Energy consumption monitoring method and air conditioning system

Abstract

The application discloses an energy consumption monitoring method and an air conditioning system, relates to the technical field of air conditioning monitoring, and the air conditioning system comprises a monitoring system and an air conditioning device, and the air conditioning device comprises a target air conditioner. The method comprises the following steps: acquiring target data corresponding to the target air conditioner, the target data being easy to acquire and covering multiple characteristics; performing energy consumption analysis on the target data of the target air conditioner, determining a target operating condition corresponding to the target data in the case that the target air conditioner has energy consumption abnormity, and determining a change trend of a target monitoring parameter under the target operating condition; and then, in the case that the change trend of the target monitoring parameter meets preset conditions compared with a preset fault parameter change table, a target fault type causing the energy consumption abnormity is determined. The energy consumption monitoring method can realize real-time monitoring, prediction and deep analysis of abnormal conditions, so as to effectively cope with the increasing challenges in the field of building energy consumption management, improve the building energy utilization efficiency, and realize the energy-saving and emission-reducing target.

Description

Technical Field

[0001] This application relates to the field of air conditioning monitoring technology, and in particular to an energy consumption monitoring method and an air conditioning system. Background Technology

[0002] With increasingly stringent requirements for building energy management, heating, ventilation, and air conditioning (HVAC) systems, as a major component of building energy consumption, especially air conditioning systems, are facing increasingly severe energy consumption challenges. Currently, air conditioning systems mainly rely on sensors, meters, and other equipment for energy consumption monitoring, and energy consumption prediction is based on mechanistic models and data-driven models.

[0003] Mechanism models predict energy consumption by constructing mathematical relationships between energy consumption and numerous factors such as air conditioning structural parameters, indoor and outdoor temperatures, and set temperatures. However, establishing such models requires not only a deep understanding of the structure and operating principles of air conditioning systems but also the acquisition of a large number of accurate system parameters. Given the significant structural differences among different air conditioning systems, the mechanistic models face weak generalization capabilities in practical applications.

[0004] Data-driven models, on the other hand, rely on massive amounts of historical data and employ techniques such as statistical analysis, machine learning, and deep learning to establish a mapping relationship between energy consumption and related influencing factors, thereby enabling energy consumption prediction. However, this model is a black box model and has relatively poor explanatory power for the causes of abnormal energy consumption.

[0005] The methods mentioned above are insufficient to fully meet the complex and ever-changing energy consumption monitoring needs of air conditioning systems. Therefore, there is an urgent need for a monitoring method that can achieve real-time energy consumption monitoring, prediction, and in-depth analysis of abnormal conditions. Summary of the Invention

[0006] This application provides an energy consumption monitoring method and an air conditioning system, which can realize real-time energy consumption monitoring, prediction, and in-depth analysis of abnormal conditions.

[0007] Firstly, an energy consumption monitoring method is provided, which is applied to an air conditioning system. The air conditioning system includes a monitoring system and air conditioning equipment. The air conditioning equipment includes one or more target air conditioners, and the monitoring system is communicatively connected to the air conditioning equipment.

[0008] The energy consumption monitoring method includes: acquiring target data corresponding to the target air conditioner; performing energy consumption analysis on the target data of the target air conditioner to determine whether there is an energy consumption anomaly in the target air conditioner; determining the target operating condition corresponding to the target data when the target air conditioner has an energy consumption anomaly, and determining the changing trend of the target monitoring parameters under the target operating condition, wherein the target monitoring parameters are the operating status parameters corresponding to the target data; comparing and analyzing the changing trend of the target monitoring parameters with the reference parameters in the preset fault parameter changing table, and determining the target fault type causing the energy consumption anomaly when the changing trend of the target monitoring parameters meets the preset conditions, wherein the preset conditions are used to indicate that there is a correspondence between the changing trend of the target monitoring parameters and the reference parameters in the preset fault parameter changing table.

[0009] Among them, the target data is used to indicate the operating status of the target air conditioner, such as temperature status, setting parameters, environmental parameters, etc. This operating status data is easy to obtain and is abundant.

[0010] In this embodiment, the target data is readily available, and the larger the target data, the more features and changes it encompasses. This allows for more accurate energy consumption analysis of the target air conditioner in the air conditioning system. Furthermore, under conditions of abnormal energy consumption, the method can determine the changing trends of the target monitoring parameters corresponding to the target data. Then, if the changing trends of the target monitoring parameters meet preset conditions, the type of target fault causing the energy consumption anomaly can be determined. This energy consumption monitoring method enables real-time monitoring, prediction, and in-depth analysis of abnormal conditions, effectively addressing the growing challenges in building energy management, improving building energy efficiency, and achieving energy conservation and emission reduction goals.

[0011] In one possible implementation, energy consumption analysis is performed on the target data of the target air conditioner to determine whether there is an energy consumption anomaly. This includes: inputting the target data into a benchmark energy consumption prediction model to obtain the benchmark energy consumption value corresponding to the target air conditioner, and obtaining the first actual energy consumption value of the target air conditioner through the electricity meter corresponding to the target air conditioner; performing a deviation analysis on the benchmark energy consumption value and the first actual energy consumption value to obtain a first deviation degree value; and determining that there is an energy consumption anomaly in the target air conditioner if there are M energy consumption deviations within a preset first time period during the operation of the target air conditioner and M / N is greater than a preset anomaly ratio. Here, if the first deviation degree value is greater than a preset deviation threshold, it indicates that the corresponding target data has an energy consumption deviation, and N is the number of detections within the preset first time period.

[0012] The system analyzes multiple sets of target data acquired within a preset first time period. If a certain proportion of these target data shows energy consumption deviation, it can be determined that the target air conditioner has an energy consumption anomaly within that time period. This process can improve the accuracy and comprehensiveness of energy consumption anomaly analysis.

[0013] In one possible implementation, energy consumption analysis is performed on the target data of the target air conditioner to determine whether there is an energy consumption anomaly. This includes: inputting the target data into a benchmark energy consumption prediction model to obtain the benchmark energy consumption value corresponding to the target air conditioner; inputting the target data and target monitoring parameters into a real-time energy consumption prediction model to obtain the second actual energy consumption value corresponding to the target air conditioner; performing a deviation analysis on the benchmark energy consumption value and the second actual energy consumption value to obtain a second deviation degree value; and determining that there is an energy consumption anomaly in the target air conditioner if there are M energy consumption deviations within a preset first time period during the operation of the target air conditioner and M / N is greater than a preset anomaly ratio. Here, if the second deviation degree value is greater than a preset deviation threshold, it indicates that the corresponding target data has an energy consumption deviation, and N is the number of detections within the preset first time period.

[0014] The system analyzes multiple sets of target data acquired within a preset first time period. If a certain proportion of these target data shows energy consumption deviation, it can be determined that the target air conditioner has an energy consumption anomaly within that time period. This process can improve the accuracy and comprehensiveness of energy consumption anomaly analysis.

[0015] Furthermore, in the event of meter failure, the target data and monitoring parameters of the target device are input into a pre-trained real-time energy consumption prediction model to indirectly obtain the energy consumption result. Through the aforementioned methods of determining the first and second actual energy consumption values, the actual energy consumption value can be determined from different dimensions, greatly improving the stability of the target air conditioner's energy consumption analysis and making the energy consumption analysis results more reliable and accurate.

[0016] In one possible implementation, the real-time energy consumption prediction model is trained using historical operating data within a preset third operating period of the target air conditioner, and a time-series model optimized during the operation of the target air conditioner.

[0017] It should be understood that the baseline energy consumption prediction model is trained using historical operating data of the target air conditioner, enabling it to learn the normal energy consumption levels that should exist under different environments. During the actual operation of the target air conditioner, the baseline energy consumption prediction model can be updated and optimized periodically based on the target data corresponding to the operation of the target air conditioner, in order to ensure the timeliness of the baseline energy consumption prediction model.

[0018] In one possible implementation, the baseline energy consumption prediction model is trained using historical operating data within a preset third operating period of the target air conditioner, and a regression model optimized during the operation of the target air conditioner.

[0019] The real-time energy consumption prediction model can also be understood as a virtual electricity meter. It is trained using the historical operating data of the target air conditioner. Its significance lies in learning the actual energy consumption level that should exist under different environments. The main significance of the real-time energy consumption prediction model is to continue to monitor energy consumption in real time when the electricity meter fails.

[0020] In one possible implementation, under the target operating condition, determining the changing trend of the target monitoring parameters corresponding to the target data includes: under the target operating condition, performing kernel density estimation on historical monitoring parameters to obtain the target parameter value and parameter mean corresponding to the maximum density of historical monitoring parameters, wherein the historical monitoring parameters are the operating status parameters corresponding to historical operating data, and the historical operating data are the operating data within a preset third time period of the target air conditioning operation; when the target density is less than a preset density threshold, comparing the target value with the target parameter value and the parameter mean, wherein the target density is the density value obtained by kernel density estimation of the target value corresponding to the target monitoring parameter; if the target value is greater than the target parameter value and the target value is greater than the parameter mean, then it is determined that the target monitoring parameter has an upward changing trend; if the target value is less than the target parameter value and the target value is less than the parameter mean, then it is determined that the target monitoring parameter has an abnormal downward changing trend.

