WO2022073307A1 - 空调的预启动时间控制方法、装置、设备及非易失性存储介质 - Google Patents
空调的预启动时间控制方法、装置、设备及非易失性存储介质 Download PDFInfo
- Publication number
- WO2022073307A1 WO2022073307A1 PCT/CN2020/141404 CN2020141404W WO2022073307A1 WO 2022073307 A1 WO2022073307 A1 WO 2022073307A1 CN 2020141404 W CN2020141404 W CN 2020141404W WO 2022073307 A1 WO2022073307 A1 WO 2022073307A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- parameters
- time
- air conditioner
- current
- duration
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 60
- 238000012549 training Methods 0.000 claims description 83
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 70
- 238000007667 floating Methods 0.000 claims description 44
- 238000013528 artificial neural network Methods 0.000 claims description 43
- 238000012360 testing method Methods 0.000 claims description 42
- 230000007613 environmental effect Effects 0.000 claims description 38
- 230000005855 radiation Effects 0.000 claims description 26
- 238000012545 processing Methods 0.000 claims description 25
- 238000013507 mapping Methods 0.000 claims description 12
- 230000002159 abnormal effect Effects 0.000 claims description 11
- 238000004422 calculation algorithm Methods 0.000 claims description 11
- 238000003062 neural network model Methods 0.000 claims description 8
- 238000012216 screening Methods 0.000 claims description 7
- 238000006243 chemical reaction Methods 0.000 claims description 2
- 238000004378 air conditioning Methods 0.000 description 28
- 238000004891 communication Methods 0.000 description 9
- 230000008569 process Effects 0.000 description 8
- 230000003203 everyday effect Effects 0.000 description 6
- 238000001816 cooling Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 238000012423 maintenance Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000005265 energy consumption Methods 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000004140 cleaning Methods 0.000 description 2
- 238000005094 computer simulation Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000012937 correction Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000017525 heat dissipation Effects 0.000 description 2
- 238000005338 heat storage Methods 0.000 description 2
- 230000010365 information processing Effects 0.000 description 2
- 238000013178 mathematical model Methods 0.000 description 2
- 238000012821 model calculation Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 230000005477 standard model Effects 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 238000010438 heat treatment Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/50—Control or safety arrangements characterised by user interfaces or communication
- F24F11/61—Control or safety arrangements characterised by user interfaces or communication using timers
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/62—Control or safety arrangements characterised by the type of control or by internal processing, e.g. using fuzzy logic, adaptive control or estimation of values
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/62—Control or safety arrangements characterised by the type of control or by internal processing, e.g. using fuzzy logic, adaptive control or estimation of values
- F24F11/63—Electronic processing
- F24F11/64—Electronic processing using pre-stored data
Definitions
- the present application relates to the technical field of air conditioners, and in particular, to a method, device, device and non-volatile storage medium for controlling the pre-start time of an air conditioner.
- the pre-cooling time of the central air conditioner is usually estimated by the property staff based on experience, and a pre-turn-on time is set on the building automation system.
- the pre-cooling time estimated by experience is quite different from the actual pre-cooling time required. Turning on the air-conditioning system will result in a waste of energy, and turning on the air-conditioning system too late will cause the indoor temperature to be too high during working hours, resulting in poor personnel comfort.
- the main purpose of the present application is to provide a method, device, device and non-volatile storage medium for controlling the pre-start time of an air conditioner, aiming to solve the technical problem that the pre-start time of an air conditioner cannot be accurately determined in the art.
- the present application provides a method for controlling the pre-start time of an air conditioner, and the method for controlling the pre-start time of the air conditioner includes the following steps:
- a time prediction is performed through a preset time prediction model to determine the target pre-start time of the air conditioner, and the preset time prediction
- the model is obtained by training the initial neural network model
- the air conditioner is controlled to realize timing start according to the target pre-start time.
- the method before the step of acquiring the current indoor environment parameters, the current outdoor environment parameters and the current equipment operating parameters where the air conditioner is located, the method further includes:
- the initial neural network is trained according to each group of prediction model parameters and the standard value of the training duration corresponding to each group of prediction model parameters to obtain an initial time prediction neural network;
- the initial time prediction neural network is used as a preset time prediction model.
- the step of acquiring the training subset of indoor environment parameters, the training subset of outdoor environment parameters and the training subset of equipment operating parameters includes:
- the plurality of the sample indoor environment parameters, the plurality of sample outdoor environment parameters, and a plurality of the sample equipment operating parameters meet the preset data continuity conditions
- the plurality of the sample indoor environment parameters, the plurality of sample equipment operating parameters are respectively performing data screening on the sample outdoor environmental parameters and a plurality of the sample equipment operating parameters to determine a plurality of indoor environmental parameters, a plurality of outdoor environmental parameters and a plurality of equipment operating parameters;
- An indoor environment parameter training subset is constructed according to a plurality of the indoor environment parameters
- an outdoor environment parameter training subset is constructed according to the plurality of the outdoor environment parameters
- a device operation parameter training subset is constructed according to the plurality of the device operation parameters.
- the method before the step of using the initial prediction time neural network as a preset time prediction model, the method further includes:
- the step of using the initial time prediction neural network as a preset time prediction model is performed.
- the time prediction is performed by a preset time prediction model to determine the target pre-start of the air conditioner. time steps, including:
- the target pre-start time of the air conditioner is determined according to the start time period and the expected time.
- the step of processing the start-up duration code to obtain the start-up duration of the air conditioner includes:
- a start-up duration sample corresponding to the successfully matched sample time floating point is used as the start-up duration of the air conditioner.
