fault types, abnormal parameter values and suggested processing schemes, and supports early warning level setting and processing state tracking; And the authority management unit (450) adopts three-level authority hierarchical control, an administrator, an operator and a viewer, wherein the administrator has all operation authorities, the operator can perform test operation and data viewing, the viewer only has data viewing and report viewing authorities, all operation behaviors record logs, and the log retention time is more than or equal to 2 years.
10. The intelligent analysis system for real-time collection of heat meter performance by heat supply condition multisource parameters according to any one of claims 1 to 9, wherein the intelligent analysis for real-time collection of heat meter performance by heat supply condition multisource parameters comprises the following steps: Step S1, working condition simulation and parameter acquisition preparation, namely configuring a target heating working condition through a working condition simulation subsystem (100), wherein the target heating working condition comprises setting water quality parameters, a temperature range, a flow rate range and a pressure range, connecting a tested heat meter (10) with a multi-source parameter real-time acquisition module (200), and starting a standard ultrasonic heat meter (230) for reference calibration; S2, synchronously collecting multisource parameters, namely starting an acquisition module, collecting pipeline pressure, temperature and flow rate data by an environment parameter sensor array (210) according to the frequency of 10Hz, reading parameters such as instantaneous flow, accumulated heat, temperature difference of water supply and return, battery voltage and the like of a heat meter by a working parameter acquisition unit (220) according to the frequency of 5Hz, realizing data alignment by a timestamp synchronous protocol, and generating a multi-dimensional parameter data set; and S3, data fusion and quality assessment, wherein a Kalman filtering algorithm is adopted to carry out fusion processing on the acquired multi-sensor data, abnormal values and noise are removed, and a calculation formula is as follows: , ; Wherein the method comprises the steps of As the state estimation value after the fusion, In order for the kalman gain to be achieved, In order for the sensor to observe a value, In order to observe the matrix, For the covariance matrix at the previous moment, Performing quality evaluation on the fused data, and entering the next step when the data qualification rate is more than or equal to 99.5%, otherwise, re-acquiring; Step S4, heat meter performance analysis and evaluation: S4.1, predicting performance degradation trend, namely inputting the fused data into an LSTM model with enhanced attention mechanism, calculating a performance degradation index based on a PDI formula, and predicting the performance change trend of a heat meter 3-12 months in the future; S4.2, calculating the real-time health degree score of the heat meter through a fuzzy logic formula, and analyzing the working condition adaptability by combining the water quality influence factors; S4.3, fault early warning detection, namely adopting an improved isolated forest algorithm to perform abnormal detection on the real-time data stream, and outputting fault probability and early warning level; s4.4, optimizing calibration parameters, namely solving the optimal calibration parameters through a digital twin virtual calibration module to generate a hotlist calibration suggestion; And S5, outputting and feeding back a result, namely storing analysis data by a data management and visualization interaction interface (400), displaying a performance curve, a health evaluation result and fault early warning information, automatically generating a comprehensive performance analysis report, pushing a maintenance/replacement suggestion and a working condition optimization scheme of the unqualified hotlist to an equipment management terminal, and forming a closed loop analysis flow of acquisition, analysis, feedback and optimization.

Description

Heat supply working condition multisource parameter real-time acquisition heat meter performance intelligent analysis system Technical Field The invention relates to the technical field of heat meter performance test and operation and maintenance management, in particular to an intelligent heat meter performance analysis system for acquiring heat meter performance in real time under a heating working condition. Background In a central heating system, a heat meter is used as core equipment for heat metering, the metering precision and the running stability of the heat meter are directly related to the economic benefits of heating enterprises and users, and the energy utilization efficiency of the heating system is also affected. Along with the rapid development of the heat supply industry, heat supply working conditions are more complex, water supply temperatures, flow rates, pressures and water quality parameters in different areas and different seasons are obviously different, and higher requirements are provided for multi-working condition adaptability and long-term operation reliability of the heat meter. However, the conventional heat meter performance testing and analyzing technology still has a plurality of limitations, and is difficult to meet the actual application requirements. The conventional performance test of the heat meter is mostly dependent on a single fixed working condition, a working condition simulation system cannot accurately reproduce conventional, severe and extreme working conditions in an actual heating scene, particularly, the dynamic coupling change of water quality parameters (such as calcium ion, magnesium ion concentration and turbidity) and temperature and flow rate is difficult to simulate, so that the performance of the heat meter in an actual running environment cannot be truly reflected by a test result, and the problem of disjointing between the test and the actual application exists. Meanwhile, the multisource parameter acquisition link lacks an effective synchronization mechanism, the acquisition of environmental parameters and heat meter working parameters has time deviation, the sensor precision is insufficient, the data noise filtering capability is weak, the accuracy and reliability of acquired data are difficult to guarantee, and hidden danger is buried for subsequent analysis work. In the aspect of heat meter performance analysis and calibration, the prior art relies on manpower to finish data arrangement and analysis, has low efficiency and is easy to generate human errors, and the accurate prediction of the heat meter performance degradation trend can not be realized. The traditional analysis method focuses on the evaluation of a single performance index, lacks comprehensive quantitative analysis on the health state of the heat meter, and has the defects that the fault early warning is based on threshold judgment, the response is delayed, and the running risk is difficult to avoid in advance. In addition, the heat meter is calibrated in an off-line disassembly and calibration mode, so that the operation is complex, the cost is high, dynamic calibration cannot be performed by combining real-time operation data, the calibration accuracy and timeliness are poor, and the heat meter on-line operation and maintenance requirements are difficult to adapt. In the aspect of data management and interaction, the existing system mostly adopts a distributed data storage mode, so that the data integration difficulty is high, and effective association analysis of historical data and real-time data cannot be realized. The visual display function is single, the performance change rule and fault information of the heat meter are difficult to visually present, manual intervention is needed for report generation, the early warning information pushing channel is single, the response is slow, closed-loop management from test, analysis to operation and maintenance cannot be formed, the operation and maintenance management efficiency of the heat meter is low, and the operation cost of a heating system is increased. To sum up, aiming at the problems of incomplete working condition simulation, asynchronous data acquisition, low intelligent degree of analysis and calibration, inconvenient interaction of data management and the like in the existing heat meter performance test technology, there is a great need to develop a heat meter performance analysis system capable of covering multiple working conditions, realizing high-precision synchronous acquisition, intelligent analysis and calibration and full-flow closed-loop management so as to improve the refinement level of heat meter test and operation and maintenance management and ensure the stable and efficient operation of a heat supply system. Disclosure of Invention The invention aims to provide an intelligent analysis system for heat supply working condition multisource parameter real-time acquisition heat meter performance, which aim