[0021] By analyzing the target operating conditions described above, we can determine the specific type of abnormal change trend of the target monitoring parameters, such as an upward or downward trend. Then, we can compare and analyze the change trend of the target monitoring parameters with the reference parameters in the preset fault parameter change table to determine the specific target fault type corresponding to the energy consumption abnormality.

[0022] In one possible implementation, the reference parameters in the preset fault parameter change table include key parameters and auxiliary parameters. The preset conditions include at least one of the following: the change trend of the first parameter in the target monitoring parameters is the same as the change trend of the key parameter, and the first parameter is the parameter in the target monitoring parameters corresponding to the key parameter; the change trend of the second parameter in the target monitoring parameters is the same as the change trend of the third parameter in the auxiliary parameters, and the second parameter is the parameter in the target monitoring parameters corresponding to the third parameter, wherein the ratio of the number of third parameters to the number of auxiliary parameters is greater than a preset first proportion; and the ratio of the number of times the change trend of the first parameter is the same as the change trend of the key parameter and the change trend of the second parameter is the same as the change trend of the third parameter in a preset third duration to the number of monitoring times in the preset third duration exceeds a preset second proportion.

[0023] The process of setting multiple coexisting conditions and determining the target fault type that causes energy consumption anomalies based on these conditions can improve the accuracy and comprehensiveness of fault attribution analysis under energy consumption anomalies.

[0024] In one possible implementation, the air conditioning system also includes a communication module for acquiring outdoor environmental data, including outdoor relative humidity and solar radiation intensity; target data includes preset temperature, preset windshield, operating mode, operating load rate, indoor temperature, outdoor temperature, outdoor relative humidity, and solar radiation intensity.

[0025] Since outdoor relative humidity and solar radiation intensity are important environmental factors affecting air conditioning energy consumption, these two parameters are also included as target data. Relative humidity affects users' perception of temperature and also influences the condensation process and heat exchange efficiency of the air conditioning system. Solar radiation intensity directly affects the heat gain of buildings, thus affecting the cooling load of the air conditioning system. Therefore, a more comprehensive consideration of the impact of the actual operating environment on air conditioning energy consumption can improve the accuracy of energy consumption prediction.

[0026] In one possible implementation, before obtaining the target data corresponding to the target air conditioner, the process includes: obtaining the initial data corresponding to the target air conditioner; after the preset temperature, preset fan speed, operating mode and / or start-up load rate in the initial data are adjusted, identifying abnormal data, wherein the abnormal data is the initial data obtained within a preset second time period from the time the adjustment is completed; deleting the abnormal data in the initial data to obtain the target data.

[0027] In order to improve the stability of energy consumption analysis of target data, this application preprocesses the initial data and deletes abnormal data that is not suitable for energy consumption anomaly analysis, thereby obtaining target data with high stability.

[0028] Secondly, an energy consumption monitoring device is provided, which includes a controller and a memory. The memory is used to store computer programs, and the controller is used to call and run the computer programs from the memory, so that the energy consumption monitoring device performs the energy consumption monitoring method of the first aspect.

[0029] Thirdly, an air conditioning system is provided, including a monitoring system and air conditioning equipment, wherein the monitoring system is used to perform the energy consumption monitoring method of any one of the first aspects.

[0030] It is understood that the beneficial effects of the second and third aspects mentioned above can be found in the relevant descriptions in the first aspect above, and will not be repeated here. Attached Figure Description

[0031] Figure 1 A schematic diagram of an air conditioning system provided in an embodiment of this application is shown;

[0032] Figure 2 A schematic diagram of yet another air conditioning system provided in an embodiment of this application is shown;

[0033] Figure 3 A flowchart illustrating an energy consumption monitoring method provided in an embodiment of this application is shown;

[0034] Figure 4 A flowchart illustrating another energy consumption monitoring method provided in an embodiment of this application is shown;

[0035] Figure 5 A flowchart illustrating another energy consumption monitoring method provided in an embodiment of this application is shown;

[0036] Figure 6 A flowchart illustrating another energy consumption monitoring method provided in an embodiment of this application is shown;

[0037] Figure 7 This illustration shows a schematic diagram of kernel density estimation of historical monitoring parameters provided in an embodiment of this application;

[0038] Figure 8 This illustration shows a schematic diagram of kernel density estimation of historical monitoring parameters provided in an embodiment of this application;

[0039] Figure 9 A flowchart illustrating another energy consumption monitoring method provided in an embodiment of this application is shown;

[0040] Figure 10 A schematic diagram of the energy consumption monitoring device provided in an embodiment of this application is shown. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0042] It should be understood that "multiple" as mentioned in this application refers to two or more. In the description of this application, unless otherwise stated, " / " indicates "or," for example, A / B can mean A or B; "and / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist, for example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Furthermore, to facilitate a clear description of the technical solutions of this application, the terms "first," "second," etc., are used to distinguish identical or similar items with essentially the same function and effect. Those skilled in the art will understand that the terms "first," "second," etc., do not limit the quantity or execution order, and that "first," "second," etc., do not necessarily imply differences.

[0043] Before providing a detailed explanation of the energy consumption monitoring method and air conditioning system provided in the embodiments of this application, the application scenarios and related technologies of the energy consumption monitoring method and air conditioning system will be explained first.

[0044] With the increasing demands for building energy management, heating, ventilation and air conditioning (HVAC) systems, as a major component of building energy consumption, especially air conditioning systems, are facing increasingly severe energy consumption issues.

[0045] It should be understood that heating, ventilation, and air conditioning (HVAC) systems are an indispensable infrastructure in modern buildings, widely used in various architectural settings such as residences, office buildings, shopping malls, hospitals, schools, and industrial plants. These systems integrate the three core functions of heating, ventilation, and air conditioning.

[0046] An air conditioning system is a device and system used to regulate indoor environmental parameters to provide a comfortable space. It is a comprehensive system composed of multiple subsystems and devices. Figure 1 A schematic diagram of an air conditioning system provided in an embodiment of this application is shown, such as... Figure 1 As shown, the air conditioning system mainly includes a monitoring system 100 and an air conditioning unit 200. The monitoring system 100 and the air conditioning unit 200 are connected by wired or wireless means to realize the status monitoring and dynamic control of the air conditioning unit 200.

[0047] Figure 2 A schematic diagram of another air conditioning system provided in an embodiment of this application is shown, such as... Figure 2 The diagram illustrates an air conditioning system as an example. The air conditioning system mainly includes a monitoring system 100 and air conditioning equipment 200. The monitoring system 100 is a crucial subsystem of the air conditioning system, used to monitor and control the operation of the entire system to ensure stable and efficient operation according to set parameters and modes. On one hand, the monitoring system 100 can perform logical calculations and judgments based on preset parameters such as temperature and operating mode, as well as feedback signals from various sensors, meters, and other devices, and then quickly issue corresponding control commands to achieve precise regulation of the operating status of each component of the air conditioning system. On the other hand, the monitoring system 100 continuously collects and deeply analyzes the operating data of the air conditioning equipment during operation, monitoring and analyzing the energy consumption of the air conditioning equipment. In other words, the monitoring system 100 is used to execute any of the energy consumption monitoring methods described below.

[0048] Air conditioning equipment 200 is the core component for achieving air conditioning functions. It includes various types of air conditioning units, ventilation equipment, cold and heat source equipment, etc., and is directly responsible for processing indoor air to achieve the purpose of regulating temperature, humidity, and purifying air.

[0049] It should be understood that air conditioning equipment 200 includes one or more air conditioning units, for example, air conditioning equipment 200 includes three multi-split air conditioners, such as... Figure 2 As shown.

[0050] An air conditioning unit is a device used to regulate environmental parameters such as indoor temperature, humidity, cleanliness, and airflow speed. According to different classification standards, it can be divided into many types, such as multi-split air conditioning units (referred to as multi-split air conditioners) and ducted air conditioners (referred to as ducted units). It can achieve independent temperature regulation of multiple spaces and has advantages such as convenient installation, space saving, and flexible control.

[0051] Currently, air conditioning systems mainly rely on devices such as sensors and meters to monitor energy consumption, and use methods such as mechanistic models and data-driven models to predict energy consumption.

[0052] Mechanistic models, also known as white-box models, describe the energy consumption behavior of a system by establishing mathematical equations based on the fundamental physical and chemical principles of the system. For example, based on thermodynamic laws and heat transfer principles, they consider the working processes of equipment such as compressors, heat exchangers, and fans, establishing mathematical relationships between energy consumption and factors such as indoor and outdoor temperature, humidity, air conditioning operating time, and set temperature to achieve energy consumption prediction. While these models have clear physical meaning and can deeply reflect the essence of energy consumption, establishing them requires not only a deep understanding of the air conditioning structure and operating principles but also a large amount of accurate system parameters. Given the significant structural differences among different air conditioners, mechanistic models face problems such as weak generalization ability and poor predictive accuracy in practical applications.

