Operation state detection system and evaluation system of air conditioner

Abstract

The invention discloses an air conditioner operation state detection system (100) and an air conditioner operation state evaluation system. The air conditioner operation state detection system comprises an Internet of Things system (1), a data access platform (2), a service platform (3), a model filing server (4) and a business platform (5). The Internet of Things system (1) is configured to receive equipment data reported by the air conditioner (14). The data access platform (2) is configured to receive equipment data reported by the air conditioner (14) from the Internet of Things system (1), generate a prediction request based on the equipment data and push the prediction request to the service platform (3); and receiving the prediction result from the service platform (3) and reporting the prediction result to the business platform (5). The service platform (3) is configured to receive a concurrent prediction request from the data access platform (2) through the prediction interface, read the prediction model, predict the operation state of the air conditioner (14) and return a prediction result to the data access platform (2). The model archiving server (4) is configured to store and archive the prediction data and the trained prediction model. And the business platform (5) is configured to obtain a prediction result from the data access platform (2), implement target management and issue a prediction model training instruction to the service platform (3).

Description

Air conditioner operation status detection system and evaluation system

[0001] This application claims priority to Chinese patent application No. 202310252643.4 filed on March 15, 2023, and priority to Chinese patent application No. 202321286603.3 filed on May 24, 2023, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The present disclosure relates to the technical field of air conditioning, and in particular to an air conditioner operation status detection system and an evaluation system. Background Art

[0003] After the air conditioner is installed, after-sales service is usually carried out based on the customer's report of air conditioner failure. This after-sales service is based on the customer's active initiation and the after-sales service department's passive execution.

[0004] Summary of the Invention

[0005] A system for detecting the operating status of an air conditioner is provided. The system includes an Internet of Things (IoT) system, a data access platform, a service platform, a model archiving server, and a business platform. The IoT system is configured to access an air conditioner gateway and receive device data reported by the air conditioner. The data access platform is connected to the IoT system and configured to receive the device data reported by the air conditioner from the IoT system, generate prediction requests based on the device data, and push them to the service platform; further, receive prediction results from the service platform and report them to the business platform. The service platform is connected to the data access platform and includes a prediction interface and a training interface. The service platform is configured to receive concurrent prediction requests from the data access platform via the prediction interface, read the prediction model, predict the operating status of the air conditioner, and return the prediction results to the data access platform; further, receive training tasks from the business platform via the training interface and train the prediction model using an asynchronous multi-process approach. The model archiving server is connected to the service platform and configured to store and archive the prediction data and the trained prediction model. The business platform is connected to the data access platform and the service platform, and is configured to obtain prediction results from the data access platform and implement target management in combination with the prediction results; and to issue prediction model training instructions to the service platform.

[0006] An air conditioner evaluation system is also provided, comprising a data collector, a cloud platform, and a user platform. The data collection platform is connected to the air conditioner and configured to collect operating data of the air conditioner and upload it to the cloud platform. The cloud platform is connected to the data collection platform and configured to determine the health level of the air conditioner based on the operating data. The user platform is communicatively coupled to the cloud platform and configured to send a request to the cloud platform to obtain the health level of the air conditioner. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] FIG1 is a schematic diagram of an operating status detection system for an air conditioner according to some embodiments;

[0008] FIG2 is a prediction flow chart of an operating state prediction system for an air conditioner according to some embodiments;

[0009] FIG3 is another architecture diagram of an air conditioner operation status detection system according to some embodiments;

[0010] FIG4 is another architecture diagram of an air conditioner operation status detection system according to some embodiments;

[0011] FIG5 is another architecture diagram of an air conditioner operation status detection system according to some embodiments;

[0012] FIG6 is a prediction flow chart of an air conditioner operating state prediction system according to some embodiments;

[0013] FIG7 is another architecture diagram of an air conditioner operation status detection system according to some embodiments;

[0014] FIG8 is another architecture diagram of an air conditioner operation status detection system according to some embodiments;

[0015] FIG9 is another architecture diagram of an air conditioner operation status detection system according to some embodiments;

[0016] FIG10 is another architecture diagram of an air conditioner operation status detection system according to some embodiments;

[0017] FIG11 is another prediction flow chart of an operating state prediction system for an air conditioner according to some embodiments;

[0018] FIG12 is a schematic diagram showing a connection principle of an evaluation system for an air conditioner according to some embodiments;

[0019] FIG13 is another schematic diagram of a connection principle of an air conditioner evaluation system according to some embodiments;

[0020] FIG14 is a diagram illustrating another connection principle structure of an air conditioner evaluation system according to some embodiments;

[0021] FIG15 is a diagram illustrating another connection principle structure of an air conditioner evaluation system according to some embodiments;

[0022] FIG16 is a diagram illustrating another connection principle structure of an air conditioner evaluation system according to some embodiments;

[0023] FIG17 is a diagram illustrating another connection principle structure of an air conditioner evaluation system according to some embodiments;

[0024] FIG18 is a diagram illustrating another connection principle structure of an air conditioner evaluation system according to some embodiments;

[0025] FIG19 is a diagram illustrating another connection principle structure of an air conditioner evaluation system according to some embodiments;

[0026] FIG20 is a diagram illustrating another connection principle structure of an air conditioner evaluation system according to some embodiments;

[0027] FIG. 21 is another schematic diagram illustrating a connection principle of an air conditioner evaluation system according to some embodiments. DETAILED DESCRIPTION

[0028] The following will be combined with the accompanying drawings to clearly and completely describe some embodiments of the present disclosure. Obviously, the embodiments described are only some embodiments of the present disclosure, rather than all embodiments. Based on the embodiments provided by the present disclosure, all other embodiments obtained by ordinary technicians in this field are within the scope of protection of the present disclosure.

[0029] Unless the context requires otherwise, throughout the specification and claims, the term "comprise" and its other forms, such as the third person singular form "comprises" and the present participle form "comprising", are to be interpreted as open and inclusive, that is, "including, but not limited to". In the description of the specification, the terms "one embodiment", "some embodiments", "exemplary embodiments", "example", "specific example" or "some examples" are intended to indicate that the particular features, structures, materials or characteristics associated with the embodiment or example are included in at least one embodiment or example of the present disclosure. The schematic representation of the above terms does not necessarily refer to the same embodiment or example. In addition, the particular features, structures, materials or characteristics may be included in any one or more embodiments or examples in any appropriate manner.

