Air conditioning system low standby control method and device based on machine self-learning

Abstract

The invention relates to the field of air conditioners, and provides an air conditioner system low standby control method and device based on machine self-learning, and the method comprises the steps that the time interval between adjacent startup and shutdown within target time is obtained, and the time interval between the adjacent startup and shutdown comprises startup staying time and shutdown staying time; determining that the number of the shutdown residence time larger than a first preset threshold value in the target time is larger than a preset number, taking the latest shutdown time in the shutdown residence time correspondingly larger than the first preset threshold value as a to-be-switched time point of the deep low-power-consumption standby mode, and obtaining an outer environment temperature; according to the outer environment temperature, the wakeup countdown duration is determined, and the system is controlled to be switched into the deep low-energy-consumption standby mode on the basis that the wakeup countdown duration is 0 and the to-be-switched time point. According to the invention, the defect of standby energy waste at night of workdays and in holidays and festivals is overcome, the standby power of the outdoor unit of the air conditioning system is obviously reduced, and the energy-saving effect is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of air conditioners, and particularly to a low standby control method and device for an air conditioner system based on machine self-learning. Background Art

[0002] An air conditioner mainly includes a compressor, an indoor heat exchanger, a throttling component, and an outdoor heat exchanger that form a main refrigerant circuit. By circulating the refrigerant in the circuit formed by the compressor, condenser, throttling component, evaporator, and compressor, along with the phase change of the refrigerant, the temperature of the indoor space where the indoor heat exchanger is located can be adjusted.

[0003] In the existing air conditioner system, there are many regular application scenarios in actual use. For example, in an office scenario, it is usually only turned on during working hours on weekdays. However, during special time periods such as at night and on holidays, although the indoor unit is in the shutdown state, since the outdoor unit is still in the standby state, that is, there is still a certain power consumption in the compressor, electric heating, communication, drive module, etc., so considering the annual energy consumption, a large amount of power will still be consumed during the annual standby time, which not only causes an increase in electricity bills for users but also increases energy waste. Summary of the Invention

[0004] The present invention provides a low standby control method and device for an air conditioner system based on machine self-learning, aiming to solve the defect of standby energy waste during weekdays at night and holiday time periods in the prior art, significantly reducing the standby power of the outdoor unit of the air conditioner system and achieving an energy-saving effect.

[0005] The present invention provides a low standby control method for an air conditioner system based on machine self-learning, including: obtaining the time interval between adjacent power-on and power-off within a target time, where the time interval between adjacent power-on and power-off includes the power-on residence time and the power-off residence time; determining that the number of power-off residence times greater than a first preset threshold within the target time is greater than a preset number, taking the latest power-off time in the corresponding power-off residence times greater than the first preset threshold as the time point to be switched to the deep low-power standby mode, and obtaining the outer ring temperature; according to the outer ring temperature, determining the wake-up countdown duration, and when the wake-up countdown duration is 0, based on the time point to be switched, controlling the system to switch to the deep low-energy standby mode.

[0006] It should be noted that by determining that the number of shutdown residence times greater than the first preset threshold in the time intervals between adjacent power - on and power - off within the obtained target time is greater than the preset number, the working days and rest days are preliminarily distinguished. The latest shutdown time on working days is used as the time point to be switched, and based on the obtained outer - ring temperature, the corresponding wake - up countdown duration is determined. When the wake - up countdown duration is 0, it is directly determined that the control system switches to the deep low - power standby mode based on the time point to be switched, realizing precise control of entering and exiting the deep standby state, minimizing the standby power consumption to the greatest extent while meeting the user's usage requirements, avoiding standby energy waste during the night on working days and holiday periods, applicable to the air - conditioning system with local control, not restricted by the timed - on wake - up function, with higher applicability, stronger popularizability, and more precise and efficient energy - saving effect.

[0007] According to the low - standby control method for an air - conditioning system based on machine self - learning provided by the present invention, obtaining the time intervals between adjacent power - on and power - off within the target time includes: obtaining the air - conditioning operation records within the target time; using a pattern recognition model to recognize the air - conditioning operation records to obtain the corresponding time intervals between adjacent power - on and power - off; wherein, the pattern recognition model is pre - trained based on the air - conditioning historical operation records and the time - interval labels corresponding to the air - conditioning historical operation records.

[0008] It should be noted that by using the pattern recognition model to recognize and predict the air - conditioning operation records, while ensuring the accuracy of the predicted time intervals between adjacent power - on and power - off, it is convenient to realize precise control of entering and exiting the deep standby state based on the predicted time intervals between adjacent power - on and power - off.

[0009] According to the low - standby control method for an air - conditioning system based on machine self - learning provided by the present invention, the power - on residence time represents the duration from power - on to power - off; after determining the wake - up countdown duration according to the outer - ring temperature, it further includes: based on the wake - up countdown duration being non - zero, determining whether there is a power - on behavior within the preset time after the end of the wake - up countdown according to the time intervals between adjacent power - on and power - off within the target time and the power - on residence time; wherein, the wake - up countdown starts from the time point to be switched; if there is a power - on behavior within the preset time after the end of the wake - up countdown, the control system powers on at the corresponding power - on time and the control system is in the normal working mode.