- the method before the step of processing the startup time code to obtain the corresponding time floating point, the method further includes:
- a preset time mapping relationship table is established according to a plurality of the sample time floating point and the sample start-up duration.
- the present application also proposes a pre-start time control device for an air conditioner, and the pre-start time control device for an air conditioner includes:
- the acquisition module is used to acquire the current indoor environment parameters, the current outdoor environment parameters and the current equipment operating parameters where the air conditioner is located;
- the algorithm module is configured to perform time prediction through a preset time prediction model according to the current indoor environment parameters, the current outdoor environment parameters and the current equipment operating parameters, so as to determine the target pre-start time of the air conditioner, so
- the preset time prediction model is obtained by training an initial neural network model
- the control module is configured to control the air conditioner to realize timing start according to the target pre-start time.
- the present application also proposes a pre-start time control device for an air conditioner
- the air conditioner pre-start time control device includes: a memory, a processor, and a device stored in the memory and available in the processor
- a pre-start time control program of the air conditioner running on the air conditioner when the pre-start time control program of the air conditioner is executed by the processor, implements the steps of the above-mentioned method for controlling the pre-start time of the air conditioner.
- the present application also proposes a non-volatile storage medium, where a pre-start time control program of an air conditioner is stored on the non-volatile storage medium, and the pre-start time control program of the air conditioner is processed When the controller is executed, the steps of the method for controlling the pre-start time of the air conditioner as described above are realized.
- the current indoor environment parameters, the current outdoor environment parameters, and the current equipment operating parameters where the air conditioner is located are first obtained, and then a preset time prediction model is used to predict the current indoor environment parameters, the current outdoor environment parameters, and the current equipment operating parameters.
- Time prediction is used to determine the target pre-start time of the air conditioner, and finally the air conditioner is controlled to realize the timing start according to the target pre-start time.
- the related art starts by manually controlling the air conditioner, but the time when the air conditioner is turned on is too long or too short, resources will be wasted, and the present application uses a preset time prediction model according to indoor environmental parameters, outdoor environmental parameters and equipment operating parameters. The time prediction is performed to determine the target pre-start time of the air conditioner, thereby controlling the air conditioner to start regularly according to the precise pre-start time, thereby reducing the energy consumption of the air conditioner and improving the user experience.
- FIG. 1 is a schematic structural diagram of a pre-start time control device for an air conditioner of an air conditioner in a hardware operating environment according to an embodiment of the present application;
- FIG. 2 is a schematic flowchart of a first embodiment of a method for controlling a pre-start time of an air conditioner of the present application
- FIG. 3 is a schematic flowchart of a second embodiment of a method for controlling a pre-start time of an air conditioner of the present application
- FIG. 4 is a structural block diagram of a first embodiment of a pre-start time control device for an air conditioner of the present application.
- FIG. 1 is a schematic structural diagram of a device for controlling the pre-start time of an air conditioner in a hardware operating environment according to an embodiment of the present application.
- the pre-start time control device of the air conditioner may include: a processor 1001, such as a central processing unit (Central Processing Unit) Processing Unit, CPU), communication bus 1002 , user interface 1003 , network interface 1004 , memory 1005 .
- the communication bus 1002 is used to realize the connection and communication between these components.
- the user interface 1003 may include a display screen (Display), and the optional user interface 1003 may also include a standard wired interface and a wireless interface.
- the wired interface of the user interface 1003 may be a USB interface in this application.
- the network interface 1004 may include a standard wired interface and a wireless interface (eg, a wireless fidelity (WIreless-FIdelity, WI-FI) interface).
- the memory 1005 may be a high-speed random access memory (Random Access Memory, RAM) memory, or may be a non-volatile memory (Non-volatile Memory, NVM), such as a disk memory.
- RAM Random Access Memory
- NVM non-volatile Memory
- the memory 1005 may also be a storage device independent of the aforementioned processor 1001 .
- FIG. 1 does not constitute a limitation on the pre-start time control device of the air conditioner, and may include more or less components than the one shown, or combine some components, or different Component placement.
- the memory 1005 identified as a computer non-volatile storage medium may include an operating system, a network communication module, a user interface module and a pre-start time control program of the air conditioner.
- the network interface 1004 is mainly used to connect to a background server and perform data communication with the background server;
- the user interface 1003 is mainly used to connect user equipment;
- the time control device invokes the pre-start time control program of the air conditioner stored in the memory 1005 through the processor 1001, and executes the method for pre-start time control of the air conditioner provided by the embodiments of the present application.
- FIG. 2 is a schematic flowchart of the first embodiment of the pre-start time control method of the air conditioner of the present application, and the first embodiment of the pre-start time control method of the air conditioner of the present application is proposed.
- the method for controlling the pre-start time of the air conditioner includes the following steps:
- Step S10 Acquire current indoor environment parameters, current outdoor environment parameters, and current equipment operating parameters where the air conditioner is located.
- the executive body of this embodiment is an air conditioner pre-start time control device, wherein the device is an air conditioner pre-start time control device with functions such as data processing, data communication, and program operation, and can also be other devices. , which is not limited in this embodiment.
- the air conditioner may be a central air conditioning system, or a household air conditioner, or the like.
- the current indoor environment parameters can be the indoor dry bulb temperature, indoor wet bulb temperature, etc. collected at the current moment
- the current equipment operating parameters can be the chilled water supply temperature and chilled water return water temperature, etc.
- the current outdoor environment parameters can be collected at the current moment.
- the outdoor wet bulb temperature, outdoor dry bulb temperature, solar radiation intensity, etc. are not limited in this embodiment.