[0030] In the following, the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of the technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of the present disclosure, unless otherwise specified, "plurality" means two or more.

[0031] When describing some embodiments, the expressions "coupled" and "connected" and their derivatives may be used. The term "connected" should be understood in a broad sense. For example, "connection" can be a fixed connection, a detachable connection, or an integral connection; it can be directly connected or indirectly connected through an intermediate medium. The term "coupled" indicates, for example, that two or more components are in direct physical or electrical contact. The term "coupled" or "communicatively coupled" may also refer to two or more components that are not in direct contact with each other, but still cooperate or interact with each other. The embodiments disclosed herein are not necessarily limited to the contents of this document.

[0032] “A and / or B” includes the following three combinations: A only, B only, and a combination of A and B.

[0033] The use of "adapted to" or "configured to" herein is intended to be open and inclusive language that does not exclude devices adapted or configured to perform additional tasks or steps.

[0034] Additionally, the use of “based on” is meant to be open and inclusive, as a process, step, calculation, or other action “based on” one or more stated conditions or values ​​may, in practice, be based on additional conditions or values ​​beyond those stated.

[0035] In some embodiments of the present disclosure, the air conditioner 14 includes an indoor unit, an outdoor unit, and an expansion valve. The outdoor unit of the air conditioner 14 includes a compressor and an outdoor heat exchanger, and the indoor unit includes an indoor heat exchanger.

[0036] The compressor is configured to compress a low-temperature, low-pressure gas-phase refrigerant and discharge the compressed high-temperature, high-pressure gas-phase refrigerant, and the high-temperature, high-pressure gas-phase refrigerant flows into the condenser.

[0037] The expansion valve may be provided in the indoor unit or the outdoor unit and is configured to expand a high-pressure liquid-phase refrigerant into a low-pressure gas-liquid two-phase refrigerant.

[0038] The indoor heat exchanger exchanges heat between indoor air and the refrigerant transported through the indoor heat exchanger to liquefy or vaporize the refrigerant. The outdoor heat exchanger is configured to exchange heat between outdoor air and the refrigerant transported through the outdoor heat exchanger to liquefy or vaporize the refrigerant.

[0039] The compressor, condenser (indoor heat exchanger or outdoor heat exchanger), expansion valve and evaporator (outdoor heat exchanger or indoor heat exchanger) perform the refrigerant cycle of the air conditioner 14. The refrigerant cycle includes a series of processes such as compression, condensation, expansion and evaporation, and circulates the refrigerant to the conditioned side.

[0040] After-sales service for the air conditioner 14 after installation is typically provided by the user, who proactively initiates the service and the after-sales service department passively performs the service. However, users typically only report operational malfunctions of the air conditioner 14, making it difficult for after-sales personnel to accurately locate and type the malfunction. This necessitates a home visit to determine the location and type of malfunction. If the malfunction cannot be corrected, a second visit is required. This cumbersome and inefficient after-sales service process degrades the user experience.

[0041] In order to predict air conditioner 14 failures in advance and locate the fault type and location, related technologies collect data on the operation of the air conditioner 14 and predict and diagnose the operating status of the air conditioner 14 based on a predictive model or diagnostic algorithm. On the one hand, this allows for timely detection of failures in their early stages and proactive initiation of maintenance procedures. On the other hand, it allows for the location and type of failures that are occurring or about to occur, thus preventing the air conditioner 14 from accumulating until it becomes unusable and being discovered. This also helps maintenance personnel perform targeted and well-prepared maintenance on the failures. However, this method of reporting faults requires the collection of a large amount of operating data from the air conditioner 14 and the support of multiple predictive or diagnostic models, placing high demands on the fault prediction and diagnosis system, which current fault prediction and diagnosis systems struggle to meet.

[0042] To address the aforementioned issues, some embodiments of the present disclosure provide an operating status detection system 100 for air conditioners 14. This system supports continuous diagnostic algorithm upgrades and meets the detection system's requirements. As shown in Figure 1 , detection system 100 includes an Internet of Things system 1, which is configured to connect to an air conditioning gateway and receive device data reported by air conditioners 14.

[0043] In some embodiments of the present disclosure, the detection system 100 further includes a data access platform 2 , a service platform 3 , a model archiving server 4 and a business platform 5 .

[0044] Data access platform 2 is configured to receive device data reported by air conditioner 14 from IoT system 1, generate a prediction request based on the received device data, and push the prediction request to service platform 3; and receive the prediction result from service platform 3 and report it to business platform 5. The prediction request includes the device data reported by air conditioner 14.

[0045] The service platform 3 includes a prediction interface 31 and a training interface 32 (as shown in Figure 2). The service platform 3 is configured to receive concurrent prediction requests from the data access platform 2 through the prediction interface 31, read the prediction model from the model archive server 4, predict the operating status of the air conditioner 14, and return the prediction results to the data access platform 2; the service platform 3 is also configured to receive prediction model training instructions from the business platform 5 through the training interface 32, parse the prediction model training instructions, generate corresponding training tasks, and train the prediction model through asynchronous multi-process.

[0046] It should be noted that it usually takes a lot of time for the service platform 3 to complete the training of a prediction model. In the related art, the service platform 3 usually uses a synchronous training mode. In this way, it cannot receive new prediction requests when the prediction model is being trained, resulting in the unavailability of the request receiving service of the service platform 3 at this time. The service platform 3 provided in some embodiments of the present disclosure trains the prediction model in an asynchronous multi-process manner, that is, the service platform 3 accepts training tasks for multiple prediction models, and starts multiple background processes to process multiple training tasks at the same time. In this way, when the service platform 3 is training the prediction model, it can continue to accept prediction requests, and the processing of prediction requests will not be blocked due to time-consuming training tasks. The multi-core resources of the server can be fully utilized to improve the training efficiency of the prediction model.

[0047] The model archiving server 4 is coupled to the service platform 3 and is configured to store and archive prediction data and trained prediction models.

