CN-121977273-A - Intelligent diagnosis method and system for energy efficiency of central air conditioner

CN121977273ACN 121977273 ACN121977273 ACN 121977273ACN-121977273-A

Abstract

The invention provides a central air conditioner energy efficiency intelligent diagnosis method and system, which relate to the technical field of refrigeration equipment control, and the method comprises the steps of collecting monitoring data and equipment information of a central air conditioner water system, preprocessing and energy consumption calibration, constructing a data analysis model, analyzing and correcting the monitoring data, outputting relevant monitoring data, calculating a host COP value by adopting an improved energy efficiency algorithm based on host parameters and energy consumption parameters, calculating a system COP value by adopting a system comprehensive energy efficiency algorithm, analyzing the energy consumption parameters by adopting an energy consumption analysis model, generating an energy consumption proportion graph, carrying out anomaly identification on the relevant monitoring data, the host COP value, the system COP value and the energy consumption proportion graph by adopting a dynamic threshold algorithm, generating an alarm event, carrying out fault location on the alarm event by combining a collaborative fault tree model, and outputting a fault analysis result, thereby realizing accurate diagnosis, early warning of anomaly, quick fault location and continuous optimization of the central air conditioner energy efficiency.

Inventors

ZHAN YANPING
Dai Tianya
YUAN GUANNAN
CUI JIAN
XU PENGFEI

Assignees

南京东创节能技术有限公司

Dates

Publication Date: 20260505
Application Date: 20260117

Claims (10)

