CN116992371A - Diagnostic model data labeling method and device, electronic equipment and storage medium - Google Patents

Abstract

The application discloses a data labeling method, a device, electronic equipment and a storage medium of a diagnostic model, wherein the method comprises the steps of acquiring unlabeled system operation data; marking the unmarked system operation data according to the established physical diagnosis model to generate system operation data with a data tag; training a machine learning model according to the system operation data with the data characteristic label to generate a system intelligent diagnosis model; diagnosing the system according to the intelligent diagnosis model of the system to generate an intelligent diagnosis result; if the intelligent diagnosis result is different from the manual diagnosis result, the labeled system operation data is subjected to label correction according to the manual diagnosis result, so that the technical problems that the time and the labor are consumed for manually labeling the data in the related technology, and the data labeling is not accurate enough through a physical diagnosis model can be solved.

Description

Diagnostic model data labeling method and device, electronic equipment and storage medium

Technical Field

The application relates to the technical field of artificial intelligent system diagnosis, in particular to a data labeling method and device for a diagnosis model, electronic equipment and a storage medium.

Background

In recent years, artificial intelligence technology is increasingly applied to intelligent diagnosis of the operational effects of devices. In the related art, when a machine learning is utilized to train a diagnosis model of a system (such as a central air conditioning system), most of the diagnosis models are based on various existing physical diagnosis models to mark system operation data, but the diagnosis model is often inaccurate, cannot completely cover manually observable data characteristics, and is deficient in the aspect of actual diagnosis effect; on the other hand, the reliability of the system diagnosis model obtained through training can be guaranteed by manually marking the equipment operation data, but the manual marking is time-consuming and labor-consuming, and the requirement of system diagnosis automation cannot be met.

Aiming at the technical problems that in the related technology, when a system diagnosis model is built, manual data marking is time-consuming and labor-consuming, and data marking is not accurate enough through a physical diagnosis model, no effective solution is proposed at present.

Disclosure of Invention

The application aims to overcome the technical defects and provide a data labeling method, device, electronic equipment and storage medium of a diagnosis model, so as to solve the technical problems that the manual data labeling is time-consuming and labor-consuming in the related technology, and the data labeling is not accurate enough through a physical diagnosis model.

In order to achieve the technical purpose, the application adopts the following technical scheme:

according to one aspect of the present application, there is provided a data labeling method of a diagnostic model, including:

acquiring unmarked system operation data;

marking the unmarked system operation data according to the established physical diagnosis model to generate system operation data with a data tag;

training a machine learning model according to the system operation data with the data characteristic label to generate a system intelligent diagnosis model;

diagnosing the system according to the intelligent diagnosis model of the system to generate an intelligent diagnosis result;

if the intelligent diagnosis result is different from the manual diagnosis result, carrying out label correction on the marked system operation data according to the manual diagnosis result.

Optionally, the method further comprises:

and comparing the difference between the intelligent diagnosis result and the artificial diagnosis result, and correcting the model parameters of the physical diagnosis model.

Optionally, the method further comprises:

and receiving the field feedback information of the system operation condition so as to obtain the manual diagnosis result according to the field feedback information.

Optionally, the performing label correction on the marked system operation data according to the manual diagnosis result includes:

receiving a manually marked system operation data label, wherein the manually marked system operation data label corresponds to the manual diagnosis result;

if the system operation data label marked by the person is different from the marked system operation data label, the system operation data label marked by the person is confirmed to be a final data label.

Optionally, the method further comprises: and if the system operation data label without the manual label is detected, the labeled system operation data label is confirmed to be a final data label.

Optionally, the diagnostic model includes a diagnostic model of an air conditioning system.

Optionally, the diagnostic model of the air conditioning system includes one or more of a sensor diagnostic model, a water pump diagnostic model, a chiller diagnostic model, a control strategy diagnostic model, and a cooling tower diagnostic model.

According to another aspect of the present application, there is also provided a data labeling apparatus for a diagnostic model, including an obtaining unit configured to obtain unlabeled system operation data;

the physical diagnosis model unit is used for marking the unmarked system operation data according to the established physical diagnosis model and generating the system operation data with the data tag;

the intelligent diagnosis model unit is used for performing machine learning model training according to the system operation data with the data characteristic labels to generate a system intelligent diagnosis model;

the diagnosis result generating unit is used for diagnosing the system according to the intelligent diagnosis model of the system and generating an intelligent diagnosis result;

and the data tag correction unit is used for performing tag correction on the marked system operation data according to the manual diagnosis result if the intelligent diagnosis result is different from the manual diagnosis result.