[0048] The business platform 5 is configured to issue prediction model training instructions and new model publishing instructions to the service platform 3, and to obtain prediction results from the data access platform 2, and implement target management based on the prediction results.

[0049] The system 100 for detecting the operating status of an air conditioner 14 provided in some embodiments of the present disclosure comprises an Internet of Things system 1, a data access platform 2, a service platform 3, a model archiving server 4, and a business platform 5. The Internet of Things system 1 receives device data generated by the operation of the air conditioner 14 from the air conditioning gateway and reports the device data to the data access platform 2. As shown in FIG2 , the data access platform 2 generates a prediction request based on the received device data and pushes the prediction request to the service platform 3 in a concurrent manner. The service platform 3 receives concurrent prediction requests from the data access platform 2 via a prediction interface 31, reads the prediction model from the model archiving server 4, and predicts the operating status of the air conditioner 14 based on the prediction model and the device data in the prediction request, obtains a prediction result, and returns the prediction result to the data access platform 2. The data access platform 2 reports the received prediction result to the business platform 5, which implements target management such as alarms and generates maintenance processes based on the prediction result. The service platform 3 also receives prediction model training instructions from the business platform 5 through the training interface 32, generates corresponding training tasks, trains the prediction model using an asynchronous multi-process method, and stores the trained prediction model in the model archive server 4.

[0050] The operating status detection system 100 of the air conditioner 14 provided in some embodiments of the present disclosure provides complete system support for the current fault diagnosis algorithm, and realizes the reception of device data reported by the Internet of Things system, online prediction, online training of the prediction model, and timely update of the model.

[0051] In some embodiments of the present disclosure, the prediction interface 31 of the service platform 3 includes a real-time prediction interface and a prediction model acquisition interface. The real-time prediction interface is configured to receive concurrent prediction requests from the data access platform 2, and the prediction model acquisition interface is configured to read the prediction model from the model archive server 4.

[0052] As shown in Figure 3, the online prediction function of the air conditioner 14 operating status detection system 100 provided in some embodiments of the present disclosure requires the service platform 3 to support a large number of concurrent requests and quickly return prediction results. In some embodiments of the present disclosure, the service platform 3 includes a prediction interface 31 that can receive concurrent requests. This allows the service platform 3 to simultaneously receive multiple prediction requests issued concurrently by the data access platform 2, thereby improving the efficiency of the service platform 3 in training the prediction model and enabling the rapid generation of prediction results and their return to the data access platform 2.

[0053] As shown in FIG4 , different from the detection system 100 for the operating status of the air conditioner in FIG3 , the detection system 100 provided in some embodiments of the present disclosure further includes a cache platform 6 , which is connected to the data access platform 2 and is configured to cache matching data.

[0054] It should be noted that the matching data is obtained by the data access platform 2 by identifying and matching the device data, and the data access platform 2 generates a prediction request based on the matching data.

[0055] In some embodiments of the present disclosure, the device data generated by the operation of the air conditioner includes first sub-device data of the air conditioner's outdoor unit and second sub-device data of the air conditioner's indoor unit. The first sub-device data of the outdoor unit may include parameters such as operating efficiency, and the second sub-device data of the indoor unit may include parameters such as indoor temperature and wind speed.

[0056] IoT system 1 obtains the first sub-device data of the outdoor unit and the second sub-device data of the indoor unit and reports them to data access platform 2. After receiving the first sub-device data of the outdoor unit and the second sub-device data of the indoor unit from IoT system 1, data access platform 2 identifies and matches the first sub-device data of the outdoor unit and the second sub-device data of the indoor unit, respectively, to obtain complete matching data, and caches the matching data in cache platform 6. Data access platform 2 generates a prediction request based on the matching data and pushes it to service platform 3 in a concurrent manner.

[0057] As shown in Figures 5 and 6, in the air conditioner operating status detection system 100 provided in some embodiments of the present disclosure, the business platform 5 is further configured to issue a new model release instruction to the service platform 3. It should be noted that, in order to optimize the model effect, the service platform 3 can continuously train and optimize the same prediction model, and a new version of the prediction model will be generated each time the prediction model training is completed.

[0058] In some embodiments of the present disclosure, the service platform 3 receives a prediction model training instruction from the business platform 5 via the training interface 32, trains the prediction model using an asynchronous multi-process approach, stores the trained prediction model in the model archiving server 4, and feeds back a message indicating that the prediction model training has been completed to the business platform 5. To increase the speed at which the service platform 3 calls the trained prediction model, the business platform 5, after receiving feedback generated by the trained prediction model, issues a new model publishing instruction to the service platform 3. Based on the new model publishing instruction issued by the business platform 5, the service platform 3 publishes the trained prediction model to the cache platform 6. Before reading the prediction model, the service platform 3 first determines whether the cache platform 6 has a newly published prediction model. That is, the service platform 3 determines whether the version of the prediction model currently in use is lower than the latest version of the prediction model published on the cache platform 6. If so, it indicates that the version of the prediction model currently in use by the service platform 3 is lower than the latest version of the prediction model. The service platform 3 then reads the newly published prediction model from the cache platform 6. If not, it indicates that the version of the prediction model currently in use by the service platform 3 is already the latest version of the prediction model. At this point, the prediction model used by the service platform 3 does not need to be updated. In this way, the service platform 3 can quickly retrieve the trained prediction model from the buffer platform 6 to improve the prediction capability of the detection system 100.

[0059] As shown in Figure 7, in some embodiments of the present disclosure, the service platform 3 also includes a prediction result buffer 33. When the number of air conditioners 14 is only one, the prediction result buffer 33 can be implemented using memory, that is, the prediction results are stored in the memory of the server where the service platform 3 itself is located to slow down the impact on the prediction speed of the service platform 3.

[0060] The service platform 3 trains the prediction model multiple times to obtain multiple prediction results, caches the multiple prediction results in the prediction result buffer 33, and returns the prediction results to the data access platform 2 when the multiple prediction results are consistent. This can be used to improve prediction accuracy, smooth the prediction results, reduce fluctuations in the prediction results, and prevent abnormal data from destroying the stability of the prediction results.