1. An intelligent diagnosis method for energy efficiency of a central air conditioner is characterized by comprising the following steps: Collecting multi-dimensional monitoring data and equipment file information of a central air conditioner water system, wherein the multi-dimensional monitoring data at least comprise a host operation parameter, a water pump operation parameter, a cooling tower operation parameter, a machine room energy consumption parameter and an environment parameter; Preprocessing and energy consumption calibration are carried out on the multi-dimensional monitoring data, a data association analysis model is constructed, analysis and correction are carried out on the multi-dimensional monitoring data, and water system association monitoring data are output; Based on the corrected host operation parameters and the machine room energy consumption parameters, calculating a host COP value by adopting an improved energy efficiency algorithm, calculating a system COP value by adopting a system comprehensive energy efficiency algorithm, and analyzing the machine room energy consumption parameters by an energy consumption proportion analysis model to generate a sub-term energy consumption proportion graph; and carrying out anomaly identification on the water system associated monitoring data, the host COP value, the system COP value and the subitem energy consumption duty ratio graph by adopting a dynamic threshold algorithm, generating an anomaly alarm event, carrying out fault location on the anomaly alarm event by combining a collaborative fault tree model, and outputting a fault analysis result.
2. The intelligent diagnosis method of central air conditioner energy efficiency according to claim 1, wherein the preprocessing and energy consumption calibration are performed on the multi-dimensional monitoring data, a data association analysis model is constructed, the multi-dimensional monitoring data is analyzed and corrected, and the water system association monitoring data is output, specifically comprising: abnormal value rejection is carried out on the multi-dimensional monitoring data based on a dynamic threshold method of working condition partition, noise suppression is carried out on the multi-dimensional monitoring data by adopting a self-adaptive filtering algorithm, and missing data complement is carried out on the multi-dimensional monitoring data by adopting a time sequence interpolation method; Performing precision calibration on the machine room energy consumption parameters according to a preset electric energy meter error coefficient, and performing normalization processing on the multi-dimensional monitoring data; Constructing the data association analysis model, carrying out correlation analysis and principal component extraction on the normalized multi-dimensional monitoring data by adopting a multivariate statistical analysis method, and identifying and quantifying the internal coupling relation between the multi-dimensional monitoring data; And extracting parameter association characteristics of the multi-dimensional monitoring data through a characteristic mapping algorithm based on the internal coupling relation, establishing a parameter association matrix, dynamically correcting the multi-dimensional monitoring data based on the parameter association matrix, and outputting the water system association monitoring data.
3. The intelligent diagnosis method for central air conditioner energy efficiency according to claim 2, wherein the dynamic correction of the multi-dimensional monitoring data based on the parameter association matrix comprises the following steps: Based on the parameter association matrix, calculating association confidence coefficient among all the association feature items, and reserving the association feature items with the association confidence coefficient larger than a preset confidence coefficient threshold value; Building an environment-refrigerating capacity association model, dynamically correcting the host refrigerating capacity Q in the host operation parameters, and generating corrected host refrigerating capacity The corresponding calculation formula is as follows: Wherein k represents a cooling water temperature influence coefficient in the parameter association matrix; the rated cooling water inlet temperature under the standard working condition is represented; Representing the real-time cooling water inlet temperature in the operation parameters of the host; representing equipment aging attenuation coefficients in the parameter association matrix; representing the accumulated running time of the equipment in the equipment file information; Representing a natural exponential function based on e; constructing an aging-host power correlation model for compressor operating power among the host operating parameters Dynamic correction is carried out to generate corrected compressor running power The corresponding calculation formula is as follows: In the formula, Representing the compressor aging power attenuation coefficient in the parameter association matrix; representing actual maintenance times in the equipment archive information; Representing the number of times of maintenance of a preset annual standard in equipment archive information; constructing a load-auxiliary machine energy consumption correlation model, and performing operation power on a chilled water pump in the machine room energy consumption parameters Dynamically correcting to generate corrected running power of the chilled water pump The corresponding calculation formula is as follows: In the formula, Representing a load influence coefficient in the parameter association matrix; Representing real-time cold water flow in the water pump operating parameters; The rated cold water flow of the water chilling unit in the equipment file information is represented; Representing a correction coefficient corresponding to the type-energy consumption response correlation characteristic of the cooling water pump in the parameter correlation matrix; and the calculation mode for dynamically correcting the running power of the cooling water pump and the running power of the cooling tower in the machine room energy consumption parameters is the same as the above.
4. The intelligent diagnosis method for central air conditioner energy efficiency according to claim 3, wherein based on the corrected host operation parameter and the machine room energy consumption parameter, a modified energy efficiency algorithm is adopted to calculate a host COP value System COP value is calculated by adopting system comprehensive energy efficiency algorithm Analyzing the machine room energy consumption parameters through an energy consumption duty ratio analysis model to generate an i-th equipment energy consumption duty ratio The corresponding calculation formula is as follows: In the formula, Representing the power of a host oil pump in the host operating parameters; Indicating the corrected total input electric power of the host, ; Representing the corrected running power of the cooling water pump; Representing the corrected operating power of the cooling tower; Representing the operating power of the i-th type equipment; , representing the host power duty cycle; , representing the energy consumption ratio of the chilled water pump; , representing the energy consumption ratio of the cooling water pump; , Representing the energy consumption ratio of the cooling tower; Integrating the host energy consumption ratio, the chilled water pump energy consumption ratio, the cooling water pump energy consumption ratio and the cooling tower energy consumption ratio, and generating the subitem energy consumption ratio map through a visual mapping algorithm.