- the indoor dry bulb temperature is 30, the indoor wet bulb temperature is 31, the chilled water supply temperature is 23, the chilled water return water temperature is 24, the outdoor wet bulb temperature is 0, and the outdoor dry bulb temperature is 0. If it is 1 and the solar radiation intensity is 17, the outdoor wet bulb temperature and outdoor dry bulb temperature are abnormal data, and the collection of indoor and outdoor environmental parameters at 6:00 is invalid, and the indoor and outdoor environmental parameters need to be collected again.
- Step S20 According to the current indoor environment parameters, the current outdoor environment parameters and the current equipment operating parameters, time prediction is performed through a preset time prediction model, so as to determine the target pre-start time of the air conditioner, the pre-start time of the air conditioner. Let the time prediction model be obtained by training the initial neural network model.
- the prediction model parameter training set includes several groups of prediction model parameters. Obtain the standard value of the training duration corresponding to each group of prediction model parameters, that is, the standard value of the air conditioner startup duration.
- the initial neural network is trained to obtain an initial time prediction neural network, and the initial time prediction neural network is used as a preset time prediction model.
- the indoor environmental parameters can be indoor dry bulb temperature, indoor wet bulb temperature
- equipment operating parameters can be chilled water supply temperature and chilled water return water temperature
- outdoor environmental parameters can be outdoor wet bulb temperature and outdoor dry bulb temperature and solar radiation Intensity, etc.
- one of the prediction model parameters is indoor dry bulb temperature, indoor wet bulb temperature, chilled water supply temperature, chilled water return water temperature, outdoor wet bulb temperature, outdoor dry bulb temperature and solar radiation intensity.
- the preset data continuous condition is that when collecting parameters such as indoor dry bulb temperature, indoor wet bulb temperature, chilled water supply temperature, chilled water return water temperature, outdoor wet bulb temperature, outdoor dry bulb temperature and solar radiation intensity, the data will not be interrupted. Completely collect a set of data, and then judge whether the collected data belongs to abnormal data or missing data.
- the parameters that meet the requirements are screened to construct an indoor environment parameter training subset, an outdoor environment parameter training subset and an equipment operating parameter training subset.
- the step of using the initial prediction time neural network as the preset time prediction model Before the step of using the initial prediction time neural network as the preset time prediction model, obtain indoor environment test parameters, outdoor environment test parameters and equipment operation test parameters, and input the indoor environment test parameters, outdoor environment test parameters and equipment operation test parameters To the initial time prediction neural network, to obtain the predicted value of the starting time of the air conditioner, determine the difference of the starting time according to the predicted value of the starting time and the standard value of the predicted time, and judge whether the difference of the starting time is less than the preset time. When the value is smaller than the preset duration threshold, the step of using the initial time prediction neural network as the preset time prediction model is performed.
- the preset time prediction model is tested to determine whether the preset time prediction model meets the standard model conditions, and the indoor environment test parameters, outdoor environment test parameters and equipment operation test parameters are collected in real time at a certain moment, and the indoor environment test parameters are obtained. parameters, outdoor environment test parameters and equipment operation test parameters corresponding to the standard value of the prediction duration, after that, input the indoor environment test parameters, outdoor environment test parameters and equipment operation test parameters into the initial time prediction neural network to obtain the start-up of the air conditioner
- the predicted value of the duration, and the difference of the startup duration is determined according to the predicted value of the startup duration and the standard value of the predicted duration.
- the startup duration difference is 10 minutes and the preset duration threshold is 30 minutes
- the startup duration difference of 10 minutes is less than the predicted duration threshold of 30 minutes.
- the subsequent initial neural network is used as the preset time prediction model. If the start-up time difference of 40 minutes is greater than the prediction time threshold of 30 minutes, the initial neural network needs to be retrained according to the collected prediction model parameters.
- the optimal startup time prediction model of the central air conditioner is established, that is, the preset time prediction model. Since the temperature in the building is affected by the outdoor temperature and humidity, solar radiation, indoor equipment, lighting, personnel heat dissipation and maintenance structure heat storage , making the whole building a very complex thermal system, and it is difficult to describe the temperature change process in the building with an accurate mathematical model. Therefore, the optimal start-up time of the air-conditioning system cannot be expressed by an ordinary functional relationship.
- the artificial neural network algorithm has nonlinear and adaptive information processing capabilities, which makes up for the shortcomings of traditional algorithms and can infinitely approximate any multi-variable nonlinear relationship. Therefore, it is very feasible to use the neural network algorithm to predict the optimal start-up time of the air-conditioning system. sex.
- the artificial neural network is structurally divided into an input layer, a hidden layer and an output layer, and sometimes a special structure layer is set up, which can store the information in the previous moment of the hidden layer.
- the correction of the algorithm and connection weights is passed to the hidden layer, and after the processing of the hidden layer is completed, the output feedback at this time is propagated to the structural layer, and the unit value is not released until the next training time.
- the artificial neural network improves the processing capability of dynamic identification by adding internal feedback signals, and is suitable for solving the problem of dynamic simulation and prediction of air-conditioning systems.
- the construction steps of the preset time prediction model can be divided into:
- Step 1 Select the predictive model parameters that meet the conditions from the historical database of predictive model parameters, and perform data cleaning to remove data such as zero values, outliers, and missing values;
- Step 2 Screening of input and output parameters, where the input parameters include: outdoor dry bulb temperature at t0, outdoor wet bulb temperature at t0, indoor dry bulb temperature at t0, indoor wet bulb temperature at t0, chilled water supply temperature at t0, and temperature at t0 Chilled water return water temperature, solar radiation intensity at t0.