[0053] Data-driven models, also known as black-box models, do not rely on the internal mechanisms of a system. Instead, they are based on massive amounts of historical data and utilize techniques such as statistical analysis, machine learning, and deep learning to establish a mapping relationship between energy consumption and related influencing factors, thereby enabling energy consumption prediction. While these models are relatively simple to build, can handle complex nonlinear relationships, and have lower data requirements, they lack physical interpretability, resulting in relatively poor explanatory power for the causes of energy consumption anomalies.

[0054] The methods used in the aforementioned technologies are all insufficient to fully meet the complex and ever-changing needs of air conditioning energy consumption monitoring.

[0055] In view of this, embodiments of this application provide an energy consumption monitoring method and an air conditioning system. The air conditioning system includes a monitoring system and air conditioning equipment. The air conditioning equipment includes one or more target air conditioners, and the monitoring system is communicatively connected to the air conditioning equipment. The energy consumption monitoring method includes: acquiring target data corresponding to the target air conditioners; performing energy consumption analysis on the target data of the target air conditioners to determine whether there is an energy consumption anomaly in the target air conditioners; if there is an energy consumption anomaly in the target air conditioners, determining the target operating condition corresponding to the target data, and determining the changing trend of the target monitoring parameters under the target operating condition; comparing and analyzing the changing trend of the target monitoring parameters with the reference parameters in a preset fault parameter changing table, and determining the target fault type causing the energy consumption anomaly if the changing trend of the target monitoring parameters meets preset conditions.

[0056] The target data is readily available, and its greater volume encompasses more features and variations, enabling more accurate energy consumption analysis of target air conditioners within the air conditioning system. Furthermore, it allows for the determination of the changing trends of target monitoring parameters corresponding to the target data under energy consumption anomalies. Subsequently, if the changing trends of the target monitoring parameters meet preset conditions, the type of target fault causing the energy consumption anomaly can be identified. This energy consumption monitoring method enables real-time monitoring, prediction, and in-depth analysis of anomalies, effectively addressing the growing challenges in building energy management, improving building energy efficiency, and achieving energy conservation and emission reduction goals.

[0057] To facilitate a further understanding of the technical solutions in some embodiments of this application, the technical solutions of the energy consumption monitoring method and how these solutions solve the aforementioned technical problems are described in detail below with reference to specific embodiments and accompanying drawings. The embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. Obviously, the described embodiments are only some, not all, of the embodiments of this application.

[0058] Figure 3 This document illustrates a flowchart of an energy consumption monitoring method provided in an embodiment of this application. Figure 3 As shown, the energy consumption monitoring method includes the following steps:

[0059] S310. Obtain the target data corresponding to the target air conditioner.

[0060] In this context, "target air conditioner" refers to any one of the air conditioners in the air conditioning system. When the air conditioning system contains multiple air conditioners, "target air conditioner" can refer to any one of them.

[0061] The target air conditioner can be a multi-split air conditioner, a ducted air conditioner, etc. For ease of description, this application uses a multi-split air conditioner as an example. A multi-split air conditioner is an air conditioning device that connects one or more outdoor units to several indoor units (refrigerant-air heat exchangers) of different or the same type and capacity, and can directly supply air to one or more areas.

[0062] The target data corresponding to the target air conditioner is obtained through real-time monitoring of the target air conditioner and is used to indicate the operating status of the target air conditioner. The real-time operating data obtained from the real-time monitoring of the target air conditioner includes numerous data items, such as data monitored by sensors and other devices on the target air conditioner, control data of the target air conditioner, operating environment data of the target air conditioner, monitoring parameters of the target air conditioner (such as target monitoring parameters), etc., while the target data only includes a portion of the real-time operating data.

[0063] Target data may include preset temperature, preset windshield, operating mode, operating load rate, indoor temperature, outdoor temperature, outdoor relative humidity, and solar radiation intensity.

[0064] It should be noted that in other embodiments of this application, the composition of the target data can be varied. The target data may contain more data items than those listed above, or it may cover only a few data items. Alternatively, the target data may be a combination of selected data from the above-mentioned data, thereby flexibly adapting to specific scenarios and analysis purposes.

[0065] Among them, the preset temperature, preset fan speed, operating mode, and start-up load rate are key data used to indicate the desired operating status of the target air conditioner, which are set and adjusted by the user or determined based on the initial parameters of the target air conditioner.

[0066] The preset temperature is the target indoor temperature set by the user based on their own comfort needs. For example, in summer, people usually set the preset temperature to around 26℃; in winter, it might be set to 22℃ to obtain a warm indoor environment. The preset fan speed refers to the operating speed setting of the indoor unit fan in the target air conditioner. Common speeds include low, medium, high, and automatic.

[0067] The target air conditioner offers multiple operating modes to meet different usage needs. Common operating modes include cooling mode, heating mode, dehumidification mode, and ventilation mode. In hot summers, selecting cooling mode allows the target air conditioner to transfer indoor heat to the outdoors, lowering the indoor temperature; in cold winters, heating mode provides warmth to the room. Dehumidification mode primarily reduces indoor humidity. In ventilation mode, the target air conditioner does not cool or heat, but only uses a fan to circulate indoor and outdoor air, serving a ventilation function.

[0068] Operating load rate represents the ratio of the total capacity of all operating indoor units in a target air conditioning system to the capacity of the outdoor unit. It is a key indicator for measuring the operating status of multi-split air conditioners or similar systems. This ratio directly reflects the load of the indoor units carried by the outdoor unit at a specific moment.

[0069] It should be understood that for multi-split air conditioners, the operating load rate is closely related to energy consumption. Generally speaking, within a certain range, the higher the operating load rate, the higher the energy consumption. By monitoring the operating load rate, the energy consumption status of the air conditioning system can be better analyzed and judged.

[0070] For example, when the operating load rate is consistently high, such as approaching 100%, it indicates that the outdoor unit is operating at full capacity to match the total demand from the indoor units. In this state, the system is running at full load. Further increasing the number of indoor units or increasing their load may lead to a decrease in cooling (heating) efficiency, a significant increase in energy consumption, and even affect the equipment's lifespan. Conversely, if the operating load rate is low, such as only 30%, it indicates that the actual operating load of the outdoor unit is far below its rated capacity. Prolonged low-load operation may result in reduced energy efficiency and energy waste.

[0071] It should be understood that the air conditioning system also includes a communication module, which establishes a connection with external systems (such as the local meteorological bureau) via a network to achieve data exchange. The communication module can be used to acquire outdoor environmental data, including at least outdoor relative humidity and solar radiation intensity.

[0072] Since outdoor relative humidity and solar radiation intensity are important environmental factors affecting air conditioning energy consumption, these two parameters are also included as target data. Relative humidity affects users' perception of temperature and also influences the condensation process and heat exchange efficiency of the air conditioning system. Solar radiation intensity directly affects the heat gain of buildings, thus affecting the cooling load of the air conditioning system. Therefore, a more comprehensive consideration of the impact of the actual operating environment on air conditioning energy consumption can improve the accuracy of energy consumption prediction.

[0073] S320. Perform energy consumption analysis on the target data of the target air conditioner to determine whether there is any energy consumption anomaly in the target air conditioner.

[0074] The energy consumption analysis of the target air conditioner's data includes: obtaining the baseline energy consumption value corresponding to the target data, and the actual energy consumption value of the target air conditioner when the target data was obtained; then, performing a deviation analysis on the baseline energy consumption value and the actual energy consumption value to obtain the corresponding deviation degree value. If the deviation degree value is greater than a preset deviation threshold, it indicates that the corresponding target data has an energy consumption deviation. In other words, when the target data has an energy consumption deviation, it indicates that the target air conditioner has an energy consumption anomaly.

[0075] The target data can be input into a pre-trained baseline energy consumption prediction model to obtain the corresponding baseline energy consumption value, also known as the normal energy consumption value. It should be understood that the baseline energy consumption value serves as a reference standard for analyzing actual energy consumption values. The baseline energy consumption prediction model is trained using historical operating data from the target air conditioner.

[0076] Because the target data can cover more features and changes, the baseline energy consumption prediction model can learn more complex and subtle patterns and rules, reduce the risk of overfitting, and improve its generalization ability on unknown data, thereby improving the accuracy of the baseline energy consumption prediction model.

[0077] It should be understood that historical operating data encompasses at least three corresponding data items: target data-related items, monitoring-related items, and energy consumption-related items. Historical operating data may be historical operating data of the target air conditioner and / or laboratory data of the target air conditioner during the design and development phase.