According to another aspect of the present application, there is also provided an electronic apparatus including: a processor and a memory;

the memory has stored thereon a computer readable program executable by the processor;

the steps of the method described above are implemented by the processor when executing the computer readable program.

According to another aspect of the present application, there is also provided a computer-readable storage medium storing one or more programs executable by one or more processors to implement the steps in the above-described method.

According to the data labeling method, the device, the electronic equipment and the storage medium of the diagnostic model, on one hand, the automatic data labeling based on the physical diagnostic model improves the labeling efficiency of data compared with a purely manual labeling mode, meanwhile, the manual data labeling is assisted to correct the automatic labeling result of the diagnostic model, so that the quality of a data label can be optimized, the diagnostic model trained based on the data label can be optimized, the accuracy of the diagnostic model is improved, and the technical problems that the manual labeling of the data is time-consuming and labor-consuming, and the data labeling is not accurate enough through the physical diagnostic model in the related art can be solved.

Drawings

FIG. 1 is a schematic flow chart of a method for labeling data of a diagnostic model according to an embodiment of the present application;

FIG. 2 is a user interface diagram of an intelligent diagnostic system for air conditioner according to an embodiment of the present application;

FIG. 3 is a schematic diagram of a physical diagnostic model in a method for labeling data of a diagnostic model according to an embodiment of the present application;

FIG. 4 is a flowchart of a method for labeling data of a diagnostic model according to another embodiment of the present application;

FIG. 5 is a flow chart of intelligent diagnosis in the data labeling method of the diagnosis model in FIG. 4;

FIG. 6 is a schematic structural diagram of a diagnostic model data labeling apparatus according to an embodiment of the present application;

fig. 7 is a block diagram of a terminal that can implement a data labeling method of a diagnostic model according to an embodiment of the present application.

Detailed Description

In order that those skilled in the art will better understand the present application, a technical solution in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present application without making any inventive effort, shall fall within the scope of the present application.

In the related art, when a system diagnosis model is built, manual data marking consumes time and labor, and data marking is not accurate enough through a physical diagnosis model. At present, a scheme capable of better solving the technical problems is not available.

Based on the above problems, the present application provides a data labeling method for a diagnostic model, so as to solve the technical problems in the related art. The following is a detailed description.

Example 1

According to an embodiment of the present application, there is provided a data labeling method of a diagnostic model, with reference to fig. 1, the method including:

step S101, obtaining unlabeled system operation data;

in step S101, as an example, the server may obtain, from a data transmission unit (such as a DTU unit, etc.) in the air conditioning system, a history running source dataset of the air conditioning system without labels.

And step S103, marking the unmarked system operation data according to the established physical diagnosis model, and generating the system operation data with the data tag.

In step S103, as an example, a physical diagnosis model of the system may be suggested and stored in advance in the server. Taking an air conditioning system as an example, as shown in fig. 3, the physical diagnosis model may include a sensor diagnosis model, a water pump diagnosis model, a chiller diagnosis model, a control strategy diagnosis model, a cooling tower diagnosis model, and the like. Taking cold station energy efficiency diagnosis as an example, the physical model of the cold station energy efficiency diagnosis is based on the EER (refrigerating capacity/total electric power of an air conditioning system), if EER 4.6 is excellent, EER 3.8< 4.6 is good, EER 3.2< 3.8 is general, EER 3.2 is poor, and the energy efficiency level is marked according to the above basis by intelligent marking without manual intervention. In some embodiments, multiple physical diagnosis models can be simultaneously built and stored in the server, so that after the air conditioner operation data are collected for cleaning and classification, the collected air conditioner operation data can be input into multiple types of physical diagnosis models to be applied to diagnosis of multiple air conditioner operation states, and the utilization rate of the air conditioner operation data is improved.

Step S105, training a machine learning model according to the system operation data with the data characteristic labels, and generating a system intelligent diagnosis model.