5. The intelligent diagnosis method of central air conditioner energy efficiency according to claim 1, wherein the adopting of dynamic threshold algorithm carries out anomaly identification on the water system associated monitoring data, the host COP value, the system COP value and the itemized energy consumption duty ratio map, and generates an anomaly alarm event, and the method specifically comprises: Based on the water system association monitoring data, the host COP value, the system COP value and historical time sequence data of the equipment energy consumption ratio extracted from the sub-item energy consumption ratio chart, a sliding window statistical method is adopted to execute a dynamic threshold algorithm, and corresponding dynamic threshold upper limits are respectively calculated And a dynamic threshold lower limit ; Acquiring real-time outdoor temperature in the environment parameters and system real-time total flow in the water pump operation parameters, and comparing the real-time outdoor temperature and the system real-time total flow with a preset working condition classification threshold value to generate an operation working condition class L; Based on the operating condition level L, the dynamic threshold is limited And the dynamic threshold lower limit Performing working condition compensation to generate a working condition self-adaptive dynamic threshold, including an upper limit of the working condition self-adaptive dynamic threshold And a working condition adaptive dynamic threshold lower limit The corresponding calculation formula is as follows: In the formula, Representing a threshold relaxation function; And comparing the water system associated monitoring data, the host COP value, the system COP value and the equipment energy consumption duty ratio extracted from the sub-term energy consumption duty ratio chart with the corresponding working condition self-adaptive dynamic threshold value respectively, and generating the abnormal alarm event when any one of the water system associated monitoring data, the host COP value, the system COP value or the equipment energy consumption duty ratio exceeds the corresponding working condition self-adaptive dynamic threshold value range.
6. The intelligent diagnosis method for central air conditioner energy efficiency according to claim 1, wherein the combining of the collaborative fault tree model performs fault location for the abnormal alarm event and outputs a fault analysis result, specifically comprising: inputting the abnormal alarm event, the water system associated monitoring data and a parameter associated matrix into the collaborative fault tree model; The collaborative fault tree model carries out association analysis on the abnormal alarm event according to the quantized equipment coupling relation in the parameter association matrix, positions equipment nodes associated with abnormal parameters and generates a candidate fault set; invoking a preset fault-symptom mapping rule library, and sequentially matching a certainty rule based on a device physical mechanism and a probability rule based on historical time sequence data statistics for candidate faults in the candidate fault set to generate a fault rule matching degree and a fault statistics probability; Based on the fault rule matching degree and the fault statistical probability, calculating the comprehensive confidence degree of the candidate faults by adopting a weighted fusion algorithm, sequencing the candidate faults according to the comprehensive confidence degree, generating a comprehensive confidence degree evaluation result, combining the fault type and the abnormal parameter list of the candidate faults, and outputting the structured fault analysis result.
7. The intelligent diagnosis method of central air conditioner energy efficiency according to claim 6, wherein the collaborative fault tree model performs association analysis on the abnormal alarm event according to the quantized device coupling relation in the parameter association matrix, locates device nodes associated with abnormal parameters, and generates a candidate fault set, and specifically comprises: Extracting abnormal parameters of the abnormal alarm event to form an abnormal parameter set, inquiring equipment nodes directly related to the abnormal parameters in the abnormal parameter set based on the parameter association matrix, and generating an initial association equipment set; calculating the comprehensive association strength of the device node r and the abnormal parameter set in the initial association device set The corresponding calculation formula is as follows: Wherein N represents the total number of abnormal parameters in the abnormal parameter set; Representing the coupling weight between the equipment node r and the nth abnormal parameter in the parameter association matrix; representing the severity of the anomaly for the nth anomaly parameter; D represents the total number of coupling association parameters which have coupling relation with the equipment node and do not trigger alarm; representing the deviation amount of the real-time coupling weight between the equipment node r and the d coupling correlation parameter and the reference coupling weight in the parameter correlation matrix; selecting the comprehensive association strength And the equipment nodes larger than a preset association threshold value form the candidate fault set.
8. A central air conditioner energy efficiency intelligent diagnosis system applied to the central air conditioner energy efficiency intelligent diagnosis method as set forth in any one of claims 1-7, characterized in that the system comprises: The data acquisition module is used for acquiring multi-dimensional monitoring data and equipment file information of the central air conditioner water system, wherein the multi-dimensional monitoring data at least comprises a host operation parameter, a water pump operation parameter, a cooling tower operation parameter, a machine room energy consumption parameter and an environment parameter; The data processing module is used for preprocessing the multi-dimensional monitoring data and calibrating energy consumption, constructing a data association analysis model, analyzing and correcting the multi-dimensional monitoring data and outputting water system association monitoring data; The energy consumption analysis module is used for calculating a host COP value by adopting an improved energy efficiency algorithm based on the corrected host operation parameters and the machine room energy consumption parameters, calculating a system COP value by adopting a system comprehensive energy efficiency algorithm, and analyzing the machine room energy consumption parameters by using an energy consumption proportion analysis model to generate a sub-term energy consumption proportion graph; The fault early warning module is used for carrying out abnormal recognition on the water system associated monitoring data, the host COP value, the system COP value and the subitem energy consumption proportion graph by adopting a dynamic threshold algorithm, generating an abnormal alarm event, carrying out fault positioning on the abnormal alarm event by combining a collaborative fault tree model, and outputting a fault analysis result.
9. An electronic device comprising a memory and a processor, said memory having stored therein a computer program, characterized in that when said processor runs said computer program stored in said memory, said processor performs the steps of a central air conditioning energy efficiency intelligent diagnosis method according to any one of claims 1-7.
10. A readable storage medium having a computer program stored therein, wherein the computer program is executed by a processor for implementing the steps of a central air conditioning energy efficiency intelligent diagnosis method according to any one of claims 1 to 7.