[0078] Target data related items refer to data whose categories are completely consistent with the target data mentioned above, such as preset temperature, preset fan speed, operating mode, operating load rate, indoor temperature, outdoor temperature, outdoor relative humidity, and solar radiation intensity. Monitoring related items are the operating status parameters of the target air conditioner (i.e., monitoring parameters), also known as historical monitoring parameters; energy consumption related items can be specifically understood as the electricity consumption of the target air conditioner.

[0079] In some embodiments, historical operating data originates from all operating data generated within a preset third time period after the target air conditioner has been put into use; wherein, the monitoring-related items are the operating status parameters of the target air conditioner within the preset third time period; and the energy consumption-related items can be specifically understood as the electricity consumption of the target air conditioner within the preset third time period. In other embodiments, historical operating data is operating status data collected by researchers during the research and development process. It should be understood that the duration for which the operating status data was collected during the experiment is greater than or equal to the preset third time period.

[0080] To improve the accuracy and comprehensiveness of energy consumption anomaly analysis, in some embodiments, analysis can be performed on multiple sets of target data acquired within a specific time period (such as a preset first time period). When a certain proportion of these target data shows energy consumption deviation, it can be determined that the target air conditioner has an energy consumption anomaly within that time period.

[0081] S330. When the target air conditioner has abnormal energy consumption, determine the target operating condition corresponding to the target data.

[0082] It should be understood that before determining the target operating condition corresponding to the target data, it is necessary to use historical operating data to determine all possible operating conditions that the target air conditioner may present during operation, and then determine the target operating condition corresponding to the target data from these operating conditions.

[0083] The operating conditions refer to the various states and conditions exhibited by the analyzed object (target air conditioner) during operation, also known as the operating environment. In this embodiment, the operating conditions of the target air conditioner encompass several key elements, specifically outdoor temperature, indoor temperature, operating mode, number of indoor units activated, and preset temperature. Outdoor temperature, as an external environmental condition, is a crucial factor affecting the energy consumption and cooling (heating) effect of the target air conditioner; indoor temperature reflects the actual temperature of the current indoor environment, and its difference from the outdoor temperature affects the air conditioner's load. Each operating mode corresponds to a different working principle and function. The number of indoor units activated determines the spatial range and load that the target air conditioner needs to handle, etc.; these elements collectively constitute the operating conditions of the target air conditioner. Furthermore, the number of operating conditions can be determined using the Elbow Method.

[0084] In some embodiments, operating conditions can be determined by performing unsupervised clustering analysis on target data related items in historical operating data (i.e., a second historical dataset). Unsupervised clustering is a modeling framework in deep learning that involves dividing samples in a second historical dataset into different clusters based on similarity without given labels or prior knowledge; that is, determining operating conditions. Samples within each cluster have high similarity, while samples between different clusters have significant differences.

[0085] Among them, the unsupervised clustering algorithm can be one of the following: K-Means Clustering Algorithm (K-Means algorithm), hierarchical clustering algorithm, and Density-Based Spatial Clustering of Applications with Noise (DBSCAN algorithm). Taking the K-Means algorithm as an example, the initial cluster centers are first randomly selected, and then each data point is assigned to the cluster containing the nearest cluster center. Then, the cluster center of each cluster is updated to the mean of all data points in the cluster. The assignment and update steps are repeated until the cluster centers no longer change or the preset number of iterations is reached.

[0086] Furthermore, for each operating condition, a cluster center is determined. The cluster center is usually a statistical measure such as the mean or median of all data points in that operating condition, which represents the typical characteristics of that operating condition.

[0087] After determining the operating conditions and their cluster centers, the cluster center closest to the target data can be calculated based on Euclidean distance to obtain the target cluster center corresponding to the target data, as well as the operating conditions corresponding to the target cluster center, i.e., the target operating conditions corresponding to the target data.

[0088] S340. Under the target operating conditions, determine the changing trend of the target monitoring parameters.

[0089] It should be understood that target monitoring parameters refer to the operating status parameters of the target air conditioner when the target data is acquired. These operating status parameters correspond to the target data in time and match the acquisition time of the target data.

[0090] In other words, the target monitoring parameters are the operating status parameters corresponding to the target data, such as compressor frequency, condenser outlet subcooling, indoor unit electronic expansion valve opening, outdoor unit electronic expansion valve opening, compressor discharge pressure, evaporation temperature, condensation temperature, compressor suction pressure, air outlet temperature, compressor discharge superheat, difference between condenser outlet temperature and ambient temperature, condenser outlet temperature, and operating current.

[0091] Under the target operating conditions, the changing trend of the target monitoring parameters corresponding to the target data can be determined, such as an upward trend or a downward trend.

[0092] S350. Compare and analyze the changing trend of the target monitoring parameters with the reference parameters in the preset fault parameter change table. If the changing trend of the target monitoring parameters meets the preset conditions, determine the target fault type that causes abnormal energy consumption.

[0093] Among them, the preset conditions are used to indicate that there is a corresponding relationship between the changing trend of the target monitoring parameters and the reference parameters in the preset fault parameter change table.

[0094] It should be understood that the preset fault parameter change table is constructed based on engineering experience and R&D data, and is a table comparing parameter changes. It clearly presents the changing trends of the corresponding monitoring parameters under various fault conditions, providing a key reference for fault diagnosis.

[0095] Taking a multi-split air conditioner as an example, the corresponding preset fault parameter change tables are specifically introduced using Tables 1, 2, and 3. The reference parameters in the preset fault parameter change tables include key parameters and auxiliary parameters. Among them, key parameters play a core and decisive role in the operating status and fault judgment of the target equipment; while auxiliary parameters, although not as decisive as key parameters, can provide supplementary information to help improve the analysis and judgment of faults.

[0096] It should be understood that the preset faults are not limited to the refrigerant overcharge fault, refrigerant undercharge fault, and indoor heat exchanger blockage fault presented in Tables 1, 2, and 3, but also include other faults, such as outdoor unit heat exchanger blockage faults, etc., which will not be listed here. For faults that conform to the parameter change trend, there is a corresponding preset fault parameter change table. Therefore, the preset fault parameter change table is not limited to Tables 1, 2, and 3, but also includes many other fault parameter change tables.

[0097] The parameter change tables for each fault include numerous parameter items. Tables 1, 2, and 3 mainly show the changing trends of key parameters. It should be understood that the parameter change tables for each fault also include the changing trends of auxiliary parameters. Due to the large amount of data, they are not listed here.

[0098] Table 1. Parameter Changes in Refrigerant Overcharge Fault

[0099]

[0100] Table 2. Parameter Changes for Refrigerant Insufficiency Faults

[0101]

[0102] Table 3. Parameter Changes in Indoor Heat Exchanger Clog Faults

[0103]

[0104] The changing trends of the target monitoring parameters are compared and analyzed with the reference parameters in the preset fault parameter change table. If the changing trends of the target monitoring parameters meet preset conditions, the type of target fault causing abnormal energy consumption is determined. The preset conditions indicate a correspondence between the changing trends of the target monitoring parameters and the reference parameters in the preset fault parameter change table.

[0105] This application provides an energy consumption monitoring method, which involves acquiring target data corresponding to a target air conditioner; performing energy consumption analysis on the target data of the target air conditioner to determine whether the target air conditioner has an energy consumption anomaly; determining the target operating condition corresponding to the target data when the target air conditioner has an energy consumption anomaly, and determining the changing trend of the target monitoring parameters under the target operating condition; comparing and analyzing the changing trend of the target monitoring parameters with the reference parameters in the preset fault parameter changing table, and determining the target fault type that causes the energy consumption anomaly when the changing trend of the target monitoring parameters meets the preset conditions.

[0106] It should be understood that the target data in this application embodiment is readily available, and the larger the target data, the more features and changes it can cover. This allows for more accurate energy consumption analysis of the target air conditioner in the air conditioning system, and in the event of abnormal energy consumption, it can determine the changing trend of the target monitoring parameters corresponding to the target data. Furthermore, if the changing trend of the target monitoring parameters meets preset conditions, it can determine the target fault type causing the abnormal energy consumption. This energy consumption monitoring method can achieve real-time monitoring, prediction, and in-depth analysis of abnormal conditions, effectively addressing the growing challenges in the field of building energy management, improving building energy efficiency, and achieving energy conservation and emission reduction goals.

[0107] The above Figure 3 In step 320, energy consumption analysis is performed on the target data of the target air conditioner to determine whether there is an energy consumption anomaly in the target air conditioner. First, it is necessary to obtain the baseline energy consumption value corresponding to the target data, as well as the actual energy consumption value of the target air conditioner when the target data is obtained.

[0108] The acquisition of baseline energy consumption and actual energy consumption can be achieved by inputting the target data into a pre-trained baseline energy consumption prediction model to obtain the corresponding baseline energy consumption value; and by obtaining the actual energy consumption value through the electricity meter corresponding to the target air conditioner.

[0109] In the event of a failure such as a malfunction, damage, or unexplained shutdown of the meter, the target data and corresponding target monitoring parameters can be input into a pre-trained real-time energy consumption prediction model to obtain the actual energy consumption value.