In step S105, as an example, the data feature recognition may be performed on the system history operation data set based on the established system physical diagnosis model, for example, the recognition result may include a type of system failure, a low energy efficiency and a time of occurrence thereof, and then the air conditioning system history operation data set is labeled according to the recognition result, so as to generate the air conditioning system operation data set with the data feature tag. The data is initially and intelligently marked by means of the physical diagnosis model, and compared with a purely manual data marking mode, the data marking efficiency is improved. And then, inputting the labeled data set into a machine learning model, and training the periodic machine learning model to initially obtain the intelligent diagnosis model of the system. For example, in some embodiments, historical operating data of the air conditioner is collected for labeling and training, and then an intelligent diagnostic model of the air conditioning system can be generated for diagnosing the type of failure, low energy efficiency, and time of occurrence of the air conditioner.

And step S107, diagnosing the system according to the intelligent diagnosis model of the system, and generating an intelligent diagnosis result.

As described above, in step S105, the system intelligent diagnosis model is obtained through preliminary training, and in step S107, the operation condition of the system may be diagnosed by using the system intelligent diagnosis model, for example, when the model trained in step S105 is the intelligent diagnosis model of the air conditioning system, in step S107, the server may collect real-time operation data of the air conditioner through the data transmission unit of the air conditioning system, and input the real-time operation data into the intelligent diagnosis model of the air conditioning system, so as to be used for diagnosing the type of failure, low energy efficiency and occurrence time of the air conditioner. Still taking cold station energy efficiency diagnosis as an example, after the server collects the cold station accumulated refrigeration capacity, cold station accumulated power consumption, cold machine accumulated power consumption, and calculated cold machine power consumption duty ratio, cold station EER and other operation data, the operation data can be input into a system intelligent diagnosis model, diagnosis results such as excellent, good, general, poor and other results are output, and corresponding data labeling is performed. Compared with a physical diagnosis model, the diagnosis result of the intelligent diagnosis model is obtained through machine learning training, is more intelligent, and is more accurate in labeling result.

And step S109, if the intelligent diagnosis result is different from the manual diagnosis result, carrying out label correction on the marked system operation data according to the manual diagnosis result.

In step S109, when the diagnosis result identified by the intelligent diagnosis model is inconsistent with the manual diagnosis result, the labeled data is subjected to label correction according to the manual diagnosis result. In general, the manually labeled label results should be in control. Because expert labeling is adopted in manual labeling of supervised machine learning, the results are generally more accurate.

Compared with the data labeling based on the diagnostic model alone, the data label subjected to manual correction optimization has better accuracy, and the embodiment also auxiliarily introduces a mode of manual data labeling to correct the data label with poor quality of the diagnostic result, so that the intelligent diagnostic model obtained based on the optimized data label training has higher accuracy, and the quality of the data labeling is also considered while the improvement of the data identification efficiency is ensured.

It should be understood that, whether it is a physical diagnostic model or an intelligent diagnostic model obtained after machine learning training, some more complex factors are often not considered in the identification and diagnosis process, so that the identification result is not accurate enough. Taking the cold station energy efficiency diagnosis as an example, the energy efficiency evaluation of the refrigeration machine room by the traditional physical model is generally only based on the size of EER, for example, EER 4.6 is excellent, EER 4.6 is good, EER 3.2 is general, EER 3.8 is poor, and intelligent labeling without manual intervention is carried out according to the energy efficiency grade according to the above basis. However, in practice, when evaluating the energy efficiency of the refrigeration machine room, only focusing on one parameter of the EER is not comprehensive enough, and when performing manual diagnosis, it is generally necessary to comprehensively evaluate the energy consumption ratio of the water chiller in the refrigeration machine room.

As an example, on-site operation and maintenance personnel verify the validity of the diagnosis result according to the intelligent diagnosis result generated by the intelligent diagnosis system and feed back information to the server through the user terminal, and the server can receive on-site feedback information of the system operation condition, so that expert personnel logged in to the server can make more accurate manual diagnosis results by combining on-site feedback information and acquired system operation data, meanwhile, the expert personnel do not need to wait on duty in the system operation site in person, and the efficiency of data correction is improved.