[0061] As shown in FIG8 , the air conditioner operating status detection system 100 provided in some embodiments of the present disclosure further includes a database 7, which is configured to store the prediction results reported by the data access platform 2. The business platform 5 reads the prediction results from the database 7. In some embodiments of the present disclosure, the data access platform 2 stores the received prediction results in the database 7. The database 7 stores the prediction results within a preset number of times so that the business platform 5 can call for query or application.

[0062] As shown in FIG9 , some embodiments of the present disclosure provide an air conditioner operation status detection system 100 in a stand-alone deployment form. The detection system 100 can be used to perform operation status detection on all air conditioners in a user's room.

[0063] Some embodiments of the present disclosure use Python to design the operating program of the detection system 100, and the data access platform 2 can use Sqlite to store prediction data and prediction results.

[0064] The service platform 3 can use FastAPI (a modern, high-performance web framework for building application programming interfaces) to implement the functions of the prediction interface 31 and the training interface 32; it can also use Celery (a task scheduling framework) to implement asynchronous multi-process training services, and Celery can be deployed on the same server as the training interface.

[0065] The cache platform 6 can cache the information of the currently used prediction model (model ID, version, download address, training time, etc.), and use List (dataset) as the prediction result buffer 33 in the cache platform 6; the model archive server 4 can be in the cloud, or it can be simply built using Nginx (engine x, an open source web server and reverse proxy server) and Lua-retry-upload components.

[0066] In some embodiments of the present disclosure, the data access platform 2 receives device data reported by the IoT system 1, generates a prediction request based on the device data, and pushes the prediction request to the service platform 3. The service platform 3 receives concurrent prediction requests from the data access platform 2 through the prediction interface 31, reads the prediction model from the model archive server 4 built with Nginx and Lua-retry-upload components, combines the device data in the prediction request, predicts the operating status of the air conditioner 14 according to a preset number of times, uses the List of the cache platform 6 to record the prediction results for each preset number of times, and determines whether all prediction results within the predicted number of times are consistent. If it is determined that all prediction results within the predicted number of times are consistent, the service platform 3 returns the prediction result to the data access platform 2. If it is determined that the prediction results within the predicted number of times are not completely consistent, the service platform 3 continues to train the prediction model until all prediction results within the preset number of times are consistent. The data access platform 2 stores the obtained prediction results in the SQLite (a database) component. The business platform 5 obtains the prediction results from the SQLite component and implements target management such as alarms and generation of maintenance processes based on the prediction results. The service platform 3 also receives prediction model training instructions from the business platform 5 through the training interface 32, uses Celery to implement model training, and stores the trained prediction model in the model archive server 4 built by Nginx and Lua-retry-upload components.

[0067] As shown in FIG10 , some embodiments of the present disclosure provide an air conditioner 14 operation status detection system 100 in a distributed deployment. The detection system 100 can be used by enterprises to perform operation status detection on all users' air conditioners 14 to increase the availability of the detection system 100 .

[0068] In some embodiments of the present disclosure, the data access platform 2 accesses the Internet of Things system 1 through the MQ (Message Queue) protocol, the MQTT (Message Queuing Telemetry Transport) protocol or the API (Application Program Interface). In some embodiments of the present disclosure, the data access platform 2 also needs to call the CoolProp library (a database) to implement some characteristic parameters in the reported device data, including but not limited to pressure, gas, and temperature data.

[0069] The service platform 3 is deployed on a high-performance server and uses Redis (Remote Dictionary Server) as the prediction result buffer 33. The Python ASGI framework (such as Starlette, FastAPI, etc.) can be used to implement concurrent access to prediction requests.

[0070] Service platform 3 uses the Celery framework to implement asynchronous multi-process training services, with the Celery server deployed as a separate service. Service configuration can be selected based on the number of training tasks. Prediction models trained on service platform 3 are exported as *.joblib files and uploaded to model archive server 4. It should be noted that because service platform 3 uses asynchronous multi-process training services, after completing the assigned task, a callback to the training interface 32 is required to notify the prediction model of the training results.

[0071] When the service platform 3 is started, the trained prediction model is preloaded to increase the responsiveness of the prediction interface 31 ; the training interface 32 can be deployed separately to reduce the impact on the prediction interface 31 .

[0072] The model archiving server 4 may be implemented using a file management server to ensure the security and availability of the model files.

[0073] The cache platform 6 can be Redis; the database 7 can use MySQL (a relational database management system), PostgreSQL (a free object-relational database management system), MongoDB (a database based on distributed file storage), and ElasticSearch (a search server) to store prediction results.

[0074] As shown in Figure 11, in some embodiments of the present disclosure, the Internet of Things system 1 reports device data to the data access platform 2 through the MQ, MQTT protocol or API interface. The data access platform 2 identifies and matches the data of each device, obtains matching data, caches the matching data in Redis, and generates a prediction request based on the matching data and pushes it to the service platform 3 in a concurrent manner. The service platform 3 uses Python's ASGI (Asynchronous Server Gateway Interface) framework (such as Starlette, FastAPI, etc.) to implement concurrent access to prediction requests, reads the prediction model from the file management server of the model archive server 4, combines the device data, and predicts the operating status of the air conditioner 14 according to a preset number of times. Redis is used to store the prediction results for each of the preset number of times, and determines whether all the prediction results are consistent. If it is determined that all the prediction results are consistent, the service platform 3 returns the prediction result to the data access platform 2. If it is determined that the prediction results are not completely consistent, the service platform 3 continues to train the prediction model until the prediction results within the preset number of times are consistent. Data access platform 2 reports the prediction results to business platform 5, which then implements targeted management based on the prediction results, such as alerting and generating maintenance processes. Service platform 3 also receives prediction model training instructions from business platform 5 via training interface 32, implements model training using Celery, and exports the trained prediction model as a *.joblib file, uploading it to the file management server of model archive server 4 for storage.

[0075] In some embodiments of the present disclosure, the air conditioner 14 may be a multi-split air conditioner. In this case, the air conditioner 14 includes an outdoor unit and an indoor unit, and the indoor unit includes a plurality of indoor units connected in parallel.