[0010] It should be noted that by determining whether there is a power - on behavior within the preset time after the end of the wake - up countdown to determine the normal working days, the control system powers on and the control system is in the normal working mode, realizing precise control of the deep low - power standby state, ensuring that the standby power consumption is minimized to the greatest extent while meeting the user's usage requirements.

[0011] A low standby control method for an air conditioning system based on machine self-learning according to the present invention determines whether there is a startup behavior within a preset time after the wake-up countdown ends, including: if there is no startup behavior within the preset time after the wake-up countdown ends, the control system switches to a deep low-power standby mode based on the time point to be switched.

[0012] It should be noted that by determining that there is no startup behavior within the preset time after the wake-up countdown ends, it is determined as a rest day such as a weekend or a holiday, and thus based on the time point to be switched, the control system switches to a deep standby mode to minimize standby power consumption while meeting the user's usage requirements and avoid standby energy waste during weekdays at night and holiday periods.

[0013] A low standby control method for an air conditioning system based on machine self-learning according to the present invention, after obtaining the time interval between adjacent power-on and power-off within the target time, further includes: determining whether the number of shutdown stay times greater than a first preset threshold within the target time is greater than a preset number; based on the number of shutdown stay times greater than the first preset threshold within the target time being less than or equal to the preset number, updating the target time and re-obtaining the time interval between adjacent power-on and power-off within the updated target time; according to the re-obtained time interval between adjacent power-on and power-off within the updated target time, re-determining whether the number of shutdown stay times greater than the first preset threshold within the updated target time is greater than the preset number.

[0014] It should be noted that by determining whether the number of shutdown stay times greater than the first preset threshold within the target time is greater than the preset number, it is determined whether to use the latest shutdown time in the corresponding shutdown stay times greater than the first preset threshold as the time point to be switched to the deep low-power standby mode, or to update the target time to re-obtain the time interval between adjacent power-on and power-off within the updated target time, so as to preliminarily screen weekdays and rest days, thereby facilitating using the latest shutdown time on weekdays as the time point to be switched to the deep low-power standby mode, and determining whether to switch to the deep low-power standby mode at this time point based on the outdoor ambient temperature, so as to achieve precise control of the deep low-power standby mode, minimize standby power consumption while meeting the user's usage requirements, and avoid standby energy waste during weekdays at night and holiday periods.

[0015] A low standby control method for an air conditioning system based on machine self-learning according to the present invention determines the wake-up countdown duration according to the outdoor ambient temperature, including: if the outdoor ambient temperature is greater than a first preset temperature, determining that the wake-up countdown duration is 0; otherwise, determining the corresponding wake-up countdown duration according to the temperature range where the outdoor ambient temperature is located in combination with the wake-up setting rule; wherein, the wake-up setting rule is used to define the wake-up countdown duration corresponding to the outdoor ambient temperature in different temperature ranges.

[0016] It should be noted that by determining whether the ambient temperature of the outer ring is greater than the first preset temperature, it is convenient to initially determine whether the wake-up countdown is set to 0, so as to facilitate subsequent switching to the deep low-power standby mode directly based on the set wake-up countdown duration at the time point to be switched, or further determining whether it can be switched to the deep low-power standby mode according to the non-zero wake-up countdown duration, and further precisely controlling the entry and exit of the deep low-power standby mode.

[0017] According to a low standby control method for an air-conditioning system based on machine self-learning provided by the present invention, the wake-up setting rule includes: if the ambient temperature of the outer ring is greater than the second preset temperature and less than or equal to the first preset temperature, it is determined that the wake-up countdown duration is the first preset duration; if the ambient temperature of the outer ring is greater than the third preset temperature and less than or equal to the second preset temperature, it is determined that the wake-up countdown duration is the second preset duration; if the ambient temperature of the outer ring is less than or equal to the third preset temperature, it is determined that the wake-up countdown duration is the third preset duration; wherein, the first preset temperature is greater than the second preset temperature, the second preset temperature is greater than the third preset temperature, the first preset duration is greater than 0 and less than the second preset duration, the second preset duration is less than the third preset duration, and the wake-up countdown duration is set in units of days and in 24-hour timekeeping.

[0018] It is worth noting that by setting different wake-up countdown durations according to the temperature range of the ambient temperature of the outer ring, it is convenient to precisely control the switching of the deep low-power mode in combination with user habits and different regional climates, and improve the energy-saving effect while meeting user needs.