Taking the cold station energy efficiency diagnosis as an example, if the cold station energy efficiency condition is judged to be "general" after diagnosis by the intelligent diagnosis model, but after the on-site personnel feedback information of the air conditioning system, an expert combines the on-site personnel feedback information and the operation data acquired by the server, and based on the professional experience, the cold station energy efficiency condition is manually diagnosed to be "good", and the data label of the corresponding cold station operation data needs to be corrected by taking the "good" as standard.

In step S109, as an example, when performing tag correction, a manually-noted system operation data tag may be received first, where the manually-noted system operation data tag corresponds to the manual diagnosis result. For example, if the manual diagnosis result is that the energy efficiency condition of the cold station is good, the corresponding data tag is "good".

And then, comparing the manually marked tag with the marked data tag in the system, and if the manually marked system operation data tag is judged to be different from the marked system operation data tag, determining the manually marked system operation data tag as a final data tag.

In other words, in the intelligent diagnosis system established by the server, the intelligent labeling result, the manual labeling result and the final labeling result can be stored at the same time, so that the labeling result is classified and stored, the system is facilitated to evaluate the perfection degree of the intelligent diagnosis model according to the correction condition of the data labeling, and the optimization process of the intelligent diagnosis model is better controlled. And the three labeling results can be further simultaneously output to a user using the intelligent diagnosis system. For example, in one intelligent diagnostic system shown in FIG. 2, after each piece of collected cold station operational data, intelligent labeling, manual labeling, and final labeling results are listed. The manual labeling result is an editable control, and is used for receiving correction operation of the manual label based on the manual diagnosis result. For example, the control may be configured as a menu item containing preset data labels, such as cold station operating energy efficiency, which may include excellent, good, general, and poor. When the user manually selects the options, the manually selected options are used as manual labeling results and are used for comparing with intelligent labeling results, so that the final labeling results are judged, and the final labeling results are output to the user. The judgment logic is as follows: if the manual labeling exists, the final diagnosis result is equal to the manual labeling result.

Further, if the system operation data label without manual marking is detected, the marked system operation data label is confirmed to be a final data label. For example, in the system shown in fig. 2, a "none" option may be configured in a menu item in a column of the manual labeling result, and before the manual labeling input, the "none" option may be set as a default result of the manual labeling, which corresponds to a system running data tag that the system does not detect the manual labeling, and at this time, the final labeling result is confirmed as a labeled intelligent labeling result according to the judgment logic. The method has the advantages that when the expert considers that the intelligent labeling result is accurate, operation feedback in the system is not needed, the intelligent labeling result can be directly defaulted, and the data labeling efficiency is further improved.

Optionally, the method further includes S111, comparing the difference between the intelligent diagnosis result and the manual diagnosis result, and correcting the model parameters of the physical diagnosis model.

In the embodiment, the diagnosis effect feedback based on the intelligent diagnosis model of the system is compared with the difference between the artificial diagnosis and the physical diagnosis model according to the comparison result, so that the artificial auxiliary data label is not only optimized for the data label, but also fed back to the physical diagnosis model to correct unreasonable or imperfect model parameters in the physical diagnosis model, thereby being beneficial to building a comprehensive and accurate physical diagnosis model for a long time. In addition, the physical diagnosis model is used for automatically labeling the data and is further used for training the intelligent diagnosis model, so that the accuracy of the physical diagnosis model is improved, and the accuracy of the intelligent diagnosis model of the system is further optimized. In some embodiments, the above steps of modifying and optimizing the model and the data tag may be repeatedly performed until a feedback with relatively accurate diagnosis results is obtained, thereby obtaining an accurate and complete intelligent diagnosis model of the system. For example, taking the cold station energy efficiency diagnosis as described above as an example, the conventional physical model generally evaluates the energy efficiency of the refrigeration machine room only based on the size of the EER, and after the manual diagnosis is performed, if the physical diagnosis model based on the size of the EER is found to be inaccurate, parameters that can affect the energy efficiency evaluation, such as the power consumption ratio of the water chilling unit in the refrigeration machine room, can be manually introduced into the physical diagnosis model.