[0076] The workload of air conditioner 14 and changes in its environment, such as changes in ambient temperature, humidity, and air pollution, may affect the normal operation of air conditioner 14, reducing its performance and potentially causing health problems. Furthermore, aging, wear, or damage to air conditioner 14 components can also affect its performance. For example, refrigerant leaks can reduce the cooling effect of air conditioner 14, and compressor wear can reduce its cooling capacity. These performance degradations of air conditioner 14 may not be immediately noticeable and are difficult to detect and diagnose.

[0077] When the air conditioner 14 has health problems, the user can usually only rely on his or her own experience or seek help from maintenance personnel to solve the health problems of the air conditioner 14, which is time-consuming and labor-intensive, and has a low level of intelligence.

[0078] To solve the above problems, some embodiments of the present disclosure provide an evaluation system 10 for an air conditioner 14 .

[0079] As shown in Figure 12, in some embodiments of the present disclosure, the evaluation system 10 for the air conditioner 14 includes a data acquisition platform 11 and a cloud platform 12. The data acquisition platform 11 is coupled to the air conditioner 14 and the cloud platform 12 respectively.

[0080] The data collection platform 11 is configured to collect operating data (such as temperature, humidity, pressure and other parameters) of the air conditioner 14 and upload it to the cloud platform 12. The data collection platform 11 includes, for example, sensors and the like.

[0081] In some embodiments of the present disclosure, the operating data of the air conditioner 14 includes first sub-operational data for the outdoor unit and second sub-operational data for the indoor unit. The first sub-operational data for the outdoor unit includes, for example, parameters such as temperature, humidity, and pressure, while the second sub-operational data for the indoor unit includes, for example, parameters such as indoor temperature, wind speed, and air quality. The operating data to be collected for the outdoor and indoor units can be adjusted based on actual needs and is not limited in this disclosure.

[0082] The cloud platform 12 has data processing and analysis capabilities and is configured to analyze and evaluate the operating data of the air conditioner 14 to determine the health level of the air conditioner 14. In some embodiments of the present disclosure, the cloud platform 12 may use data mining or artificial intelligence algorithm models to analyze and evaluate the operating data of the air conditioner 14.

[0083] It should be noted that the health level of the air conditioner 14 indicates the health status of the air conditioner 14 . Generally, a higher health level indicates a better health status of the air conditioner 14 and a better reliability of the air conditioner 14 .

[0084] In some embodiments of the present disclosure, referring to FIG. 12 , the evaluation system 10 for the air conditioner 14 further includes a user platform 13 , which is coupled to the cloud platform 12 .

[0085] The user platform 13 is configured to send a request to the cloud platform 12 to obtain the health level of the air conditioner 14. In some embodiments of the present disclosure, users can interact with the cloud platform 12 through the user platform 13. For example, a user can send a request to the cloud platform 12 through the user platform 13 to obtain the health level of the air conditioner 14. This facilitates the user to query and understand information related to the health status of the air conditioner 14.

[0086] The data collection platform 11 is configured to collect first sub-operation data and second sub-operation data, and upload the first sub-operation data and second sub-operation data to the cloud platform 12. The cloud platform 12 evaluates the first sub-operation data and second sub-operation data through data processing and analysis methods to determine the health level of the air conditioner 14. The user sends a request to the cloud platform 12 through the user platform 13 to obtain the health level information of the air conditioner 14. In this way, an evaluation system 10 is constructed that can help users intuitively understand the health status of the air conditioner 14 and provide users with corresponding maintenance recommendations. This allows users to easily obtain and monitor the health status of the air conditioner 14, facilitating timely implementation of appropriate measures to maintain and repair the air conditioner 14.

[0087] In some embodiments of the present disclosure, as shown in FIG13 , the data acquisition platform 11 includes at least one air-conditioning sensor component 111 and at least one first communication component 112 .

[0088] The air conditioning sensor component 111 is configured to collect operating data of the air conditioner 14 and transmit the operating data to the first communication component 112. In some embodiments of the present disclosure, the air conditioning sensor component 111 may include multiple sensors (e.g., sensors for measuring temperature, humidity, pressure, etc.). For example, at least one air conditioning sensor component 111 may include one or more sub-air conditioning sensor components.

[0089] The first communication component 112 is configured to transmit the operating data of the air conditioner 14 to the cloud platform 12 according to a preset communication method, and to receive feedback information from the cloud platform 12 regarding the receipt of the operating data. For example, the first communication component 112 may be a gateway in a centralized controller, and the preset communication method may include Bluetooth communication, wireless communication, or other communication methods. The type and communication method of the first communication component 112 can be adjusted according to actual needs and are not limited in this disclosure.

[0090] In some embodiments of the present disclosure, as shown in FIG14 , the cloud platform 12 includes a second communication component 121 , a data management component 122 and a data processing component 123 , and the data management component 122 is connected to the second communication component 121 and the data processing component 123 , respectively.

[0091] The second communication component 121 can serve as a medium for data transmission, and is configured to receive the first sub-operation data and the second sub-operation data sent by the first communication component 112 , and transmit the first sub-operation data and the second sub-operation data to the data management component 122 .

[0092] The data management component 122 is configured to perform data distribution management on the first sub-operation data and the second sub-operation data. In some embodiments of the present disclosure, the data management component 122 can distribute the first sub-operation data and the second sub-operation data to corresponding components for processing as needed.

[0093] The data processing component 123 is configured to receive the first and second sub-operational data subscribed from the data management component 122 and perform data preprocessing on the first and second sub-operational data to obtain target operational data. Data preprocessing may include operations such as data cleaning, data format conversion, outlier detection, and feature extraction to ensure that the operational data is compatible with subsequent analysis and processing.

[0094] In some embodiments of the present disclosure, the distribution relationship of the operating data of the air conditioner 14 depends on the data subscription relationship. The data processing component 123 needs to subscribe to the operating data of the data management component 122 in advance. The data management component 122 will send the operating data (including the first and second sub-operation data) to the data processing component 123 only after receiving the first and second sub-operation data uploaded by the data acquisition platform 11.

[0095] In some embodiments of the present disclosure, as shown in FIG15 , the data processing component 123 includes a data conversion subcomponent 1231 , a data matching subcomponent 1232 and a data processing subcomponent 1233 , and the data matching subcomponent 1232 is coupled to the data conversion subcomponent 1231 and the data processing subcomponent 1233 , respectively.