[0019] The present invention also provides a low standby control device for an air-conditioning system based on machine self-learning, including: a data acquisition module, which acquires the time interval between adjacent power-on and power-off within the target time, and the time interval between adjacent power-on and power-off includes the power-on stay time and the power-off stay time; a judgment module, which determines that the number of power-off stay times greater than the first preset threshold within the target time is greater than the preset number, takes the latest power-off time in the corresponding power-off stay times greater than the first preset threshold as the time point to be switched to the deep low-power standby mode, and acquires the ambient temperature of the outer ring; a control module, which determines the wake-up countdown duration according to the ambient temperature of the outer ring, and based on the wake-up countdown duration being 0, controls the system to switch to the deep low-power standby mode based on the time point to be switched.

[0020] It should be noted that by determining that the number of shutdown stay times greater than the first preset threshold in the time interval between adjacent power - on and power - off within the target time obtained by the data acquisition module through the judgment module is greater than the preset number, the working days and rest days are preliminarily distinguished. The latest shutdown time on working days is used as the time point to be switched. And based on the obtained outer - ring temperature through the control module, the corresponding wake - up countdown duration is determined. When the wake - up countdown duration is 0, it is directly determined that the control system switches to the deep low - power standby mode based on the time point to be switched, realizing precise control of entering and exiting deep standby. While meeting the user's usage requirements, the standby power consumption is minimized to the greatest extent, avoiding standby energy waste during the night on working days and holiday periods. It is applicable to local - controlled air - conditioning systems, without being restricted by the timed power - on wake - up function, with higher applicability and stronger popularizability, and more precise and efficient energy - saving effects.

[0021] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, the steps of the low - standby control method for an air - conditioning system based on machine self - learning as described in any one of the above are implemented.

[0022] The present invention also provides a non - transitory computer - readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the low - standby control method for an air - conditioning system based on machine self - learning as described in any one of the above are implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0024] Figure 1 is one of the flow charts of the low - standby control method for an air - conditioning system based on machine self - learning provided by the present invention; Figure 2 is another flow chart of the low - standby control method for an air - conditioning system based on machine self - learning provided by the present invention; Figure 3 is the flow chart of the deep low - power standby control method for an air - conditioning system on working days provided by the present invention; Figure 4 is the flow chart of the deep low - power standby control method for an air - conditioning system on rest days provided by the present invention; Figure 5 is the structural diagram of the low - standby control device for an air - conditioning system based on machine self - learning provided by the present invention; Figure 6 It is a schematic structural diagram of the electronic device provided by the present invention. Specific embodiments

[0025] To make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below with reference to the accompanying drawings in the present invention. Apparently, the described embodiments are some but not all of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present invention without making creative efforts shall fall within the scope of protection of the present invention.

[0026] Figure 1 A flowchart showing a low standby control method for an air-conditioning system based on machine self-learning according to the present invention is shown. The method includes: S11, obtaining the time interval between adjacent power-on and power-off operations within a target time. The time interval between adjacent power-on and power-off operations includes the power-on stay time and the power-off stay time; S12, determining that the number of power-off stay times greater than a first preset threshold within the target time is greater than a preset number, taking the latest power-off time among the corresponding power-off stay times greater than the first preset threshold as the time point to be switched to the deep low-power standby mode, and obtaining the outside ring temperature; S13, determining the wake-up countdown duration according to the outside ring temperature, and based on the wake-up countdown duration being 0, and based on the time point to be switched, controlling the system to switch to the deep low-energy standby mode.

[0027] It should be noted that the step numbers "S1N" in this specification do not represent the sequence of the low standby control method for the air-conditioning system based on machine self-learning. The following specifically combines Figures 2 - 4 to describe the low standby control method for the air-conditioning system based on machine self-learning of the present invention.

[0028] In step S11, the time interval between adjacent power-on and power-off operations within the target time is obtained. The time interval between adjacent power-on and power-off operations includes the power-on stay time and the power-off stay time.

[0029] In this embodiment, obtaining the time interval between adjacent power-on and power-off operations within the target time includes: obtaining the air-conditioning operation records within the target time; using a pattern recognition model to recognize the air-conditioning operation records to obtain the corresponding time interval between adjacent power-on and power-off operations; wherein, the pattern recognition model is previously trained based on the air-conditioning historical operation records and the time interval labels corresponding to the air-conditioning historical operation records.

[0030] It should be noted that by using the pattern recognition model to recognize and predict the operation records of the air conditioner, while ensuring the time interval accuracy between adjacent predicted power - on and power - off operations, it facilitates subsequent precise control of entering and exiting deep standby based on the time interval between predicted adjacent power - on and power - off operations.

[0031] Specifically, obtaining the operation records of the air conditioner within the target time includes: obtaining the operation records of the air conditioner within the target time based on a preset time period. It should be added that the preset time period can be designed according to the actual design cycle, such as one week, one month, etc. The target time is different time periods determined based on the preset time period, and no further limitation is made here; the operation records of the air conditioner are all the operation records within the design cycle.