Optionally, in one intelligent diagnosis system, multiple intelligent diagnosis models can be configured at the same time, so as to facilitate the establishment of comprehensive and rich diagnosis on the system. For example, in the intelligent air conditioner diagnostic system shown in fig. 2, a cold station energy efficiency diagnosis, a cold water pump energy efficiency diagnosis, a cooling tower energy efficiency diagnosis module may be simultaneously configured. And after receiving the selection operation of the user on the diagnosis type, switching to a corresponding diagnosis interface, and outputting data required by manual annotation according to the diagnosis type. Taking cold station energy efficiency diagnosis as an example, the interface can output running data such as cold station accumulated refrigerating capacity, cold station accumulated power consumption, cold machine accumulated power consumption, calculated cold machine power consumption ratio, cold station EER and the like to a user in a form mode, and display corresponding intelligent labeling, manual labeling and final standard results. The control of the manual annotation is editable and is used for manually correcting the intelligent annotation result, the final annotation result is generated by the system according to the manual annotation and the intelligent annotation result, and specific generating logic is described in detail above and is not repeated here. The use flow of the system can be summarized as follows: 1. a diagnostic category is selected. 2. Different data required for manual diagnosis are displayed according to the diagnosis category. 3. And correcting the intelligent labeling result manually. 4. And the system generates a final labeling result according to the intelligent labeling and the manual labeling.

Fig. 4 shows a data labeling method of an intelligent diagnosis model of an air conditioning system according to an embodiment, firstly, a data transmission unit inputs source data into a physical diagnosis model, labels the data through the physical diagnosis model to form a labeled data set, then machine learning model training is performed by using the labeled data set to obtain the intelligent diagnosis model of the air conditioning system, the intelligent diagnosis model of the air conditioning system is configured into the intelligent diagnosis module of the air conditioning system to perform diagnosis effect test, diagnosis results are fed back to an expert, the expert performs manual diagnosis based on experience, and if the manual diagnosis results are inconsistent with the intelligent diagnosis results, the labeled data set is manually labeled and corrected according to the manual diagnosis results so as to optimize label quality. Meanwhile, the physical diagnosis model can be subjected to model correction according to expert experience. In this embodiment, the diagnostic method of the intelligent diagnostic module of the air conditioning system may be as shown in fig. 5, firstly, the data transmission unit still collects the source data, then the source data is transmitted to the server, the intelligent diagnostic model of the air conditioning system performs automatic diagnosis, and after the diagnostic result is output, the operation and maintenance personnel of the air conditioning system feedback the diagnostic effect.

Compared with purely manual data marking, the embodiment greatly improves the data marking speed while ensuring the data marking precision, namely, the diagnosis model is adopted to automatically mark data and simultaneously assist manual data marking, the automatic marking result of the diagnosis model is corrected, the quality of the data label can be optimized, the diagnosis model trained based on the data label is optimized, the accuracy of the diagnosis model is improved, and therefore the technical problems that time and effort are consumed for manually marking the data in the related technology, and the data marking is not accurate enough through the physical diagnosis model can be solved.

From the description of the above embodiments, it will be clear to a person skilled in the art that the method according to the above embodiments may be implemented by means of software plus the necessary general hardware platform, but of course also by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. ROM/RAM, magnetic disk, optical disk) comprising instructions for causing a terminal device (which may be a mobile phone, a computer, a server, or a network device, etc.) to perform the method according to the embodiments of the present application.

Example two

According to an embodiment of the present application, there is provided a data labeling apparatus for a diagnostic model, in conjunction with fig. 6, including an obtaining unit 21 for obtaining unlabeled system operation data;

a physical diagnosis model unit 23, configured to label the unlabeled system operation data according to an established physical diagnosis model, and generate system operation data with a data label;

an intelligent diagnosis model unit 25, configured to perform machine learning model training according to the system operation data with the data feature tag, and generate a system intelligent diagnosis model;

a diagnostic result generating unit 27, configured to diagnose the system according to the system intelligent diagnostic model, and generate an intelligent diagnostic result;

and the data tag correction unit 29 is configured to perform tag correction on the labeled system operation data according to the manual diagnosis result if the intelligent diagnosis result is different from the manual diagnosis result.

Compared with purely manual data marking, the embodiment greatly improves the data marking speed while ensuring the data marking precision, namely, the diagnosis model is adopted to automatically mark data and simultaneously assist manual data marking, the automatic marking result of the diagnosis model is corrected, the quality of the data label can be optimized, the diagnosis model trained based on the data label is optimized, the accuracy of the diagnosis model is improved, and therefore the technical problems that time and effort are consumed for manually marking the data in the related technology, and the data marking is not accurate enough through the physical diagnosis model can be solved.