Description

Intelligent diagnosis method and system for energy efficiency of central air conditioner Technical Field The invention relates to the technical field of refrigeration equipment control, in particular to a central air conditioner energy efficiency intelligent diagnosis method and system. Background The central air conditioning system is used as main energy consumption equipment of a large building, and the energy efficiency level of the central air conditioning system directly influences the overall operation cost and the energy saving effect. In the prior art, a data acquisition module and an energy efficiency analysis platform are deployed to evaluate the energy efficiency and alarm the abnormity of a central air conditioner host and auxiliary equipment. However, the existing central air conditioning energy efficiency diagnosis technology has significant technical defects. Firstly, the data acquisition dimension is single, and only the temperature and power parameters of a water chilling unit are covered, and the operation data of auxiliary machines such as a cooling tower and a water pump and the related information such as the environmental temperature and humidity and the system load rate are not fully acquired, so that the energy efficiency analysis lacks systematicness. Secondly, the acquisition frequency is low, generally once every minute, and transient processes such as equipment start-stop and load abrupt change are difficult to capture, and early failure characteristics are easy to miss. Moreover, the data processing capability is weaker, only simple outlier rejection is performed, and data completion, noise suppression and multi-parameter correlation analysis are lacked, so that the data reliability is affected. In addition, the energy efficiency calculation model adopts a fixed formula, is not dynamically corrected according to equipment aging, environmental change and operation conditions, so that the energy efficiency evaluation deviation is larger, abnormal alarm depends on a static threshold value, false alarm and missing alarm occur frequently, and a fault source cannot be positioned. Finally, the system functions are limited in post alarm and report generation, lack of closed-loop management capabilities such as fault early warning, energy efficiency optimization suggestions and visual analysis, and are difficult to support the actual requirements of continuous energy conservation and intelligent operation and maintenance. Therefore, it is necessary to provide a method and a system for intelligent diagnosis of central air conditioner energy efficiency to solve the above technical problems. Disclosure of Invention In order to solve the technical problems, the invention provides an intelligent diagnosis method and system for energy efficiency of a central air conditioner, which are used for solving the problems of narrow data acquisition dimension, weak data processing capability, low diagnosis precision and single system function in the prior art. The invention provides a central air conditioner energy efficiency intelligent diagnosis method, which comprises the following steps: Collecting multi-dimensional monitoring data and equipment file information of a central air conditioner water system, wherein the multi-dimensional monitoring data at least comprise a host operation parameter, a water pump operation parameter, a cooling tower operation parameter, a machine room energy consumption parameter and an environment parameter; Preprocessing and energy consumption calibration are carried out on the multi-dimensional monitoring data, a data association analysis model is constructed, analysis and correction are carried out on the multi-dimensional monitoring data, and water system association monitoring data are output; Based on the corrected host operation parameters and the machine room energy consumption parameters, calculating a host COP value by adopting an improved energy efficiency algorithm, calculating a system COP value by adopting a system comprehensive energy efficiency algorithm, and analyzing the machine room energy consumption parameters by an energy consumption proportion analysis model to generate a sub-term energy consumption proportion graph; and carrying out anomaly identification on the water system associated monitoring data, the host COP value, the system COP value and the subitem energy consumption duty ratio graph by adopting a dynamic threshold algorithm, generating an anomaly alarm event, carrying out fault location on the anomaly alarm event by combining a collaborative fault tree model, and outputting a fault analysis result. Preferably, the preprocessing and the energy consumption calibration are performed on the multi-dimensional monitoring data, a data association analysis model is constructed, the multi-dimensional monitoring data is analyzed and corrected, and the water system association monitoring data is output, which specifically comprises: abnormal valu