[0096] The data conversion sub-component 1231 is configured to receive the first sub-operation data and the second sub-operation data subscribed from the data management component 122 , and store the first sub-operation data and the second sub-operation data in a cache database.

[0097] The data conversion subcomponent 1231 converts and stores the received operational data for subsequent matching and processing. For example, the data conversion subcomponent 1231 converts the operational data into a format or structure that is more convenient for processing. The conversion process may involve operations such as converting the operational data format, cleaning the operational data, and standardizing the operational data to ensure consistency and accuracy.

[0098] In some embodiments of the present disclosure, the data conversion subcomponent 1231 includes a cache database. After the data conversion subcomponent 1231 converts the operating data of the air conditioner 14, it transfers it to the cache database. The cache database has fast read and write operations and high concurrency performance, and can provide fast data access and response. This allows for more efficient access and retrieval of operating data during subsequent operating data query and analysis, improving the efficiency and performance of processing the operating data of the air conditioner 14.

[0099] The data matching subcomponent 1232 is configured to match and integrate the first sub-operation data and the second sub-operation data in the cache database to obtain target operating data. The matching of the first sub-operation data and the second sub-operation data can be performed based on key attributes or timestamps of the air conditioner 14. Key attributes can include the device code or system number of the air conditioner 14. This ensures the accuracy and consistency of the operating data.

[0100] In some embodiments of the present disclosure, the data matching sub-component 1232 will receive the first sub-operation data of the outdoor unit and the second sub-operation data of the indoor unit from different air conditioners 14. The brands of different air conditioners 14 may be different, and the device codes, device system numbers and timestamps may all be different. Therefore, the operation data of the air conditioners 14 with the same device code, device system number and timestamp can be taken as a group of data for data combination, that is, the operation data belonging to the same air conditioner 14 at the same time can be taken as the target operation data.

[0101] The data processing subcomponent 1233 is configured to perform feature processing on the target operation data to obtain feature data of the target operation data.

[0102] It should be noted that feature processing involves extracting characteristic data (e.g., temperature, humidity, pressure, flow rate, etc.) from the raw operating data of air conditioner 14 that reflects the current state and performance of air conditioner 14, thereby supporting subsequent analysis and decision-making. Feature processing can include operations such as operating data cleaning, operating data transformation, feature extraction, and dimensionality reduction. Through the processing of operating data by data conversion subcomponent 1231, data matching subcomponent 1232, and data processing subcomponent 1233, the operating data can be better understood and utilized, enabling detection, diagnosis, and prediction of the health of air conditioner 14.

[0103] In some embodiments of the present disclosure, as shown in Figure 16, the cloud platform 12 also includes a diagnostic component 124, which is coupled to the data processing component 123 and is configured to obtain characteristic data of the target operating data from the data processing sub-component 1233, determine the health level of the air conditioner 14 based on the characteristic data, and send the health level of the air conditioner 14 to the data processing sub-component 1233.

[0104] In some embodiments of the present disclosure, after acquiring characteristic data of target operating data, the diagnostic component 124 uses a pre-trained algorithm model to determine whether the air conditioner 14 is faulty and to determine the health level of the air conditioner 14. For example, by analyzing and comparing the characteristic data, the diagnostic component 124 can detect abnormal behavior, health patterns, or potential problems of the air conditioner 14.

[0105] When the diagnosis component 124 determines that the air conditioner 14 has a health problem, it assigns a corresponding health level to the health problem and sends the health level to the data processing subcomponent 1233. It should be noted that the health level reflects the severity or priority of the health problem, for example, the health level is divided into mild, moderate and severe levels.

[0106] In some embodiments of the present disclosure, based on FIG. 16 , as shown in FIG. 17 , the diagnosis component 124 includes a diagnosis subcomponent 1241 and a scoring subcomponent 1242 , and the diagnosis subcomponent 1241 is coupled to the scoring subcomponent 1242 .

[0107] The diagnosis subcomponent 1241 is configured to obtain characteristic data of the target operation data from the data processing subcomponent 1233 and determine whether the air conditioner 14 has a fault based on the characteristic data of the target operation data.

[0108] Diagnostic subcomponent 1241 invokes a fault detection model from a database and determines whether air conditioner 14 is faulty based on the characteristic data of the target operating data. If a fault is determined in air conditioner 14, diagnostic subcomponent 1241 outputs the fault type. In some embodiments of the present disclosure, diagnostic subcomponent 1241 outputs a predefined code or label indicating the type of health issue, such as fault code 01 indicating a sensor fault, fault code 02 indicating a compressor fault, etc. If a fault is determined in air conditioner 14, diagnostic subcomponent 1241 returns an indication of normal operation.

[0109] Scoring subcomponent 1242 is configured to, if a fault occurs in air conditioner 14, call the fault diagnosis model and the characteristic data of the target operating data from diagnosis subcomponent 1241 to determine a health assessment score for air conditioner 14, determine the health level of air conditioner 14 based on the health assessment score, and transmit the health level and health assessment score to data processing subcomponent 1233. Specifically, scoring subcomponent 1242 analyzes the characteristic data of the target operating data to evaluate the overall operating status of air conditioner 14, that is, to determine the fault type of air conditioner 14 in the current state.

[0110] In some embodiments of the present disclosure, the scoring subcomponent 1242 deducts corresponding points from the original score of the air conditioner 14 based on the severity of the fault type of the air conditioner 14, with reference to the score evaluation table shown in Table 1, to calculate the health assessment score of the air conditioner 14, and then determines the health level of the air conditioner 14 according to the score range of the health assessment score.

[0111] It should be noted that the health assessment score is usually a numerical indicator, which can be a percentage or other relative quantity, indicating the current health status of the air conditioner 14.

[0112] Table 1

[0113] For example, if only the indoor expansion valve of air conditioner 14A experiences an abnormality, Table 1 indicates that 15 points should be deducted from the health assessment score of air conditioner 14A. Assuming the initial health assessment score is 100, the health assessment score is now 100-15 = 85. A score of 85 indicates that the health level of air conditioner 14 is, for example, good.