- Time t0 is a certain time before the host is powered on every day, such as 6:00.
- Step 3 The artificial neural network model is trained according to the input and output parameters, and the weight coefficients in the model are automatically identified. In order to ensure the accuracy of the prediction time, the neural network model is retrained according to the newly added historical data at regular intervals, and the model weight coefficients are updated.
- the building automation system judges whether the current time has reached the start-up time t4, if not, it will wait. If it has reached the start-up time t4, it will send a start-up command to the chiller, that is to say, it will judge whether the difference between the staff's working time and the current time is less than or equal to the predicted time. When the air conditioner is turned on, if it is less than or equal to the predicted time, the start command will be sent, otherwise it will not wait.
- the time prediction is performed through a preset time prediction model to determine the target pre-start time of the air conditioner.
- the parameters and the current outdoor key parameters are input into the preset time prediction model to obtain the start-up time code of the air conditioner, and the start-up time code is processed to obtain the start-up time of the air conditioner and the expected time corresponding to the indoor comfortable temperature.
- the expected time determines a target pre-start time of the air conditioner.
- the steps of processing the start-up duration code to obtain the start-up duration of the air conditioner, processing the start-up duration code to obtain the corresponding time floating point, and mapping the time floating point to the sample time in the preset time mapping relationship table The floating point is matched. If the matching is successful, the start time sample corresponding to the floating point of the successfully matched sample time is used as the start time of the air conditioner.
- the startup duration establishes a preset time mapping relationship table.
- an intelligent building management system is set up locally.
- the cloud technology is used to deploy the optimal startup time prediction model of the central air-conditioning system on the cloud computing platform, and the remote invocation service of the intelligent building management system is established.
- the intelligent building management system regularly sends data to the cloud prediction model, and returns the predicted time value after the model calculation to realize the dynamic optimization of the startup time of the central air-conditioning system.
- the cloud computing platform remote optimal startup time prediction control steps are as follows:
- the intelligent building management system sends a data request to the cloud computing platform at a certain time (for example, 6:00) before starting up every day, and reports the indoor temperature and humidity, outdoor temperature and humidity, solar radiation intensity, and water supply and return water temperature of the chiller at that time. and other parameters are sent to the cloud computing platform prediction model, and the data is processed into a specific code;
- the cloud computing platform model After the cloud computing platform model is calculated, it returns the code containing the optimal startup time to the intelligent building management system, and the intelligent building management system automatically parses the code into a time value, and modifies the set value of the central air-conditioning system startup time of the day;
- the intelligent building management system sends a start-up command to the central air-conditioning system to complete the optimal start-up time prediction control.
- Step S30 controlling the air conditioner to realize timing start according to the target pre-start time.
- the current indoor environment parameters, the current outdoor key parameters and the current equipment operating parameters are input into the preset time prediction model to obtain the start-up duration code of the air conditioner, and the start-up duration code is processed to obtain the start-up duration of the air conditioner.
- obtain the expected time corresponding to the indoor comfortable temperature determine the target pre-start time of the air conditioner according to the start-up duration and the expected time, and then control the air conditioner to realize timing start according to the target pre-start time.
- the current indoor environment parameters, the current outdoor environment parameters and the current equipment operating parameters where the air conditioner is located are first obtained, and then the current indoor environment parameters, the current outdoor environment parameters and the current equipment operating parameters are obtained through a preset time prediction model.
- Time prediction is used to determine the target pre-start time of the air conditioner, and finally the air conditioner is controlled to realize the timing start according to the target pre-start time. Since the related art starts by manually controlling the air conditioner, but the time when the air conditioner is turned on is too long or too short, resources will be wasted.
- the preset time is used to predict The model performs time prediction to determine the target pre-start time of the air conditioner, so as to control the air conditioner to start regularly according to the precise pre-start time, thereby reducing the energy consumption of the air conditioner.
- FIG. 3 shows the first embodiment of the pre-start time control method of the air conditioner based on the above-mentioned, and the second embodiment of the pre-start time control method of the air conditioner of the present application is proposed.
- step S20 in the method for controlling the pre-start time of the air conditioner includes:
- Step S201 Input the current indoor environment parameters, the current outdoor environment parameters, and the current equipment operating parameters into a preset time prediction model to obtain a start-up duration code of the air conditioner.
- the current indoor environment parameters can be the indoor dry bulb temperature, indoor wet bulb temperature, etc. collected at the current moment
- the equipment operating parameters can be the chilled water supply temperature and chilled water return water temperature, etc.
- the current outdoor environment parameters can be collected at the current moment.
- the outdoor wet bulb temperature, outdoor dry bulb temperature, solar radiation intensity, etc. are not limited in this embodiment.
- each set of start-up duration codes corresponding to indoor dry bulb temperature, indoor wet bulb temperature, chilled water supply temperature, chilled water return water temperature, outdoor wet bulb temperature, outdoor dry bulb temperature and solar radiation intensity are also encoded. That is to say, each group of prediction model parameters is different, and the corresponding startup time coding is also different.
- Step S202 Process the start-up duration code to obtain the start-up duration of the air conditioner.
- the startup duration code Process the startup duration code to obtain the corresponding time floating point, and match the time floating point with the sample time floating point in the preset time mapping relationship table.
- the start-up duration sample is taken as the start-up duration of the air conditioner.
- the start-up duration code may also be encoded to obtain the corresponding time floating point, and then converted into the corresponding start-up duration of the air conditioner according to the time floating point according to certain data conversion rules, which is not limited in this embodiment.