[0032] In addition, the pattern recognition model includes: a feature extraction layer that extracts features from the input operation records of the air conditioner to obtain power - on and power - off features; a feature classification layer that classifies the extracted power - on and power - off features to obtain power - on features and power - off features; a pattern recognition layer that obtains the power - on stay time and the power - off stay time based on the power - on features and the power - off features. Among them, the power - on stay time is obtained based on the previous power - on feature and the adjacent subsequent power - off feature, and the power - off stay time is obtained based on the previous power - off feature and the adjacent subsequent power - on feature.

[0033] Correspondingly, using the pattern recognition model to recognize the operation records of the air conditioner to obtain the time interval between corresponding adjacent power - on and power - off operations includes: using the feature extraction layer to extract features from the input operation records of the air conditioner to obtain power - on and power - off features; using the feature classification layer to classify the extracted power - on and power - off features to obtain power - on features and power - off features; using the pattern recognition layer to obtain the power - on stay time and the power - off stay time based on the power - on features and the power - off features.

[0034] In an alternative embodiment, before using the pattern recognition model to recognize the operation records of the air conditioner to obtain the time interval between corresponding adjacent power - on and power - off operations, it includes: training the pattern recognition model. Specifically, training the pattern recognition model includes: obtaining the historical operation records of the air conditioner and the time interval labels corresponding to the historical operation records of the air conditioner; using the historical operation records of the air conditioner as the input data for training, and using the time interval labels corresponding to the historical operation records of the air conditioner as the labels for training, and training the model to be trained to obtain a pattern recognition model for recognizing and predicting the time interval between adjacent power - on and power - off operations.

[0035] It should be noted that an existing network can be built into the model to be trained. The existing network usually includes a network structure or other networks specified by the user, such as neural network algorithms. The model to be trained usually includes a feature extraction layer for extracting corresponding power-on and power-off features respectively, a feature classification layer for classifying the extracted power-on and power-off features, a pattern recognition layer for recognizing patterns based on the classified power-on features and power-off features, and a loss function. According to the preset iteration rules, the above-mentioned air conditioner historical operation records are input into the model to be trained for training, and a trained pattern recognition model is obtained.

[0036] In an alternative embodiment, after obtaining the time intervals between adjacent power-on and power-off operations within the target time, it further includes: determining whether the number of power-off stay times greater than a first preset threshold within the target time is greater than a preset number; based on the fact that the number of power-off stay times greater than the first preset threshold within the target time is less than or equal to the preset number, updating the target time and re-obtaining the time intervals between adjacent power-on and power-off operations within the updated target time; according to the re-obtained time intervals between adjacent power-on and power-off operations within the updated target time, re-determining whether the number of power-off stay times greater than the first preset threshold within the updated target time is greater than the preset number.

[0037] It should be noted that since the target time is different time periods determined based on a preset time cycle, the target time needs to be updated. By determining whether the number of power-off stay times greater than the first preset threshold within the target time is greater than the preset number, it is determined whether to use the latest power-off time in the corresponding power-off stay times greater than the first preset threshold as the time point to be switched to the deep low-power standby mode, or to update the target time to re-obtain the time intervals between adjacent power-on and power-off operations within the updated target time, so as to preliminarily screen weekdays and rest days, thereby facilitating using the latest power-off time on weekdays as the time point to be switched to the deep low-power standby mode, and determining whether to switch to the deep low-power standby mode at this time point based on the outdoor ambient temperature, so as to achieve precise control of the deep low-power standby mode, minimize standby power consumption to the greatest extent while meeting the user's usage requirements, and avoid standby energy waste during weekdays at night and holiday periods.

[0038] In addition, the preset number can be set based on the number of weekdays, such as set to 4, 5, etc., and no further limitation is made here; the first preset threshold can be set based on the user's actual air conditioner usage situation, such as any value within [8, 16h] is selected, and no further limitation is made here.

[0039] Step S12, when it is determined that the number of power-off stay times greater than the first preset threshold within the target time is greater than the preset number, use the latest power-off time in the corresponding power-off stay times greater than the first preset threshold as the time point to be switched to the deep low-power standby mode, and obtain the outdoor ambient temperature.

[0040] In this embodiment, obtaining the ambient temperature outside the ring includes: obtaining the value of the ambient temperature sensor outside the ring within the target time range, where the lower limit value of the target time range is the time point to be switched; obtaining the ambient temperature outside the ring according to the average value of the values of the ambient temperature sensor outside the ring within the target time range.

[0041] It should be noted that the target time range can be set based on actual design requirements. For example, the upper limit value is the time point 24 hours before the time point to be switched, and the lower limit value is the time point to be switched. No further limitation is made here. In addition, through the values of the ambient temperature sensor outside the ring within the target time range, it is convenient to determine the average value of the ambient temperature outside the ring within the target time range and use it as the obtained ambient temperature outside the ring, so as to facilitate setting the corresponding wake-up countdown duration according to the temperature range interval where the ambient temperature outside the ring is located.