[0110] In some embodiments, when the electricity meter is functioning normally, the target data and the corresponding target monitoring parameters can be input into the trained real-time energy consumption prediction model to obtain the corresponding actual energy consumption value.

[0111] It should be understood that both the baseline energy consumption prediction model and the real-time energy consumption prediction model are models trained from historical operating data.

[0112] Historical operating data originates from all operating data generated within a preset third time period after the target air conditioner has been put into use. It covers at least three data items: target data-related items, monitoring-related items, and energy consumption-related items. The monitoring-related items are the operating status parameters of the target air conditioner within the preset third time period; the energy consumption-related items can be specifically understood as the electricity consumption of the target air conditioner within the preset third time period. In other embodiments, historical operating data refers to the operating status data collected by researchers during the research and development process. It should be understood that the duration of the collected operating status data is greater than or equal to the preset third time period.

[0113] In some embodiments, the preset third duration includes at least one cooling cycle and one heating cycle, for example, 10 months, 1 year, or 2 years. In other embodiments, the preset third duration includes at least one cooling cycle, for example, 3 months or 5 months.

[0114] It should be understood that the data for multi-split air conditioners vary depending on the environmental conditions and usage scenarios. These data may include outdoor temperature, indoor temperature, preset temperature, operating mode, operating load rate, and monitoring parameters such as compressor frequency and condenser outlet subcooling.

[0115] The operation of a target air conditioner is affected by various factors such as season, ambient temperature, and usage habits. By collecting historical operating data for at least one year of cooling and heating seasons, the operating status of the target air conditioner under different operating conditions can be comprehensively and realistically reflected, including changes in operating parameters under different preset temperatures and different operating load rates. This provides rich and realistic training samples for the baseline energy consumption prediction model and the real-time energy consumption prediction model, enabling the baseline energy consumption prediction model and the real-time energy consumption prediction model to learn various possible operating modes.

[0116] The baseline energy consumption prediction model is trained using historical operating data of the target air conditioner. Its significance lies in learning the normal energy consumption levels that should exist under different environments. The target data-related terms and energy consumption-related terms in the historical operating data constitute the first historical dataset. Regression training can be performed using this first historical dataset. The main goal of regression training is to construct a regression model based on the given first historical dataset, enabling the model to describe the relationship between independent and dependent variables as accurately as possible. By learning patterns and regularities in the training data, the model can predict new and unseen target data, providing corresponding baseline energy consumption values.

[0117] In other words, based on the target data related to the historical operation data of the target air conditioner, the model learning parameters are used to perform regression training on the energy consumption related items to establish a baseline energy consumption prediction model. This baseline energy consumption prediction model is a regression model, meaning it can be one of the following regression models: logistic regression, support vector regression, decision tree regression, random forest regression, or neural network regression.

[0118] It should be understood that during the actual operation of the target air conditioner, the benchmark energy consumption prediction model can be updated and optimized periodically based on the target data corresponding to the operation of the target air conditioner to ensure the timeliness of the benchmark energy consumption prediction model.

[0119] The real-time energy consumption prediction model can also be understood as a virtual electricity meter. It is trained using the historical operating data of the target air conditioner. Its significance lies in learning the actual energy consumption level that should exist under different environments. The main significance of the real-time energy consumption prediction model is to continue to monitor energy consumption in real time when the electricity meter fails.

[0120] Specifically, target data-related items and monitoring-related items from the historical operation data of the target air conditioner are used as model learning parameters to perform time-series regression training on energy consumption-related items and establish a real-time energy consumption prediction model.

[0121] The real-time energy consumption prediction model is a time-series regression model. That is, the real-time energy consumption prediction model can be a Long-Short Term Memory (LSTM) model or an Autoregressive Integrated Moving Average (ARIMA) model.

[0122] It should be understood that during the actual operation of the target air conditioner, the real-time energy consumption prediction model can be updated and optimized periodically based on the target data and target monitoring parameters corresponding to the operation of the target air conditioner, so as to ensure the timeliness of the real-time energy consumption prediction model.

[0123] After the baseline energy consumption prediction model and the real-time energy consumption prediction model are trained... Figure 4 The following is a flowchart illustrating another energy consumption monitoring method provided in an embodiment of this application, as shown below. Figure 4 As shown above, Figure 3 Step 320, which involves performing energy consumption analysis on the target data of the target air conditioner to determine whether there is any energy consumption anomaly in the target air conditioner, also includes the following steps:

[0124] S321. Input the target data into the benchmark energy consumption prediction model to obtain the benchmark energy consumption value corresponding to the target air conditioner.

[0125] S322A. Under the condition that the electricity meter corresponding to the target air conditioner is working normally, obtain the first actual energy consumption value of the target air conditioner through the electricity meter.

[0126] S323A. Deviation analysis is performed on the baseline energy consumption value and the first actual energy consumption value to obtain the first deviation degree value.

[0127] The first deviation value is determined by calculation methods such as absolute deviation or relative deviation. Taking the calculation method of relative deviation as an example, it can more intuitively reflect the magnitude of the deviation. The first deviation value is obtained by comparing the baseline energy consumption value and the first actual energy consumption value. The first deviation value is calculated through the following process:

[0128]

[0129] Where Dev1 is the first deviation value, W normal As the baseline energy consumption value, W now1 This is the first actual energy consumption value.

[0130] S324A. If there are M instances of energy consumption deviation within a preset first operating time of the target air conditioner and the ratio M / N is greater than a preset abnormality ratio, it is determined that the target air conditioner has an energy consumption abnormality.

[0131] If the first deviation value is greater than the preset deviation threshold, it indicates that the corresponding target data has energy consumption deviation. For example, if the preset deviation threshold is ε, then if Dev1>ε, it indicates that the corresponding target data has energy consumption deviation.

[0132] Where N represents the number of detections within a preset first duration, which can be set as needed, such as 1 hour or 2 hours. The frequency of acquiring target data within this preset first duration is also flexible, and can be acquired once every 2 minutes, once every 3 minutes, or once every 5 minutes.

[0133] For example, if the preset duration is 2 hours and the target data is acquired every 5 minutes, then N is 24 times. When the preset anomaly rate is 50%, the target air conditioner can be determined to have abnormal energy consumption under the following conditions:

[0134] If the number of energy consumption deviations (M) is greater than 12 during the 2-hour operation period of the target air conditioner, such as 13 or 16 times, the ratio of M to 24 will be greater than 50%, which indicates that the target air conditioner has abnormal energy consumption.

[0135] It should be noted that, for the target air conditioner, it performs high-frequency monitoring of the preset temperature, preset fan speed, operating mode, start-up load rate, indoor temperature and outdoor temperature in the target data. The monitoring frequency is such as once every 5 seconds or once every 10 seconds. The data collected each time will be stored in the target air conditioner's built-in memory in real time.

[0136] In this embodiment, the number of detections N within a preset first time period refers to the frequency at which the target air conditioner uploads target data such as the preset temperature, preset fan speed, operating mode, operating load rate, indoor temperature, and outdoor temperature to the monitoring system within the preset first time period. The frequency at which the target air conditioner uploads data to the monitoring system can be set to once every 2 minutes, once every 3 minutes, or once every 5 minutes.

[0137] Therefore, by analyzing multiple sets of target data acquired within a preset first time period, if a certain proportion of these target data shows energy consumption deviation, it can be determined that the target air conditioner has an energy consumption anomaly within that time period. This process can improve the accuracy and comprehensiveness of energy consumption anomaly analysis.

[0138] In some embodiments, Figure 5 The following is a flowchart illustrating another energy consumption monitoring method provided in an embodiment of this application, as shown below. Figure 5 As shown above, Figure 4 Following step 321, the following steps may also be included:

[0139] S322B: Input the target data and target monitoring parameters into the real-time energy consumption prediction model to obtain the second actual energy consumption value corresponding to the target air conditioner.

[0140] Typically, step 322B is performed when the electricity meter fails. In other embodiments, when the electricity meter is functioning normally, the target data and corresponding target monitoring parameters can be input into a pre-trained real-time energy consumption prediction model to obtain the corresponding second actual energy consumption value.

[0141] It should be understood that the setting of the first and second actual energy consumption values ​​is mainly to clearly distinguish the two different ways of obtaining actual energy consumption values. The first actual energy consumption value is the energy consumption data obtained through real-time collection by the electricity meter; while the second actual energy consumption value is the energy consumption result indirectly obtained by inputting the target data and target monitoring parameters of the target device into a pre-trained real-time energy consumption prediction model.

[0142] By using the above methods to determine the first and second actual energy consumption values, the actual energy consumption values ​​can be determined from different dimensions, which greatly improves the stability of the target air conditioner energy consumption analysis and makes the energy consumption analysis results more reliable and accurate.

[0143] S323B: Perform deviation analysis on the baseline energy consumption value and the second actual energy consumption value to obtain the second deviation degree value.