- the startup duration establishes a preset time mapping relationship table.
- the start-up duration can be understood as the pre-cooling or heating start-up duration of the air conditioner.
- Step S203 Obtain the expected time corresponding to the indoor comfortable temperature.
- the expected time may be the time when the user needs to make the indoor temperature reach a comfortable temperature at a certain time, and may be the user's work time at 8:00, or the rest time at 17:00, etc., which is not limited in this embodiment.
- Step S204 Determine the target pre-start time of the air conditioner according to the start-up duration and the expected time.
- the current indoor environment parameters, the current outdoor key parameters and the current equipment operating parameters are firstly input into the preset time prediction model to obtain the start-up duration code of the air conditioner, and the start-up duration code is processed to obtain the Start-up time, obtain the expected time corresponding to the indoor comfortable temperature, and determine the target pre-start time of the air conditioner according to the start-up time and the expected time. Since in the related art, manually setting the pre-start time of the air conditioner will lead to forgetting to set the pre-start time in advance.
- the code of the start-up duration of the air conditioner is processed to obtain the start-up duration of the air-conditioner, the expected time corresponding to the indoor comfortable temperature, and the target pre-start time of the air-conditioner is determined according to the start-up duration and the expected time, thereby realizing automatic control of the air conditioner the pre-start time of the device.
- an embodiment of the present application also proposes a non-volatile storage medium, where a pre-start time control program of the air conditioner is stored on the non-volatile storage medium, and when the pre-start time control program of the air conditioner is executed by the processor.
- an embodiment of the present application also proposes a pre-start time control device for an air conditioner, and the air conditioner pre-start time control device includes:
- the acquisition module 4001 is used to acquire current indoor environment parameters, current outdoor environment parameters, and current equipment operation parameters where the air conditioner is located.
- the executive body of this embodiment is an air conditioner pre-start time control device, wherein the device is an air conditioner pre-start time control device with functions such as data processing, data communication, and program operation, and can also be other devices. , which is not limited in this embodiment.
- the air conditioner may be a central air conditioning system, or a household air conditioner, or the like.
- the current indoor environment parameters can be the indoor dry bulb temperature, indoor wet bulb temperature, etc. collected at the current moment
- the current equipment operating parameters can be the chilled water supply temperature and chilled water return water temperature, etc.
- the current outdoor environment parameters can be collected at the current moment.
- the outdoor wet bulb temperature, outdoor dry bulb temperature, solar radiation intensity, etc. are not limited in this embodiment.
- the indoor dry bulb temperature is 30, the indoor wet bulb temperature is 31, the chilled water supply temperature is 23, the chilled water return water temperature is 24, the outdoor wet bulb temperature is 0, and the outdoor dry bulb temperature is 0. If it is 1 and the solar radiation intensity is 17, the outdoor wet bulb temperature and outdoor dry bulb temperature are abnormal data, and the collection of indoor and outdoor environmental parameters at 6:00 is invalid, and the indoor and outdoor environmental parameters need to be collected again.
- Algorithm module 4002 configured to perform time prediction through a preset time prediction model according to the current indoor environment parameters, the current outdoor environment parameters and the current equipment operating parameters, so as to determine the target pre-start time of the air conditioner,
- the preset time prediction model is obtained by training an initial neural network model.
- the prediction model parameter training set includes several groups of prediction model parameters. Obtain the standard value of the training duration corresponding to each group of prediction model parameters, that is, the standard value of the air conditioner startup duration.
- the initial neural network is trained to obtain an initial time prediction neural network, and the initial time prediction neural network is used as a preset time prediction model.
- the indoor environmental parameters can be indoor dry bulb temperature, indoor wet bulb temperature
- equipment operating parameters can be chilled water supply temperature and chilled water return water temperature
- outdoor environmental parameters can be outdoor wet bulb temperature and outdoor dry bulb temperature and solar radiation Intensity, etc.
- one of the prediction model parameters is indoor dry bulb temperature, indoor wet bulb temperature, chilled water supply temperature, chilled water return water temperature, outdoor wet bulb temperature, outdoor dry bulb temperature and solar radiation intensity.
- the preset data continuous condition is that when collecting parameters such as indoor dry bulb temperature, indoor wet bulb temperature, chilled water supply temperature, chilled water return water temperature, outdoor wet bulb temperature, outdoor dry bulb temperature and solar radiation intensity, the data will not be interrupted. Completely collect a set of data, and then judge whether the collected data belongs to abnormal data or missing data.
- the parameters that meet the requirements are screened to construct an indoor environment parameter training subset, an outdoor environment parameter training subset and an equipment operating parameter training subset.
- the step of using the initial prediction time neural network as the preset time prediction model Before the step of using the initial prediction time neural network as the preset time prediction model, obtain indoor environment test parameters, outdoor environment test parameters and equipment operation test parameters, and input the indoor environment test parameters, outdoor environment test parameters and equipment operation test parameters To the initial time prediction neural network, to obtain the predicted value of the starting time of the air conditioner, determine the difference of the starting time according to the predicted value of the starting time and the standard value of the predicted time, and judge whether the difference of the starting time is less than the preset time. When the value is smaller than the preset duration threshold, the step of using the initial time prediction neural network as the preset time prediction model is performed.
- the preset time prediction model is tested to determine whether the preset time prediction model meets the standard model conditions, and the indoor environment test parameters, outdoor environment test parameters and equipment operation test parameters are collected in real time at a certain moment, and the indoor environment test parameters are obtained. parameters, outdoor environment test parameters and equipment operation test parameters corresponding to the standard value of the prediction duration, after that, input the indoor environment test parameters, outdoor environment test parameters and equipment operation test parameters into the initial time prediction neural network to obtain the start-up of the air conditioner
- the predicted value of the duration, and the difference of the startup duration is determined according to the predicted value of the startup duration and the standard value of the predicted duration.