[0042] Step S13: Determine the wake-up countdown duration according to the ambient temperature outside the ring, and based on the wake-up countdown duration being 0 and the time point to be switched, control the system to switch to the deep low-power standby mode.

[0043] In this embodiment, determining the wake-up countdown duration according to the ambient temperature outside the ring includes: if the ambient temperature outside the ring is greater than the first preset temperature, determining that the wake-up countdown duration is 0; otherwise, determining the corresponding wake-up countdown duration according to the temperature range where the ambient temperature outside the ring is located in combination with the wake-up setting rule; where the wake-up setting rule is used to limit the wake-up countdown duration corresponding to the ambient temperature outside the ring in different temperature ranges.

[0044] It should be noted that by judging whether the ambient temperature outside the ring is greater than the first preset temperature, it is convenient to initially determine whether the wake-up countdown is set to 0, so as to facilitate subsequently directly switching to the deep low-power standby mode based on the wake-up countdown duration set to 0 at the time point to be switched, or further determining whether it can be switched to the deep low-power standby mode according to the non-zero wake-up countdown duration, and further precisely controlling the entry and exit of the deep low-power standby.

[0045] Furthermore, the wake-up setting rule includes: if the ambient temperature outside the ring is greater than the second preset temperature and less than or equal to the first preset temperature, determining that the wake-up countdown duration is the first preset duration; if the ambient temperature outside the ring is greater than the third preset temperature and less than or equal to the second preset temperature, determining that the wake-up countdown duration is the second preset duration; if the ambient temperature outside the ring is less than or equal to the third preset temperature, determining that the wake-up countdown duration is the third preset duration; where the first preset temperature is greater than the second preset temperature, the second preset temperature is greater than the third preset temperature, the first preset duration is greater than 0 and less than the second preset duration, the second preset duration is less than the third preset duration, and the wake-up countdown duration is set in units of days and in 24-hour timekeeping.

[0046] It should be noted that different wake-up countdown durations are set according to the temperature range of the outer ring temperature, so as to precisely control the switching of the deep low-power mode in combination with user habits and different regional climates, improving the energy-saving effect while meeting user needs.

[0047] In addition, the first preset temperature, the second preset temperature, and the third preset temperature can be set according to the user's regional climate and the corresponding outer ring temperature. For example, the first preset temperature is set to 20°C, the second preset temperature is set to 7°C, and the third preset temperature is set to -10°C. No further limitation is made here; the first preset duration, the second preset duration, and the third preset duration can be set according to the user's usage habits. For example, the first preset duration is set to 120 min, the second preset duration is set to 180 min, and the third preset duration is set to 240 min. No further limitation is made here.

[0048] In an alternative embodiment, referring to Figure 3 , the power-on stay time represents the duration from power-on to power-off; after determining the wake-up countdown duration according to the outer ring temperature, it further includes: based on the wake-up countdown duration being non-zero, determining whether there is a power-on behavior within a preset time after the end of the wake-up countdown according to the time interval between adjacent power-on and power-off within the target time; where the wake-up countdown is carried out from the time point to be switched according to the wake-up countdown; if there is a power-on behavior within the preset time after the end of the wake-up countdown, the system is powered on at the corresponding power-on time and the system is controlled to be in the normal working mode.

[0049] It should be added that the preset time can be set according to actual usage habits and design requirements. For example, it can be 2 h. No further limitation is made here.

[0050] It should be noted that by determining that there is a power-on behavior within the preset time after the end of the wake-up countdown to determine it as a normal working day, the system is thus powered on and the system is controlled to be in the normal working mode, realizing precise control of deep low-energy standby, ensuring to minimize standby power consumption while meeting user usage needs.

[0051] In addition, determining whether there is a power-on behavior within the preset time after the end of the wake-up countdown includes: if there is no power-on behavior within the preset time after the end of the wake-up countdown, the system is controlled to switch to the deep low-energy standby mode based on the time point to be switched.

[0052] It should be noted that by determining that there is no power-on behavior within the preset time after the end of the wake-up countdown to determine it as a rest day such as a weekend or a holiday, the system is thus switched to the deep standby mode based on the time point to be switched, so as to minimize standby power consumption while meeting user usage needs and avoiding standby energy waste during the night on weekdays and holiday periods.

[0053] In summary, in the embodiment of the present invention, by determining that the number of shutdown stay times greater than the first preset threshold in the time intervals between adjacent power - on and power - off within the obtained target time is greater than the preset number, the working days and rest days are preliminarily distinguished. The latest shutdown time on working days is used as the time point to be switched, and based on the obtained outdoor ambient temperature, the corresponding wake - up countdown duration is determined. When the wake - up countdown duration is 0, it is directly determined that the control system switches to the deep low - power standby mode based on the time point to be switched, realizing precise control of entering and exiting deep standby. While meeting the user's usage requirements, the standby power consumption is minimized to the greatest extent, avoiding standby energy waste during the night on working days and holiday periods. It is applicable to air - conditioning systems with local control, without being restricted by the timed power - on wake - up function, having higher applicability and stronger popularizability, and the energy - saving effect is more precise and efficient.