[0144] S324B: If there are M instances of energy consumption deviation within a preset first operating time of the target air conditioner and the ratio M / N is greater than a preset abnormality ratio, it is determined that the target air conditioner has an energy consumption abnormality.

[0145] Wherein, if the second deviation value is greater than the preset deviation threshold, it indicates that the corresponding target data has energy consumption deviation, and N is the number of detections within the preset first time period.

[0146] For details on the implementation of steps 323B and 324B, please refer to the detailed description of steps 323A and 324A above, which will not be repeated here.

[0147] It should be noted that in the above embodiments, the preset temperature, preset fan speed, operating mode, and operating load rate in the target data are all data items that the target air conditioner user can adjust themselves. When one or more of these data items are adjusted within a preset second time period, the response data exhibits transient characteristics. This type of data is highly volatile and is not suitable as target data for energy consumption anomaly analysis.

[0148] In the above embodiments, the preset temperature, preset fan speed, operating mode, and operating load rate in the target data are all data items that the target air conditioner user can adjust themselves. When one or more of these data items are adjusted, the corresponding data within a preset second time period is transient data, meaning data that changes rapidly over time and is short-lived and immediate within the process corresponding to the preset second time period. This type of data changes significantly and frequently, exhibiting poor stability, and is therefore unsuitable as target data for energy consumption anomaly analysis. Therefore, in Figure 3 Before step 310 of the process to obtain the target data corresponding to the target air conditioner, the following steps (1)-(3) are also included:

[0149] (1) Obtain the initial data corresponding to the target air conditioner.

[0150] It should be understood that the initial data and target data cover the same data items. The initial data includes preset temperature, preset windshield, operating mode, start-up load rate, indoor temperature, outdoor temperature, outdoor relative humidity, and solar radiation intensity.

[0151] It should also be understood that the data processing mainly targets the preset temperature, preset windshield, operating mode, and start-up load rate in the initial data.

[0152] (2) After the preset temperature, preset windshield, operating mode and / or start-up load rate in the initial data are adjusted, abnormal data is identified.

[0153] Abnormal data refers to the initial data obtained within a preset second time period starting from the time the adjustment is completed.

[0154] When one or more data items in the initial data are adjusted, the corresponding data acquired within a preset second time period from the time the adjustment is completed are process transient data, which are considered abnormal data. These abnormal data exhibit large and frequent fluctuations and poor stability, making them unsuitable as target data for energy consumption anomaly analysis. Therefore, these abnormal data need to be deleted.

[0155] The preset second duration is the process by which the target air conditioner stabilizes after one or more data items among the preset temperature, preset fan speed, operating mode, and operating load rate are adjusted. After the preset second duration is reached, the operating status of the target air conditioner is stable.

[0156] It should be understood that the time required for the target air conditioner to return to a stable operating state varies depending on the adjustment of different data items. Therefore, the preset second duration is not a fixed value, but a duration that can be flexibly adjusted according to the actual situation, such as 10 minutes. In some other embodiments, the preset second duration can also be set to a fixed value, which is the longest time required for the target air conditioner to return to a stable operating state.

[0157] (3) Delete the abnormal data in the initial data to obtain the target data.

[0158] In this embodiment of the application, in order to improve the stability of energy consumption analysis of target data, the initial data is preprocessed to delete abnormal data that is not suitable for energy consumption anomaly analysis, so as to obtain target data with high stability.

[0159] It should also be understood that in the process of training the baseline energy consumption prediction model and the real-time energy consumption prediction model, it is also necessary to perform corresponding preprocessing on the historical operating data used for training in order to improve the accuracy of the trained baseline energy consumption prediction model and the real-time energy consumption prediction model.

[0160] Furthermore, before determining the target operating condition corresponding to the target data in step 330, it is necessary to determine all possible operating conditions that the target air conditioner may exhibit during operation using historical operating data. This first requires preprocessing the historical operating data, i.e., steady-state cleaning. Data in the historical operating data that falls within a preset second time range after adjustment of any one or more data items related to preset temperature, preset fan speed, operating mode, and start-up load rate are removed, resulting in preprocessed historical operating data.

[0161] Furthermore, based on the unsupervised clustering analysis of the target data related items in the preprocessed historical operating data, multiple operating conditions and cluster centers for each operating condition are obtained. The cluster centers represent the typical characteristics of the operating condition.

[0162] Finally, after determining the operating conditions and their cluster centers, the cluster center closest to the target data can be calculated based on Euclidean distance to obtain the target cluster center corresponding to the target data, and the operating conditions corresponding to the target cluster center, i.e., the target operating conditions corresponding to the target data.

[0163] In this embodiment, steady-state cleaning is performed on both historical operating data and target data, which helps improve the stability of the energy consumption monitoring method at different stages, thereby enhancing overall stability. In other words, steady-state cleaning ensures that the energy consumption monitoring process is more reliable and efficient.

[0164] When the target air conditioner exhibits abnormal energy consumption, the target operating condition corresponding to the target data must be determined. Figure 6The following is a flowchart illustrating another energy consumption monitoring method provided in an embodiment of this application, as shown below. Figure 6 As shown above, Figure 3 Step 340, under the target operating conditions, determines the changing trend of the target monitoring parameters, including the following steps:

[0165] S341. Under the target operating conditions, perform kernel density estimation on the historical monitoring parameters to obtain the target parameter value and the mean value corresponding to the point where the historical monitoring parameter density is the maximum.

[0166] Among them, historical monitoring parameters are the monitoring parameters corresponding to historical operating data (i.e., data related to monitoring items), such as compressor frequency, condenser outlet subcooling, indoor unit electronic expansion valve opening, outdoor unit electronic expansion valve opening, compressor discharge pressure, evaporation temperature, condensation temperature, compressor suction pressure, air outlet temperature, compressor discharge superheat, difference between condenser outlet temperature and ambient temperature, condenser outlet temperature, operating current, etc.

[0167] It should be understood that the data items covered by historical monitoring parameters and target monitoring parameters are consistent. Historical monitoring parameters refer to the operating status parameters corresponding to the historical operating data of the target air conditioner during a preset third operating period. Target monitoring parameters, on the other hand, are the operating status data obtained from real-time monitoring of the target air conditioner while it is running. In other words, monitoring parameters existing in historical monitoring parameters can also have corresponding monitoring parameters in target monitoring parameters. For example, monitoring parameter A existing in historical monitoring parameters also has a corresponding monitoring parameter A in target monitoring parameters.

[0168] Kernel density is a nonparametric estimation method for estimating probability density functions. It constructs a smooth density estimate function by placing a kernel function at each data point and then summing these kernel functions in a weighted manner. The basic idea is to infer the probability density of the entire data distribution using known data points.

[0169] It should be understood that under the target operating condition, kernel density estimation is performed on each monitoring parameter from the historical monitoring parameters. After obtaining the kernel density estimation result of the monitoring parameter, the parameter value at the point of maximum density under the target operating condition can be determined by analyzing the density function, that is, finding the parameter value corresponding to the peak point of the probability density function (the target parameter value). Furthermore, under the target operating condition, the average value of the monitoring parameters is the parameter mean.

[0170] Figure 7This illustration shows a schematic diagram of kernel density estimation of historical monitoring parameters provided in an embodiment of this application. It is a schematic diagram of the kernel density estimation result of monitoring parameter A in the historical monitoring parameters under the target operating condition. Figure 7 As shown, under the target operating conditions, the target parameter value of monitoring parameter A is 0.36, and the average parameter value is 0.78.

[0171] S342. When the target density is less than the preset density threshold, compare the target value with the target parameter value and the average parameter value.

[0172] Among them, the target density is the density value obtained by kernel density estimation of the target value corresponding to the target monitoring parameter.

[0173] The target monitoring parameter is the monitoring parameter corresponding to the historical monitoring parameter. For example, if the historical monitoring parameter is the compressor frequency, then the target monitoring parameter is also the compressor frequency. The target value corresponding to the target monitoring parameter is the value of that target monitoring parameter.

[0174] It should be understood that for a target value, the corresponding target density can be determined based on kernel density estimation.

[0175] Figure 8 This illustration shows a schematic diagram of kernel density estimation of historical monitoring parameters provided in an embodiment of this application, such as... Figure 8 As shown, with Figure 7 The historical monitoring parameters are the same, and the value corresponding to monitoring parameter A in the target monitoring parameters is also presented. The corresponding target density can be determined by the value of 3.5 corresponding to monitoring parameter A.

[0176] The preset density threshold is used to analyze the target density and then determine whether there is a trend of change in the target monitoring parameters. If the target density is less than the preset density threshold, it is determined that there is a trend of change in the target monitoring parameters; if the target density is greater than or equal to the preset density threshold, it is determined that there is no trend of change in the target monitoring parameters; for example, the preset density threshold is 0.01, 0.05, etc.

[0177] If the target density is less than the preset density threshold, and the target monitoring parameters show a changing trend, the target value can be compared with the target parameter value and the average value of the parameters.