- the startup duration difference is 10 minutes and the preset duration threshold is 30 minutes
- the startup duration difference of 10 minutes is less than the predicted duration threshold of 30 minutes.
- the subsequent initial neural network is used as the preset time prediction model. If the start-up time difference of 40 minutes is greater than the prediction time threshold of 30 minutes, the initial neural network needs to be retrained according to the collected prediction model parameters.
- the optimal startup time prediction model of the central air conditioner is established, that is, the preset time prediction model. Since the temperature in the building is affected by the outdoor temperature and humidity, solar radiation, indoor equipment, lighting, personnel heat dissipation and maintenance structure heat storage , making the whole building a very complex thermal system, and it is difficult to describe the temperature change process in the building with an accurate mathematical model. Therefore, the optimal start-up time of the air-conditioning system cannot be expressed by an ordinary functional relationship.
- the artificial neural network algorithm has nonlinear and adaptive information processing capabilities, which makes up for the shortcomings of traditional algorithms and can infinitely approximate any multi-variable nonlinear relationship. Therefore, it is very feasible to use the neural network algorithm to predict the optimal start-up time of the air-conditioning system. sex.
- the artificial neural network is structurally divided into an input layer, a hidden layer and an output layer, and sometimes a special structure layer is set up, which can store the information in the previous moment of the hidden layer.
- the correction of the algorithm and connection weights is passed to the hidden layer, and after the processing of the hidden layer is completed, the output feedback at this time is propagated to the structural layer, and the unit value is not released until the next training time.
- the artificial neural network improves the processing capability of dynamic identification by adding internal feedback signals, and is suitable for solving the problem of dynamic simulation and prediction of air-conditioning systems.
- the construction steps of the preset time prediction model can be divided into:
- Step 1 Select the predictive model parameters that meet the conditions from the historical database of predictive model parameters, and perform data cleaning to remove data such as zero values, outliers, and missing values;
- Step 2 Screening of input and output parameters, where the input parameters include: outdoor dry bulb temperature at t0, outdoor wet bulb temperature at t0, indoor dry bulb temperature at t0, indoor wet bulb temperature at t0, chilled water supply temperature at t0, and temperature at t0 Chilled water return water temperature, solar radiation intensity at t0.
- Time t0 is a certain time before the host is powered on every day, such as 6:00.
- Step 3 The artificial neural network model is trained according to the input and output parameters, and the weight coefficients in the model are automatically identified. In order to ensure the accuracy of the prediction time, the neural network model is retrained according to the newly added historical data at regular intervals, and the model weight coefficients are updated.
- the building automation system judges whether the current time has reached the start-up time t4, if not, it will wait. If it has reached the start-up time t4, it will send a start-up command to the chiller, that is to say, it will judge whether the difference between the staff's working time and the current time is less than or equal to the predicted time. When the air conditioner is turned on, if it is less than or equal to the predicted time, the start command will be sent, otherwise it will not wait.
- the time prediction is performed through a preset time prediction model to determine the target pre-start time of the air conditioner.
- the parameters and the current outdoor key parameters are input into the preset time prediction model to obtain the start-up time code of the air conditioner, and the start-up time code is processed to obtain the start-up time of the air conditioner and the expected time corresponding to the indoor comfortable temperature.
- the expected time determines a target pre-start time of the air conditioner.
- the steps of processing the start-up duration code to obtain the start-up duration of the air conditioner, processing the start-up duration code to obtain the corresponding time floating point, and mapping the time floating point to the sample time in the preset time mapping relationship table The floating point is matched. If the matching is successful, the start time sample corresponding to the floating point of the successfully matched sample time is used as the start time of the air conditioner.
- the startup duration establishes a preset time mapping relationship table.
- an intelligent building management system is set up locally.
- the cloud technology is used to deploy the optimal startup time prediction model of the central air-conditioning system on the cloud computing platform, and the remote invocation service of the intelligent building management system is established.
- the intelligent building management system regularly sends data to the cloud prediction model, and returns the predicted time value after the model calculation to realize the dynamic optimization of the startup time of the central air-conditioning system.
- the cloud computing platform remote optimal startup time prediction control steps are as follows:
- the intelligent building management system sends a data request to the cloud computing platform at a certain time (for example, 6:00) every day before the startup, and reports the indoor temperature and humidity, outdoor temperature and humidity, solar radiation intensity, and water supply and return water temperature of the chiller at that time. and other parameters are sent to the cloud computing platform prediction model, and the data is processed into a specific code;
- the cloud computing platform model After the cloud computing platform model is calculated, it returns the code containing the optimal startup time to the intelligent building management system, and the intelligent building management system automatically parses the code into a time value, and modifies the set value of the central air-conditioning system startup time of the day;
- the intelligent building management system sends a start-up command to the central air-conditioning system to complete the optimal start-up time prediction control.
- the control module 4003 is configured to control the air conditioner to realize timing start according to the target pre-start time.
- the current indoor environment parameters, the current outdoor key parameters and the current equipment operating parameters are input into the preset time prediction model to obtain the start-up duration code of the air conditioner, and the start-up duration code is processed to obtain the start-up duration of the air conditioner.
- obtain the expected time corresponding to the indoor comfortable temperature determine the target pre-start time of the air conditioner according to the start-up duration and the expected time, and then control the air conditioner to realize timing start according to the target pre-start time.