[0114] In some embodiments of the present disclosure, as shown in FIG. 18 , the cloud platform 12 further includes a third communication component 125 .

[0115] The third communication component 125 communicates with the user platform 13 and is configured to, after receiving an acquisition request from the user platform 13, communicate with the data processing subcomponent 1233 and request the health level and health assessment score of the air conditioner 14. The third communication component 125 obtains the health level and health assessment score of the air conditioner 14 from the data processing subcomponent 1233, packages the health level and health assessment score, and sends the data to the user platform 13.

[0116] In some embodiments of the present disclosure, as shown in FIG. 19 , the user platform 13 includes an input component 131 and a fourth communication component 132 , and the input component 131 is coupled to the fourth communication component 132 .

[0117] The input component 131 is configured to generate an acquisition request in response to an instruction input by a user, and transmit the acquisition request to the fourth communication component 132 .

[0118] The fourth communication component 132 is coupled to the third communication component 125 and is configured to send an acquisition request to the third communication component 125 and receive the health level and health assessment score of the air conditioner 14 sent by the third communication component 125 .

[0119] In some embodiments of the present disclosure, the user platform 13 is a user device or interface for interacting with the evaluation system 10, such as a mobile application, computer interface, or other user interface. On the user platform 13, the user can use the health self-check function of the app and click on the smart health diagnosis function to issue an operation instruction, which is received by the input component 131 and triggers a corresponding acquisition request. The input component 131 parses the user's instructions and identifies the information the user wants to obtain, such as fault pre-diagnosis results, the condition of the air conditioning filter, and the actual operating status of the air conditioner 14.

[0120] In some embodiments of the present disclosure, as shown in FIG. 20 , the user platform 13 further includes a suggestion component 133 , which is coupled to the fourth communication component 132 .

[0121] The suggestion component 133 is configured to obtain the health level and the health assessment score from the fourth communication component 132 and generate maintenance suggestions for the air conditioner 14 based on the health level and the health assessment score.

[0122] In some embodiments of the present disclosure, the recommendation component 133 generates maintenance recommendations based on the acquired health level and health assessment score using a preset algorithm or rule. For example, the recommendation is generated based on the status of the air conditioner 14, historical data, a health assessment model, or other such algorithm or rule. The maintenance recommendations may include regularly cleaning the air conditioner 14 filter, replacing the filter element, and inspecting the air conditioner 14 piping to ensure the normal operation and maintenance of the air conditioner 14.

[0123] In some embodiments of the present disclosure, as shown in FIG21 , the user platform 13 further includes a display component 134 coupled to the suggestion component 133 . The display component 134 is configured to display the health level, health assessment score, and maintenance suggestions for the air conditioner 14 .

[0124] In some embodiments of the present disclosure, the suggestion component 133 of the user platform 13 generates a health level, a health assessment score, and maintenance suggestions for the air conditioner 14, and transmits the health level, health assessment score, and maintenance suggestions for the air conditioner 14 to the display component 134. The display component 134 uses a display controller to convert the health level, health assessment score, and maintenance suggestions for the air conditioner 14 into corresponding images or text signals, and displays them to the user through the display screen. The user can interact with the display component 134 through a user interface, for example, by selecting corresponding information display or switching display pages through a touch screen.

[0125] Some embodiments of the present disclosure provide an evaluation system 10 for an air conditioner 14. This system collects operating data of the air conditioner 14 through a data acquisition platform 11 and uploads it to a cloud platform 12 for processing and analysis. This allows users to request the health level of the air conditioner 14 at any time through a user platform 13. This allows users to promptly understand the health status of the air conditioner 14 and identify potential problems as early as possible. The cloud platform 12 processes and analyzes the collected operating data of the air conditioner 14 using data mining or artificial intelligence algorithms and models.

[0126] In this way, the evaluation system 10 can help users more accurately assess the health of the air conditioner 14 through an automated diagnostic process, reducing errors in the user's subjective judgment. Furthermore, by collecting and analyzing data from the air conditioner 14, the evaluation system 10 can detect possible malfunctions of the air conditioner 14 during use. This allows users to receive early warnings before the air conditioner malfunctions occur and take appropriate measures (such as timely repairs or component replacement) to prevent the malfunction from further deteriorating and affecting the air conditioner's performance.

[0127] In addition, the user platform 13 is coupled to the cloud platform 12, and the user can send a request to the cloud platform 12 through the user platform 13 to obtain the health level of the air conditioner 14. The user operation is simple and can obtain the health status of the air conditioner 14 at any time, which improves the convenience of the user's management and maintenance of the air conditioner 14.

[0128] The evaluation system 10 for the air conditioner 14 provided in the present disclosure provides solutions such as health monitoring, automated diagnosis, health problem level assessment, and early fault warning for the air conditioner 14 through data collection, data processing, and algorithm model analysis, thereby helping users to better manage and maintain the air conditioner 14, solving the problem of accelerated aging of the air conditioner 14, reducing maintenance and replacement costs, ensuring the overall performance of the air conditioner 14, and improving the use effect and lifespan of the air conditioner 14.

[0129] Those skilled in the art will understand that the scope of the present disclosure is not limited to the above specific embodiments, and that certain elements of the embodiments may be modified and replaced without departing from the spirit of the present application. The scope of the present application is limited by the appended claims.

Claims

1. An air conditioner operation status detection system, include: The IoT system is configured to access the air-conditioning gateway and receive device data reported by the air-conditioner; A data access platform connected to the Internet of Things system, the data access platform being configured to receive device data reported by the air conditioner from the Internet of Things system, and to generate a prediction request based on the device data and push it to the service platform; and, receiving prediction results from the service platform and reporting them to the business platform; The service platform is connected to the data access platform, the service platform includes a prediction interface and a training interface, and the service platform is configured to receive concurrent prediction requests from the data access platform through the prediction interface, read the prediction model, predict the operating status of the air conditioner, and return the prediction result to the data access platform; and, receiving training tasks from the business platform through the training interface, and training the prediction model through an asynchronous multi-process approach; A model archiving server, connected to the service platform, configured to store and archive prediction data and trained prediction models; The business platform is connected to the data access platform and the service platform, and is configured to obtain prediction results from the data access platform and implement target management in combination with the prediction results; And, send prediction model training instructions to the service platform.