Alternatively, specific examples in this embodiment may refer to examples described in the foregoing embodiments, and this embodiment is not described herein.

It should be noted that the above modules are the same as examples and application scenarios implemented by the corresponding steps, but are not limited to what is disclosed in the above embodiments. It should be noted that, the above modules may be implemented in a corresponding hardware environment as part of the apparatus, and may be implemented in software, or may be implemented in hardware, where the hardware environment includes a network environment.

Fig. 7 is a block diagram of a terminal according to an embodiment of the present application, and as shown in fig. 7, the terminal may include: one or more (only one is shown) processors 101, memory 103, and transmission means 105, as shown in fig. 7, the terminal may further comprise input output devices 107.

The memory 103 may be used to store software programs and modules, such as program instructions/modules corresponding to the methods and apparatuses in the embodiments of the present application, and the processor 101 executes the software programs and modules stored in the memory 103, thereby performing various functional applications and data processing, that is, implementing the methods described above. Memory 103 may include high-speed random access memory, but may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, memory 103 may further include memory remotely located with respect to processor 101, which may be connected to the terminal via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.

The transmission device 105 is used for receiving or transmitting data via a network, and can also be used for data transmission between the processor and the memory. Specific examples of the network described above may include wired networks and wireless networks. In one example, the transmission device 105 includes a network adapter (Network Interface Controller, NIC) that may be connected to other network devices and routers via a network cable to communicate with the internet or a local area network. In one example, the transmission device 105 is a Radio Frequency (RF) module for communicating with the internet wirelessly.

Wherein in particular the memory 103 is used for storing application programs.

The processor 101 may call an application stored in the memory 103 via the transmission means 105 to perform the following steps:

acquiring unmarked system operation data; marking the unmarked system operation data according to the established physical diagnosis model to generate system operation data with a data tag; training a machine learning model according to the system operation data with the data characteristic label to generate a system intelligent diagnosis model; diagnosing the system according to the intelligent diagnosis model of the system to generate an intelligent diagnosis result; if the intelligent diagnosis result is different from the manual diagnosis result, carrying out label correction on the marked system operation data according to the manual diagnosis result.

Alternatively, specific examples in this embodiment may refer to examples described in the foregoing embodiments, and this embodiment is not described herein.

It will be appreciated by those skilled in the art that the above-mentioned structure of the terminal is merely illustrative, and the terminal may be a smart phone (such as an Android mobile phone, an iOS mobile phone, etc.), a tablet computer, a palm computer, a mobile internet device (Mobile Internet Devices, MID), a PAD, etc. Fig. 7 is not limited to the structure of the electronic device. For example, the terminal may also include more or fewer components (e.g., network interfaces, display devices, etc.) than shown in fig. 7, or have a different configuration than shown in fig. 7.

Those of ordinary skill in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be implemented by a program for instructing a terminal device to execute in association with hardware, the program may be stored in a computer readable storage medium, and the storage medium may include: flash disk, read Only Memory (ROM), random access Memory (Random Access Memory, RAM), magnetic or optical disk, and the like.

The embodiment of the application also provides a storage medium. Alternatively, in the present embodiment, the storage medium may be used for executing the program code of the data standard method of the diagnostic model.

Alternatively, in this embodiment, the storage medium may be located on at least one network device of the plurality of network devices in the network shown in the above embodiment.

Alternatively, in the present embodiment, the storage medium is configured to store program code for performing the steps of: acquiring unmarked system operation data; marking the unmarked system operation data according to the established physical diagnosis model to generate system operation data with a data tag; training a machine learning model according to the system operation data with the data characteristic label to generate a system intelligent diagnosis model; diagnosing the system according to the intelligent diagnosis model of the system to generate an intelligent diagnosis result; if the intelligent diagnosis result is different from the manual diagnosis result, carrying out label correction on the marked system operation data according to the manual diagnosis result.

Alternatively, specific examples in this embodiment may refer to examples described in the foregoing embodiments, and this embodiment is not described herein.