[0054] Next, the low - standby control device for an air - conditioning system based on machine self - learning provided by the present invention will be described. The low - standby control device for an air - conditioning system based on machine self - learning described below can be mutually referred to with the low - standby control method for an air - conditioning system based on machine self - learning described above.

[0055] Figure 5 A schematic structural diagram of a low - standby control device for an air - conditioning system based on machine self - learning is shown. The device includes: A data acquisition module 51, which acquires the time intervals between adjacent power - on and power - off within the target time. The time intervals between adjacent power - on and power - off include power - on stay times and shutdown stay times; A judgment module 52, which determines that the number of shutdown stay times greater than the first preset threshold within the target time is greater than the preset number, uses the latest shutdown time among the corresponding shutdown stay times greater than the first preset threshold as the time point to be switched to the deep low - power standby mode, and acquires the outdoor ambient temperature; A control module 53, which determines the wake - up countdown duration according to the outdoor ambient temperature, and based on the wake - up countdown duration being 0, based on the time point to be switched, controls the system to switch to the deep low - power standby mode.

[0056] In this embodiment, the data acquisition module 51 includes: a record acquisition unit, which acquires the air - conditioning operation records within the target time; a rule recognition unit, which uses a rule recognition model to recognize the air - conditioning operation records to obtain the corresponding time intervals between adjacent power - on and power - off. The rule recognition model is trained in advance based on the air - conditioning historical operation records and the time - interval labels corresponding to the air - conditioning historical operation records.

[0057] Specifically, the record acquisition unit includes: a record acquisition subunit, which acquires the air - conditioning operation records within the target time based on a preset time period.

[0058] In addition, the pattern recognition model includes a feature extraction layer, a feature classification layer, and a pattern recognition layer; correspondingly, the pattern recognition unit includes: a feature extraction unit that uses the feature extraction layer to extract features from the input air conditioner operation record to obtain on-off features; a feature classification unit that uses the feature classification layer to classify the extracted on-off features to obtain on features and off features; a pattern recognition unit that uses the pattern recognition layer to obtain the on-stay time and the off-stay time according to the on features and the off features; wherein, the on-stay time is obtained according to the previous on feature and the adjacent subsequent off feature, and the off-stay time is obtained according to the previous off feature and the adjacent subsequent on feature.

[0059] In an alternative embodiment, the device further includes: a training module that trains the pattern recognition model before using the pattern recognition model to recognize the air conditioner operation record to obtain the time interval between adjacent on-off operations. Specifically, the training module includes: a training data acquisition unit that acquires the air conditioner historical operation record and the time interval label corresponding to the air conditioner historical operation record; a training unit that uses the air conditioner historical operation record as the input data for training and the time interval label corresponding to the air conditioner historical operation record as the label for training, and trains the model to be trained to obtain a pattern recognition model for recognizing and predicting the time interval between adjacent on-off operations.

[0060] In an alternative embodiment, the judgment module 52 includes: a judgment unit that, after obtaining the time interval between adjacent on-off operations within the target time, determines whether the number of off-stay times greater than a first preset threshold within the target time is greater than a preset number; a data re-acquisition unit that, based on the number of off-stay times greater than the first preset threshold within the target time being less than or equal to the preset number, updates the target time and re-acquires the time interval between adjacent on-off operations within the updated target time; a re-judgment unit that, according to the time interval between adjacent on-off operations within the re-acquired updated target time, re-determines whether the number of off-stay times greater than the first preset threshold within the updated target time is greater than the preset number.

[0061] Furthermore, the judgment module 52 further includes: a sensor value acquisition unit that acquires the value of the outer ring temperature sensor within the target time range, and the lower limit value of the target time range is the time point to be switched; an outer ring temperature determination unit that obtains the outer ring temperature according to the mean value of the outer ring temperature sensor values within the target time range.

[0062] The control module 53 includes: a duration determination unit that determines the wake-up countdown duration to be 0 if the outer ring temperature is greater than the first preset temperature; otherwise, it determines the corresponding wake-up countdown duration according to the temperature range of the outer ring temperature in combination with the wake-up setting rule, where the wake-up setting rule is used to define the wake-up countdown duration corresponding to the outer ring temperature in different temperature ranges.

[0063] Further, the wake-up setting rule includes: if the outer ring temperature is greater than the second preset temperature and less than or equal to the first preset temperature, it determines the wake-up countdown duration to be the first preset duration; if the outer ring temperature is greater than the third preset temperature and less than or equal to the second preset temperature, it determines the wake-up countdown duration to be the second preset duration; if the outer ring temperature is less than or equal to the third preset temperature, it determines the wake-up countdown duration to be the third preset duration. Wherein, the first preset temperature is greater than the second preset temperature, the second preset temperature is greater than the third preset temperature, the first preset duration is greater than 0 and less than the second preset duration, the second preset duration is less than the third preset duration, and the wake-up countdown duration is set in units of days and in 24-hour timing.