2. The operating status detection system according to claim 1, in, The prediction interface includes: A real-time prediction interface, configured to receive concurrent prediction requests from the data access platform; The prediction model acquisition interface is configured to read the prediction model from the model archive server.

3. The operating status detection system according to claim 1 or 2, in, The data access platform identifies and matches the device data received from the IoT system to obtain matching data, and generates a prediction request based on the matching data; The operating status detection system also includes: A cache platform is connected to the data access platform, and the cache platform is configured to cache the matching data.

4. The operating status detection system according to claim 3, in, Before the service platform reads the prediction model from the model archive server, it determines whether there is a newly released prediction model on the cache platform; If it is determined that the cache platform has a newly released prediction model, the service platform reads the newly released prediction model from the cache platform, and based on the new model release instruction issued by the business platform, publishes the trained prediction model to the cache platform.

5. The operating status detection system according to claim 4, in, The service platform includes a prediction result buffer; The service platform implements multiple predictions and caches the prediction results of a preset number of times in the prediction result buffer, and returns the prediction results when the prediction results within the preset number of times are consistent.

6. The operating status detection system according to claim 5, further comprising: include: The database is configured to store the prediction data and the prediction results reported by the data access platform, and the business platform reads the prediction results from the database.

7. The operating status detection system according to claim 6, in, The prediction data and the prediction results are stored and archived via MySQL, PostgreSQL, MongoDB, or ElasticSearch.

8. The operating status detection system according to any one of claims 1 to 7, in, The data access platform accesses the Internet of Things system through MQ, MQTT protocol or API interface.

9. The operating status detection system according to any one of claims 1 to 8, in, The pressure, gas and / or temperature data in the device data are implemented by calling the CoolProp library.

10. The operating status detection system according to any one of claims 1 to 9, in, The service platform uses Python's ASGI framework to achieve concurrent access of prediction requests; The service platform uses Python's celery framework to implement asynchronous multi-process training tasks.

11. An evaluation system for an air conditioner, include: A data collection platform connected to the air conditioner, the data collection platform is configured to collect operating data of the air conditioner and upload it to a cloud platform; a cloud platform connected to the data collection platform, the cloud platform being configured to determine a health level of the air conditioner based on the operating data; The user platform is communicatively coupled to the cloud platform device, and the user platform is configured to send an acquisition request to the cloud platform to obtain the health level of the air conditioner.

12. The evaluation system according to claim 11, in, The air conditioner comprises an outdoor unit and an indoor unit; the indoor unit comprises a plurality of indoor units connected in parallel; The data acquisition platform includes at least one air conditioning sensor component and at least one first communication component; The air conditioner sensor component is configured to collect first sub-operation data of the outdoor unit and second sub-operation data of the indoor unit, and transmit the first sub-operation data and the second sub-operation data to the first communication component; The first communication component is configured to send the first sub-operation data and the second sub-operation data to the cloud platform according to a preset communication method, and receive feedback information sent by the cloud platform; wherein the preset communication method includes one of Bluetooth communication and wireless communication.

13. An evaluation system according to claim 11 or 12, in, The cloud platform includes: a second communication component, configured to receive the first sub-operation data and the second sub-operation data sent by the first communication component, and transmit the first sub-operation data and the second sub-operation data to a data management component; The data management component is connected to the second communication component and is configured to perform data distribution management on the first sub-operation data and the second sub-operation data; The data processing component is connected to the data management component and is configured to receive the first sub-operation data and the second sub-operation data subscribed from the data management component and perform data preprocessing to obtain target operation data.

14. The evaluation system according to claim 13, in, The data processing component includes: a data conversion subcomponent, connected to the data matching subcomponent, configured to receive the first sub-operation data and the second sub-operation data subscribed from the data management component, and store the first sub-operation data and the second sub-operation data in a cache database; The data matching subcomponent is connected to the data processing subcomponent and is configured to match and integrate the first sub-operation data and the second sub-operation data in the cache database to obtain the target operation data; The data processing subcomponent is configured to perform feature processing on the target operation data to obtain feature data of the target operation data.

15. The evaluation system according to claim 11 or 12, in, The cloud platform also includes a diagnostic component coupled to the data processing component; The diagnostic component is configured to obtain characteristic data of the target operating data from the data processing subcomponent, determine the health level of the air conditioner based on the characteristic data, and send the health level of the air conditioner to The data processing subcomponent.

16. The evaluation system according to claim 15, in, The diagnosis component includes a diagnosis subcomponent and a scoring subcomponent, wherein the diagnosis subcomponent is coupled to the scoring subcomponent; The diagnosis subcomponent is configured to obtain characteristic data of the target operation data from the data processing subcomponent, and determine whether the air conditioner has a fault according to the characteristic data of the target operation data; The scoring subcomponent is configured to determine a health assessment score of the air conditioner based on characteristic data of the target operating data when a fault occurs in the air conditioner, and determine a health level of the air conditioner based on the health assessment score, and transmit the health level and the health assessment score to the data processing subcomponent.

17. The evaluation system according to claim 16, in, The cloud platform also includes a third communication component, which is configured to obtain the health level and the health assessment score from the data processing subcomponent after receiving the acquisition request sent by the user platform, and send the health level and the health assessment score to the user platform.

18. The evaluation system according to claim 17, in, The user platform includes an input component and a fourth communication component, the input component coupled to the fourth communication component; The input component is configured to generate the acquisition request in response to an instruction input by a user, and transmit the acquisition request to the fourth communication component coupling; The fourth communication component is coupled and configured to send the acquisition request to the third communication component, and receive the health level and the health assessment score sent by the third communication component.

19. The evaluation system according to claim 18, in, The user platform also includes a suggestion component, which is coupled to the fourth communication component and is configured to obtain the health level and the health assessment score from the fourth communication component, and generate maintenance suggestions for the air conditioner based on the health level and the health assessment score.

20. The evaluation system according to claim 19, in, The user platform further includes a display component, which is coupled to the suggestion component and configured to display the health level, the health assessment score, and maintenance suggestions for the air conditioner.

Images