Alternatively, in the present embodiment, the storage medium may include, but is not limited to: a usb disk, a Read Only Memory (ROM), a random access Memory (RAM, random Access Memory), a removable hard disk, a magnetic disk, or an optical disk, or other various media capable of storing program codes.

The foregoing embodiment numbers of the present application are merely for the purpose of description, and do not represent the advantages or disadvantages of the embodiments.

The integrated units in the above embodiments may be stored in the above-described computer-readable storage medium if implemented in the form of software functional units and sold or used as separate products. Based on such understanding, the technical solution of the present application may be embodied in essence or a part contributing to the prior art or all or part of the technical solution in the form of a software product stored in a storage medium, comprising several instructions for causing one or more computer devices (which may be personal computers, servers or network devices, etc.) to perform all or part of the steps of the method described in the embodiments of the present application.

In the foregoing embodiments of the present application, the descriptions of the embodiments are emphasized, and for a portion of this disclosure that is not described in detail in this embodiment, reference is made to the related descriptions of other embodiments.

In the several embodiments provided by the present application, the described embodiments of the apparatus are merely illustrative, for example, the division of the units is merely a logical function division, and there may be other manners of division in practice, for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some interfaces, units or modules, or may be in electrical or other forms.

The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in the embodiments of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.

The foregoing is merely a preferred embodiment of the present application and it should be noted that modifications and adaptations to those skilled in the art may be made without departing from the principles of the present application, which are intended to be comprehended within the scope of the present application.

Claims (10)

1. A method for labeling data of a diagnostic model, comprising:

acquiring unmarked system operation data;

marking the unmarked system operation data according to the established physical diagnosis model to generate system operation data with a data tag;

training a machine learning model according to the system operation data with the data characteristic label to generate a system intelligent diagnosis model;

diagnosing the system according to the intelligent diagnosis model of the system to generate an intelligent diagnosis result;

if the intelligent diagnosis result is different from the manual diagnosis result, carrying out label correction on the marked system operation data according to the manual diagnosis result.

2. The method for labeling data of a diagnostic model of claim 1, further comprising:

and comparing the difference between the intelligent diagnosis result and the artificial diagnosis result, and correcting the model parameters of the physical diagnosis model.

3. The method for labeling data of a diagnostic model of claim 1, further comprising:

and receiving the field feedback information of the system operation condition so as to obtain the manual diagnosis result according to the field feedback information.

4. The method for labeling data of a diagnostic model according to claim 1, wherein the performing label correction on labeled system operation data according to the manual diagnostic result comprises:

receiving a manually marked system operation data label, wherein the manually marked system operation data label corresponds to the manual diagnosis result;

if the system operation data label marked by the person is different from the marked system operation data label, the system operation data label marked by the person is confirmed to be a final data label.

5. The method for labeling data of a diagnostic model of claim 4, further comprising: and if the system operation data label without the manual label is detected, the labeled system operation data label is confirmed to be a final data label.

6. The method for labeling data of a diagnostic model of claim 1, wherein the diagnostic model comprises a diagnostic model of an air conditioning system.

7. The method of claim 6, wherein the diagnostic model of the air conditioning system comprises one or more of a sensor diagnostic model, a water pump diagnostic model, a chiller diagnostic model, a control strategy diagnostic model, and a cooling tower diagnostic model.

8. A data labeling device for a diagnostic model, comprising

The system comprises an acquisition unit, a control unit and a control unit, wherein the acquisition unit is used for acquiring system operation data without labels;

the physical diagnosis model unit is used for marking the unmarked system operation data according to the established physical diagnosis model and generating the system operation data with the data tag;

the intelligent diagnosis model unit is used for performing machine learning model training according to the system operation data with the data characteristic labels to generate a system intelligent diagnosis model;

the diagnosis result generating unit is used for diagnosing the system according to the intelligent diagnosis model of the system and generating an intelligent diagnosis result;

and the data tag correction unit is used for performing tag correction on the marked system operation data according to the manual diagnosis result if the intelligent diagnosis result is different from the manual diagnosis result.

9. An electronic device, comprising: a processor and a memory;

the memory has stored thereon a computer readable program executable by the processor;

the processor, when executing the computer readable program, implements the steps of the method according to any of claims 1-7.

10. A computer readable storage medium storing one or more programs executable by one or more processors to implement the steps of the method of any of claims 1-7.