[0064] In an alternative embodiment, the control module further includes: a power-on determination unit that, after determining the wake-up countdown duration according to the outer ring temperature, based on the wake-up countdown duration being non-zero, determines whether there is a power-on behavior within a preset time after the end of the wake-up countdown according to the time interval between adjacent power-on and power-off within the target time. Wherein, the wake-up countdown is carried out from the time point to be switched according to the wake-up countdown; a power-on control unit that, if there is a power-on behavior within the preset time after the end of the wake-up countdown, controls the system to power on at the corresponding power-on time and controls the system to be in the normal working mode.

[0065] In addition, the control module further includes: a mode switching unit that, if there is no power-on behavior within the preset time after the end of the wake-up countdown, controls the system to switch to the deep low-power standby mode based on the time point to be switched.

[0066] In summary, in the embodiment of the present invention, the judgment module determines that the number of shutdown stay times greater than the first preset threshold in the time interval between adjacent power-on and power-off within the target time obtained by the data acquisition module is greater than the preset number, so as to preliminarily distinguish weekdays and rest days, and takes the latest shutdown time on weekdays as the time point to be switched. Then, the control module determines the corresponding wake-up countdown duration based on the obtained outer ring temperature, so that when the wake-up countdown duration is 0, it directly determines that the control system switches to the deep low-power standby mode based on the time point to be switched, realizing precise control of entering and exiting deep standby, minimizing standby power consumption to the greatest extent while meeting the user's usage requirements, avoiding standby energy waste during weekdays at night and holiday periods, being applicable to local control air-conditioning systems, not being restricted by the timed power-on wake-up function, having higher applicability and stronger popularizability, and having more precise and efficient energy-saving effects.

[0067] Figure 6 Illustrates a schematic diagram of the physical structure of an electronic device, as Figure 6 shown. The electronic device may include: a processor 610, a communications interface 620, a memory 630, and a communication bus 640. Among them, the processor 610, the communications interface 620, and the memory 630 complete communication with each other through the communication bus 640. The processor 610 may call logic instructions in the memory 630 to execute a low standby control method for an air-conditioning system based on machine self-learning. The method includes: obtaining the time interval between adjacent power-on and power-off within a target time, where the time interval between adjacent power-on and power-off includes the power-on residence time and the power-off residence time; determining that the number of power-off residence times greater than a first preset threshold within the target time is greater than a preset number, taking the latest power-off time in the corresponding power-off residence times greater than the first preset threshold as the time point to be switched to the deep low-power standby mode, and obtaining the outer ambient temperature; according to the outer ambient temperature, determining the wake-up countdown duration, and based on the wake-up countdown duration being 0, and based on the time point to be switched, controlling the system to switch to the deep low-energy standby mode.

[0068] In addition, when the logic instructions in the above-mentioned memory 630 can be implemented in the form of software functional units and sold or used as an independent product, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. And the foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0069] On the other hand, the present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the low standby control method of the air-conditioning system based on machine self-learning provided by the above-mentioned various methods. The method includes: obtaining the time interval between adjacent power-on and power-off operations within a target time. The time interval between adjacent power-on and power-off operations includes the power-on residence time and the power-off residence time; determining that the number of power-off residence times greater than a first preset threshold within the target time is greater than a preset number, taking the latest power-off time among the corresponding power-off residence times greater than the first preset threshold as the time point to be switched to the deep low-power standby mode, and obtaining the external ambient temperature; determining the wake-up countdown duration according to the external ambient temperature, and based on the wake-up countdown duration being 0, and based on the time point to be switched, controlling the system to switch to the deep low-energy standby mode.

[0070] In another aspect, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it is configured to execute the low standby control method of the air-conditioning system based on machine self-learning provided by the above-mentioned various methods. The method includes: obtaining the time interval between adjacent power-on and power-off operations within a target time. The time interval between adjacent power-on and power-off operations includes the power-on residence time and the power-off residence time; determining that the number of power-off residence times greater than a first preset threshold within the target time is greater than a preset number, taking the latest power-off time among the corresponding power-off residence times greater than the first preset threshold as the time point to be switched to the deep low-power standby mode, and obtaining the external ambient temperature; determining the wake-up countdown duration according to the external ambient temperature, and based on the wake-up countdown duration being 0, and based on the time point to be switched, controlling the system to switch to the deep low-energy standby mode.

[0071] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. A person of ordinary skill in the art can understand and implement it without creative labor.