- the current indoor environment parameters, the current outdoor environment parameters and the current equipment operating parameters where the air conditioner is located are first obtained, and then the current indoor environment parameters, the current outdoor environment parameters and the current equipment operating parameters are obtained through a preset time prediction model.
- Time prediction is used to determine the target pre-start time of the air conditioner, and finally the air conditioner is controlled to realize the timing start according to the target pre-start time. Since the related art starts by manually controlling the air conditioner, but the time when the air conditioner is turned on is too long or too short, resources will be wasted.
- the preset time is used to predict The model performs time prediction to determine the target pre-start time of the air conditioner, so as to control the air conditioner to start regularly according to the precise pre-start time, thereby reducing the energy consumption of the air conditioner.
- ROM Read Only Memory image
- RAM Random Access Memory
- magnetic disk magnetic disk
- optical disk including several instructions to make a terminal device (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) execute the methods described in the various embodiments of the present application.
- a terminal device which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.
Landscapes
- Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Mechanical Engineering (AREA)
- General Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Human Computer Interaction (AREA)
- Air Conditioning Control Device (AREA)
Abstract
一种空调的预启动时间控制方法、装置、设备及非易失性存储介质，该方法包括：获取空调器所处的当前室内环境参数、当前室外环境参数及当前设备运行参数；根据当前室内环境参数、当前室外环境参数及当前设备运行参数，通过预设时间预测模型进行时间预测，以确定空调器的目标预启动时间；控制空调器按照目标预启动时间实现定时启动。
Description
本申请要求：2020年10月10日申请的、申请号为202011082502.5、名称为“空调的预启动时间控制方法、装置、设备及存储介质”的中国专利申请的优先权，在此将其引入作为参考。
本申请涉及空调技术领域，尤其涉及一种空调的预启动时间控制方法、装置、设备及非易失性存储介质。
随着人们生活水平的不断提高，人们对空调的要求也越来越高，尤其是定时开启中央空调的要求。目前中央空调的预制冷时间通常都由物业人员凭经验估计，在建筑自动化系统上设置一个预开启时间，但凭经验估计得出的预制冷时间与实际所需预制冷时间差异较大，过早开启空调系统造成能源的浪费，过晚开启空调系统则会导致工作时间室内温度过高，人员舒适性差。
上述内容仅用于辅助理解本申请的技术方案，并不代表承认上述内容是现有技术。
本申请的主要目的在于提供一种空调的预启动时间控制方法、装置、设备及非易失性存储介质，旨在解决技术中无法精准确定空调器的预开启时间的技术问题。
为实现上述目的，本申请提供一种空调的预启动时间控制方法，所述空调的预启动时间控制方法包括以下步骤：
获取空调器所处的当前室内环境参数、当前室外环境参数及当前设备运行参数；
根据所述当前室内环境参数、所述当前室外环境参数及所述当前设备运行参数，通过预设时间预测模型进行时间预测，以确定所述空调器的目标预启动时间，所述预设时间预测模型通过对初始神经网络模型进行训练获得；以及
控制所述空调器按照所述目标预启动时间实现定时启动。
在一实施例中，所述获取空调器所处的当前室内环境参数、当前室外环境参数及当前设备运行参数的步骤之前，还包括：
获取室内环境参数训练子集、室外环境参数训练子集及设备运行参数训练子集；
根据所述室内环境参数训练子集、所述室外环境参数训练子集及所述设备运行参数训练子集构建预测模型参数训练集，所述预测模型参数训练集包括若干组预测模型参数；
获取各组预测模型参数对应的训练时长标准值；
根据各组预测模型参数和各组预测模型参数对应的训练时长标准值对初始神经网络进行训练，获得初始时间预测神经网络；以及
将所述初始时间预测神经网络作为预设时间预测模型。
在一实施例中，所述获取室内环境参数训练子集、室外环境参数训练子集及设备运行参数训练子集的步骤，包括：
在预设时间阈值内采集多个样本室内环境参数、多个样本室外环境参数及多个样本设备运行参数；
判断多个所述样本室内环境参数、多个所述样本室外环境参数及多个所述样本设备运行参数是否符合预设数据连续条件；