[0178] S343. If the target value is greater than the target parameter value, and the target value is greater than the average parameter value, then it is determined that the target monitoring parameter has an upward trend.

[0179] like Figure 8 As shown, the target data is 3.5, the target parameter value is 0.36, and the average parameter value is 0.78. In other words, the target value of 3.5 is greater than the target parameter value of 0.36, and the target value of 3.5 is greater than the average parameter value of 0.78. Therefore, the monitored parameter shows an upward trend.

[0180] S344. If the target value is less than the target parameter value and the target value is less than the average parameter value, then it is determined that the target monitoring parameter has an abnormal downward trend.

[0181] In some embodiments, if the target value is compared with the target parameter value and the parameter mean, and it does not belong to the case of step 344 or the case of step 345, then the target value is normal data.

[0182] By analyzing the target operating conditions described above, we can determine the specific type of abnormal change trend of the target monitoring parameters, such as an upward or downward trend. Then, we can compare and analyze the change trend of the target monitoring parameters with the reference parameters in the preset fault parameter change table to determine the specific target fault type corresponding to the energy consumption abnormality.

[0183] The changing trends of the target monitoring parameters are compared and analyzed with the reference parameters in the preset fault parameter change table. If the changing trend of the target monitoring parameters meets the preset conditions, the target fault type causing the energy consumption anomaly is determined. If the changing trend of the target monitoring parameters does not meet the preset conditions, the corresponding data information is presented to the air conditioning system operation and maintenance personnel to facilitate timely handling of energy consumption anomalies.

[0184] The reference parameters in the preset fault parameter change table include key parameters and auxiliary parameters. Key parameters play a core and decisive role in the target equipment's operating status and fault judgment. Auxiliary parameters, while not as decisive as key parameters, can provide supplementary information to help improve the analysis and judgment of faults.

[0185] Taking a multi-split air conditioner as an example, refer to Table 1, Table 2, or Table 3. Table 1, for instance, shows the parameter changes for refrigerant overcharge faults. In cooling mode, the key parameters are compressor frequency, condenser outlet subcooling, compressor discharge pressure, and the temperature difference between the condenser outlet and ambient temperatures. Auxiliary parameters include the opening of the indoor unit's electronic expansion valve, the opening of the outdoor unit's electronic expansion valve, evaporation temperature, condensation temperature, compressor suction pressure, outlet air temperature, compressor discharge superheat, condenser outlet temperature, and operating current. In heating mode, the key parameters are compressor frequency, condenser outlet subcooling, outlet air temperature, and compressor discharge pressure. Auxiliary parameters include the opening of the indoor unit's electronic expansion valve, the opening of the outdoor unit's electronic expansion valve, evaporation temperature, condensation temperature, compressor suction pressure, compressor discharge superheat, the temperature difference between the condenser outlet and ambient temperatures, the condenser outlet temperature, and operating current.

[0186] It should be understood that the tables have limited capacity, therefore the specific details of the auxiliary parameters are not presented in Tables 1, 2 and 3 above.

[0187] For step 350 above, the changing trend of the target monitoring parameters is compared and analyzed with the reference parameters in the preset fault parameter change table. If the changing trend of the target monitoring parameters meets the preset conditions, the target fault type causing abnormal energy consumption is determined. The preset conditions include at least one of the following conditions (1) to (3):

[0188] Condition (1): The changing trend of the first parameter in the target monitoring parameters is the same as the changing trend of the key parameters.

[0189] Among them, the first parameter is the parameter that corresponds to the key parameter in the target monitoring parameters. For example, the target monitoring parameter that corresponds to the compressor frequency and condenser outlet subcooling in Table 1 is the first parameter.

[0190] Condition (2): The changing trend of the second parameter in the target monitoring parameters is the same as the changing trend of the third parameter in the auxiliary parameters.

[0191] Among them, the ratio of the number of third parameters to the number of auxiliary parameters is greater than the preset first ratio. For example, if the preset first ratio is 50%, then the number of third parameters accounts for half or more of the number of auxiliary parameters.

[0192] The second parameter is the parameter in the target monitoring parameters that corresponds to the third parameter. It should be understood that the first and second parameters are different parameters within the target monitoring parameters. The target monitoring parameters may also include other parameters besides the first and second parameters.

[0193] Condition (3): The ratio of the number of times the change trend of the first parameter is the same as the change trend of the key parameter and the change trend of the second parameter is the same as the change trend of the third parameter in the preset third time period to the number of monitoring times in the preset third time period exceeds the preset second ratio.

[0194] The preset third duration can be one day or two days.

[0195] If, during the preset third operating period, the ratio of the number of times that simultaneously meet conditions (1) and (2) to the number of monitoring times during the preset third operating period exceeds a preset second proportion (e.g., the preset second proportion is 70%), then the target fault type causing the energy consumption anomaly can be identified. Analysis is then conducted on multiple monitoring events occurring within the preset third operating period. If, among these monitoring events, a certain proportion of the monitoring events simultaneously meet conditions (1) and (2) to the number of monitoring times during the preset third operating period, then the target fault type causing the energy consumption anomaly within that preset third operating period can be determined. This process can improve the accuracy and comprehensiveness of fault attribution analysis under energy consumption anomalies.

[0196] In some embodiments, the preset conditions include the above conditions (1), (2) and (3), that is, when the changing trend of the target monitoring parameters simultaneously meets conditions (1), (2) and (3), the target fault type that causes abnormal energy consumption is determined.

[0197] For example, when there is an energy consumption anomaly in the target data, the target monitoring parameters corresponding to the target data include compressor frequency, condenser outlet subcooling, compressor discharge pressure, difference between condenser outlet temperature and ambient temperature, opening of indoor unit electronic expansion valve, opening of outdoor unit electronic expansion valve, evaporation temperature, condensation temperature, compressor suction pressure, and outlet air temperature.

[0198] In step 340, under the target operating condition (cooling mode), the changing trend of the target monitoring parameters is determined, as shown in Table 4 below:

[0199] Table 4. Trends in Target Monitoring Parameters

[0200] compressor frequency √ Condenser outlet subcooling √ Compressor discharge pressure √ Condenser outlet temperature difference with ambient temperature √ Indoor unit electronic expansion valve opening √ Compressor current √ Evaporation temperature √ Condensation temperature √ compressor suction pressure √ Air outlet temperature √

[0201] By comparing the changing trends of the target monitoring parameters in Table 4 with the reference parameters in the preset fault parameter change table, it can be seen that the changing trends of the target monitoring parameters in Table 4 and the parameter change table of refrigerant overcharge fault in Table 1 satisfy conditions (1), (2) and (3) simultaneously.

[0202] In other words, the changing trend of the first parameter in the target monitoring parameters (compressor frequency, condenser outlet subcooling, compressor discharge pressure, and the difference between condenser outlet temperature and ambient temperature) is the same as the changing trend of the key parameters in Table 1.

[0203] The changing trend of the second parameter (opening of the indoor unit electronic expansion valve, evaporation temperature, compressor suction pressure, and outlet air temperature) in the target monitoring parameters is the same as that of the third parameter (opening of the indoor unit electronic expansion valve, evaporation temperature, compressor suction pressure, and outlet air temperature) in the auxiliary parameters, and the second parameter accounts for more than 50%.

[0204] Furthermore, over the two days, the trend of change of the first parameter was the same as that of the key parameter, and the trend of change of the second parameter was the same as that of the third parameter 75 times, which is more than 70% of the total number of monitoring times of 100 over the two days (0.75).

[0205] In the example above, when there is an energy consumption anomaly in the target data, the target fault type causing the energy consumption anomaly is determined to be a refrigerant overcharge fault.

[0206] Figure 9 The following is a flowchart illustrating another energy consumption monitoring method provided in an embodiment of this application, as shown below. Figure 9 As shown, before the energy consumption monitoring method is officially used: firstly, the baseline energy consumption prediction model and the real-time energy consumption prediction model are trained by different data in the historical operation data, and the historical monitoring parameters in the historical operation data are analyzed to obtain all operating conditions corresponding to the historical operation data.

[0207] Furthermore, it acquires and stores the target data corresponding to the target air conditioner in real time.

[0208] The target data is input into the trained baseline energy consumption prediction model to obtain the corresponding baseline energy consumption value. The target data and target monitoring parameters are input into the trained real-time energy consumption prediction model to obtain the corresponding actual energy consumption value. The baseline energy consumption value and the actual energy consumption value are analyzed to determine whether there is an energy consumption anomaly.

[0209] It should be understood that, assuming the electricity meter is not malfunctioning, the actual energy consumption value can also be obtained through the electricity meter.

[0210] Finally, in the case of abnormal energy consumption, the target operating condition corresponding to the target data is determined, and under the target operating condition, the kernel density estimation analysis of historical monitoring parameters is performed. Then, in combination with the preset fault parameter change table, the cause of the abnormal energy consumption, i.e. the target fault type, is determined.