[0072] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0073] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A low standby control method for an air conditioning system based on machine self-learning, characterized in that, Including: Obtain the time interval between adjacent power - on and power - off within the target time, where the time interval between adjacent power - on and power - off includes the power - on stay time and the power - off stay time; Determine that the number of power - off stay times greater than the first preset threshold within the target time is greater than the preset number. Take the latest power - off time among the corresponding power - off stay times greater than the first preset threshold as the time point to be switched to the deep low - power standby mode, and obtain the ambient temperature outside the loop; According to the ambient temperature outside the loop, determine the wake - up countdown duration, and based on the wake - up countdown duration being 0 and based on the time point to be switched, control the system to switch to the deep low - energy consumption standby mode.

2. The low standby control method for an air conditioning system based on machine self-learning according to claim 1, characterized in that, The obtaining the time interval between adjacent power - on and power - off within the target time includes: Obtain the air - conditioner operation record within the target time; Use a pattern recognition model to recognize the air - conditioner operation record to obtain the time interval between corresponding adjacent power - on and power - off; where the pattern recognition model is pre - trained based on the air - conditioner historical operation record and the time - interval label corresponding to the air - conditioner historical operation record.

3. The low standby control method for an air conditioning system based on machine self-learning according to claim 1, wherein The power - on stay time represents the duration from power - on to power - off; after determining the wake - up countdown duration according to the ambient temperature outside the loop, it further includes: Based on the wake - up countdown duration being non - zero, determine whether there is a power - on behavior within a preset time after the end of the wake - up countdown according to the time interval between adjacent power - on and power - off within the target time and the power - on stay time; where the wake - up countdown starts from the time point to be switched according to the wake - up countdown; If there is a power - on behavior within the preset time after the end of the wake - up countdown, the system powers on at the corresponding power - on time and the system is in the normal working mode.

4. The low standby control method for an air conditioning system based on machine self-learning according to claim 3, characterized in that, The determining whether there is a power - on behavior within a preset time after the end of the wake - up countdown includes: If there is no power - on behavior within the preset time after the end of the wake - up countdown, then based on the time point to be switched, control the system to switch to the deep low - energy consumption standby mode.

5. The low standby control method for an air conditioning system based on machine self-learning according to claim 1, characterized in that After obtaining the time interval between adjacent power - on and power - off within the target time, it further includes: Determine whether the number of power - off stay times greater than the first preset threshold within the target time is greater than the preset number; Based on the number of power - off stay times greater than the first preset threshold within the target time being less than or equal to the preset number, update the target time and re - obtain the time interval between adjacent power - on and power - off within the updated target time; According to the re - obtained time interval between adjacent power - on and power - off within the updated target time, re - determine whether the number of power - off stay times greater than the first preset threshold within the updated target time is greater than the preset number.

6. The low standby control method for an air conditioning system based on machine self-learning according to claim 1, wherein Determining the wake - up countdown duration according to the ambient temperature outside the loop includes: If the ambient temperature outside the loop is greater than the first preset temperature, determine that the wake - up countdown duration is 0; Otherwise, according to the temperature range where the ambient temperature outside the loop is located, combined with the wake - up setting rule, determine the corresponding wake - up countdown duration; where the wake - up setting rule is used to limit the wake - up countdown duration corresponding to the ambient temperature outside the loop in different temperature ranges.

7. The low standby control method for an air conditioning system based on machine self-learning according to claim 6, characterized in that, The wake - up setting rule includes: If the ambient temperature outside the ring is greater than the second preset temperature and less than or equal to the first preset temperature, determine that the wake-up countdown duration is the first preset duration; If the ambient temperature outside the ring is greater than the third preset temperature and less than or equal to the second preset temperature, determine that the wake-up countdown duration is the second preset duration; If the ambient temperature outside the ring is less than or equal to the third preset temperature, determine that the wake-up countdown duration is the third preset duration; Wherein, the first preset temperature is greater than the second preset temperature, the second preset temperature is greater than the third preset temperature, the first preset duration is greater than 0 and less than the second preset duration, the second preset duration is less than the third preset duration, and the wake-up countdown duration is set in units of days and in 24-hour timekeeping.

8. An air conditioner system low standby control device based on machine self-learning, characterized in that, Including: A data acquisition module that acquires the time interval between adjacent power-on and power-off operations within the target time, and the time interval between adjacent power-on and power-off operations includes the power-on stay time and the power-off stay time; A judgment module that determines that the number of power-off stay times greater than the first preset threshold within the target time is greater than the preset number, takes the latest power-off time among the corresponding power-off stay times greater than the first preset threshold as the time point to be switched to the deep low-power standby mode, and acquires the ambient temperature outside the ring; A control module that determines the wake-up countdown duration according to the ambient temperature outside the ring, and based on the wake-up countdown duration being 0 and based on the time point to be switched, controls the system to switch to the deep low-power standby mode.

9. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein, When the processor executes the program, it implements the steps of the low standby control method for an air-conditioning system based on machine self-learning according to any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the low standby control method for an air-conditioning system based on machine self-learning according to any one of claims 1 to 7.

Images