[0211] It should be understood that the sequence numbers of the processes in the above embodiments do not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention. The various embodiments described herein can be independent solutions or combinations based on internal logic, and all such solutions fall within the protection scope of this application.

[0212] It should also be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowchart may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0213] Figure 10 A schematic diagram of an energy consumption monitoring device provided in an embodiment of this application is shown. This energy consumption monitoring device 400 can be used to perform the actions executed in the method embodiments described above. For example... Figure 10As shown, the energy consumption monitoring device 400 includes a data acquisition unit 410, an energy consumption analysis unit 420, and an anomaly attribution processing unit 430.

[0214] The data acquisition unit 410 is used to acquire target data corresponding to the target air conditioner;

[0215] The energy consumption analysis unit 420 is used to perform energy consumption analysis on the target data of the target air conditioner to determine whether there is any energy consumption anomaly in the target air conditioner.

[0216] The anomaly attribution processing unit 430 is used to determine the target operating condition corresponding to the target data when the target air conditioner has an energy consumption anomaly, and to determine the changing trend of the target monitoring parameters under the target operating condition, wherein the target monitoring parameters are the operating status parameters corresponding to the target data.

[0217] The anomaly attribution processing unit 430 is also used to compare and analyze the changing trend of the target monitoring parameters with the reference parameters in the preset fault parameter changing table. When the changing trend of the target monitoring parameters meets the preset conditions, the target fault type that causes the energy consumption anomaly is determined. The preset conditions are used to indicate that there is a corresponding relationship between the changing trend of the target monitoring parameters and the reference parameters in the preset fault parameter changing table.

[0218] The energy consumption monitoring device provided in this application embodiment is used to execute the energy consumption monitoring method of the above embodiments. Its technical principle and technical effects are similar, and will not be repeated here. It should be understood that this energy consumption monitoring device can be a terminal device or a chip within the terminal device.

[0219] This application provides a computer program product that, when run on an electronic device, causes the electronic device to execute the technical solutions described in the above embodiments. Its implementation principle and technical effects are similar to those of the related embodiments described above, and will not be repeated here.

[0220] This application provides a readable storage medium containing instructions that, when executed by an electronic device, cause the electronic device to perform the technical solutions described in the above embodiments. The implementation principle and technical effects are similar and will not be repeated here.

[0221] The computer-readable storage medium and computer program product are used to execute the corresponding methods provided above. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects of the corresponding methods provided above, and will not be repeated here.

[0222] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0223] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0224] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for example, the division of units is merely a logical functional division, and there may be other division methods in actual implementation; for example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, and the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0225] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0226] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0227] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. An energy consumption monitoring method, characterized in that, The system is applied to an air conditioning system, which includes a monitoring system and air conditioning equipment. The air conditioning equipment includes one or more target air conditioners, and the monitoring system is communicatively connected to the air conditioning equipment. The monitoring system is configured to execute the energy consumption monitoring method, including: Obtain target data corresponding to the target air conditioner, wherein the target data is used to indicate the operating status of the target air conditioner; Based on the benchmark energy consumption prediction model, energy consumption analysis is performed on the target data of the target air conditioner to determine whether the target air conditioner has any energy consumption anomalies. When the target air conditioner has abnormal energy consumption, the target operating condition corresponding to the target data is determined, and under the target operating condition, the changing trend of the target monitoring parameters is determined, wherein the target monitoring parameters are the operating status parameters corresponding to the target data; The changing trend of the target monitoring parameter is compared and analyzed with the reference parameter in the preset fault parameter changing table. If the changing trend of the target monitoring parameter meets the preset conditions, the target fault type causing the abnormal energy consumption is determined. The preset conditions are used to indicate that there is a corresponding relationship between the changing trend of the target monitoring parameter and the reference parameter in the preset fault parameter changing table.

2. The energy consumption monitoring method according to claim 1, characterized in that, The step of performing energy consumption analysis on the target data of the target air conditioner to determine whether the target air conditioner has any energy consumption anomalies includes: The target data is input into the benchmark energy consumption prediction model to obtain the benchmark energy consumption value corresponding to the target air conditioner, and the first actual energy consumption value of the target air conditioner is obtained through the electricity meter corresponding to the target air conditioner. A deviation analysis is performed between the baseline energy consumption value and the first actual energy consumption value to obtain a first deviation degree value; If there are M instances of energy consumption deviation within a preset first operating time of the target air conditioner and the ratio M / N is greater than a preset abnormality ratio, it is determined that the target air conditioner has an energy consumption abnormality. Here, if the first deviation degree value is greater than a preset deviation threshold, it indicates that the corresponding target data has an energy consumption deviation, and N is the number of detections within the preset first operating time.

3. The energy consumption monitoring method according to claim 1, characterized in that, The step of performing energy consumption analysis on the target data of the target air conditioner to determine whether the target air conditioner has any energy consumption anomalies includes: The target data is input into the benchmark energy consumption prediction model to obtain the benchmark energy consumption value corresponding to the target air conditioner; The target data and the target monitoring parameters are input into the real-time energy consumption prediction model to obtain the second actual energy consumption value corresponding to the target air conditioner; A deviation analysis is performed between the baseline energy consumption value and the second actual energy consumption value to obtain a second deviation value; If there are M instances of energy consumption deviation within a preset first operating time of the target air conditioner and the ratio M / N is greater than a preset abnormality ratio, it is determined that the target air conditioner has an energy consumption abnormality. Here, if the second deviation degree value is greater than a preset deviation threshold, it indicates that the corresponding target data has an energy consumption deviation, and N is the number of detections within the preset first operating time.

4. The energy consumption monitoring method according to claim 3, characterized in that, The real-time energy consumption prediction model is trained using historical operating data within a preset third operating period of the target air conditioner, as well as a time-series model optimized during the operation of the target air conditioner.

5. The energy consumption monitoring method according to any one of claims 1 to 3, characterized in that, The baseline energy consumption prediction model is trained using historical operating data within a preset third operating period of the target air conditioner, as well as a regression model optimized during the operation of the target air conditioner.

6. The energy consumption monitoring method according to any one of claims 1 to 4, characterized in that, Under the target operating condition, determining the changing trend of the target monitoring parameters corresponding to the target data includes: Under the target operating condition, kernel density estimation is performed on historical monitoring parameters to obtain the target parameter value and the mean value corresponding to the maximum density of the historical monitoring parameters. The historical monitoring parameters are the operating status parameters corresponding to historical operating data, and the historical operating data are the operating data within a preset third time period when the target air conditioner has been running for a certain duration. When the target density is less than a preset density threshold, the target value is compared with the target parameter value and the mean value of the parameter, wherein the target density is the density value obtained by kernel density estimation of the target value corresponding to the target monitoring parameter; If the target value is greater than the target parameter value, and the target value is greater than the average value of the parameter, then it is determined that the target monitoring parameter has an upward trend. If the target value is less than the target parameter value, and the target value is less than the average value of the parameter, then it is determined that the target monitoring parameter has an abnormal downward trend.

7. The energy consumption monitoring method according to any one of claims 1 to 4, characterized in that, The reference parameters in the preset fault parameter change table include key parameters and auxiliary parameters, and the preset conditions include at least one of the following: The changing trend of the first parameter in the target monitoring parameters is the same as the changing trend of the key parameter, and the first parameter is the parameter in the target monitoring parameters that corresponds to the key parameter; The changing trend of the second parameter in the target monitoring parameters is the same as the changing trend of the third parameter in the auxiliary parameters. The second parameter is the parameter in the target monitoring parameters that corresponds to the third parameter. The ratio of the number of the third parameter to the number of the auxiliary parameters is greater than a preset first ratio. Within a preset third time period, the ratio of the number of times the change trend of the first parameter is the same as the change trend of the key parameter, and the change trend of the second parameter is the same as the change trend of the third parameter, to the number of monitoring times within the preset third time period exceeds a preset second ratio.

8. The energy consumption monitoring method according to claim 1, characterized in that, The air conditioning system also includes a communication module, which is used to acquire outdoor environmental data, including outdoor relative humidity and solar radiation intensity. The target data includes preset temperature, preset windshield, operating mode, start-up load rate, indoor temperature, outdoor temperature, outdoor relative humidity, and solar radiation intensity.

9. The energy consumption monitoring method according to claim 8, characterized in that, Before acquiring the target data corresponding to the target air conditioner, the following steps are included: Obtain the initial data corresponding to the target air conditioner; When the preset temperature, the preset windshield, the operating mode and / or the start-up load rate in the initial data are adjusted, abnormal data is determined. The abnormal data is the initial data obtained within a preset second time period from the time the adjustment is completed. The abnormal data in the initial data is deleted to obtain the target data.

10. An air conditioning system, characterized in that, It includes a monitoring system and air conditioning equipment, wherein the monitoring system is used to perform the energy consumption monitoring method as described in any one of claims 1 to 9.

Images