在多个所述样本室内环境参数、多个所述样本室外环境参数及多个所述样本设备运行参数符合所述预设数据连续条件时，分别对多个所述样本室内环境参数、多个所述样本室外环境参数及多个所述样本设备运行参数进行数据筛选，以确定多个室内环境参数、多个室外环境参数及多个设备运行参数；以及
根据多个所述室内环境参数构建室内环境参数训练子集，根据多个所述室外环境参数构建室外环境参数训练子集，并根据多个所述设备运行参数构建设备运行参数训练子集。
在一实施例中，所述将所述初始预测时间神经网络作为预设时间预测模型的步骤之前，还包括：
获取室内环境测试参数、室外环境测试参数及设备运行测试参数；
将所述室内环境测试参数、所述室外环境测试参数及所述设备运行测试参数输入至所述初始时间预测神经网络中，以获得空调器的启动时长预测值；
根据所述启动时长预测值和预测时长标准值确定启动时长差值；
判断所述启动时长差值是否小于预设时长阈值；以及
在所述启动时长差值小于所述预设时长阈值时，执行所述将所述初始时间预测神经网络作为预设时间预测模型的步骤。
在一实施例中，所述根据所述当前室内环境参数、所述当前室外环境参数及所述当前设备运行参数，通过预设时间预测模型进行时间预测，以确定所述空调器的目标预启动时间的步骤，包括：
将所述当前室内环境参数、所述当前室外环境参数、所述当前设备运行参数输入至预设时间预测模型，以获得所述空调器的启动时长编码；
对所述启动时长编码进行处理，获得所述空调器的启动时长；
获取室内舒适温度对应的期许时间；以及
根据所述启动时长和所述期许时间确定所述空调器的目标预启动时间。
在一实施例中，所述对所述启动时长编码进行处理，获得所述空调器的启动时长的步骤，包括：
对所述启动时长编码进行处理，以获取对应的时间浮点；
将所述时间浮点与预设时间映射关系表中的样本时间浮点进行匹配；
若匹配成功，则将匹配成功的所述样本时间浮点对应的启动时长样本作为所述空调器的启动时长。
在一实施例中，所述对所述启动时间编码进行处理，以获取对应的时间浮点的步骤之前，还包括：
获取多个样本时间浮点；
根据多个所述样本时间浮点分别确定对应的样本启动时长；
根据多个所述样本时间浮点和所述样本启动时长建立预设时间映射关系表。
此外，为实现上述目的，本申请还提出一种空调的预启动时间控制装置，所述空调的预启动时间控制装置包括：
采集模块，用于获取空调器所处的当前室内环境参数、当前室外环境参数及当前设备运行参数；
算法模块，用于根据所述当前室内环境参数、所述当前室外环境参数及所述当前设备运行参数，通过预设时间预测模型进行时间预测，以确定所述空调器的目标预启动时间，所述预设时间预测模型通过对初始神经网络模型进行训练获得；以及
控制模块，用于控制所述空调器按照所述目标预启动时间实现定时启动。
此外，为实现上述目的，本申请还提出一种空调的预启动时间控制设备，所述空调的预启动时间控制设备包括：存储器、处理器及存储在所述存储器上并可在所述处理器上运行的空调的预启动时间控制程序，所述空调的预启动时间控制程序被所述处理器执行时实现如上文所述的空调的预启动时间控制方法的步骤。
此外，为实现上述目的，本申请还提出一种非易失性存储介质，所述非易失性存储介质上存储有空调的预启动时间控制程序，所述空调的预启动时间控制程序被处理器执行时实现如上文所述的空调的预启动时间控制方法的步骤。
本申请中，首先获取空调器所处的当前室内环境参数和当前室外环境参数、当前设备运行参数，然后根据当前室内环境参数和当前室外环境参数、当前设备运行参数，通过预设时间预测模型进行时间预测，以确定空调器的目标预启动时间，最后控制空调器按照目标预启动时间实现定时启动。由于相关技术通过手动控制空调器启动，但空调器定时开启时间过长或过短时，会造成资源浪费，而本申请根据室内环境参数、室外环境参数及设备运行参数，通过预设时间预测模型进行时间预测确定空调器的目标预启动时间，从而控制空调器按照精准的预启动时间定时启动，进而降低空调器的能耗，提高了用户的使用体验。
图1是本申请实施例方案涉及的硬件运行环境的空调的预启动时间控制设备的结构示意图；
图2为本申请空调的预启动时间控制方法第一实施例的流程示意图；
图3为本申请空调的预启动时间控制方法第二实施例的流程示意图；
图4为本申请空调的预启动时间控制装置第一实施例的结构框图。
本申请目的的实现、功能特点及优点将结合实施例，参照附图做进一步说明。
应当理解，此处所描述的具体实施例仅仅用以解释本申请，并不用于限定本申请。
参照图1，图1为本申请实施例方案涉及的硬件运行环境的空调的预启动时间控制设备结构示意图。
如图1所示，该空调的预启动时间控制设备可以包括：处理器1001，例如中央处理器（Central
Processing Unit，CPU），通信总线1002、用户接口1003，网络接口1004，存储器1005。其中，通信总线1002用于实现这些组件之间的连接通信。用户接口1003可以包括显示屏（Display），可选用户接口1003还可以包括标准的有线接口、无线接口，对于用户接口1003的有线接口在本申请中可为USB接口。网络接口1004可选的可以包括标准的有线接口、无线接口（如无线保真（WIreless-FIdelity，WI-FI）接口）。存储器1005可以是高速的随机存取存储器（Random Access Memory，RAM）存储器，也可以是稳定的存储器（Non-volatile Memory，NVM），例如磁盘存储器。存储器1005可选的还可以是独立于前述处理器1001的存储装置。
本领域技术人员可以理解，图1中示出的结构并不构成对空调的预启动时间控制设备的限定，可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件布置。
如图1所示，认定为一种计算机非易失性存储介质的存储器1005中可以包括操作系统、网络通信模块、用户接口模块以及空调的预启动时间控制程序。
在图1所示的空调的预启动时间控制设备中，网络接口1004主要用于连接后台服务器，与所述后台服务器进行数据通信；用户接口1003主要用于连接用户设备；所述空调的预启动时间控制设备通过处理器1001调用存储器1005中存储的空调的预启动时间控制程序，并执行本申请实施例提供的空调的预启动时间控制方法。
基于上述硬件结构，提出本申请空调的预启动时间控制方法的实施例。
参照图2，图2为本申请空调的预启动时间控制方法第一实施例的流程示意图，提出本申请空调的预启动时间控制方法第一实施例。
在第一实施例中，所述空调的预启动时间控制方法包括以下步骤：
步骤S10：获取空调器所处的当前室内环境参数和当前室外环境参数、当前设备运行参数。
需要说明的是，本实施例的执行主体是空调的预启动时间控制设备，其中，该设备是具有数据处理，数据通信及程序运行等功能的空调的预启动时间控制设备，也可为其他设备，本实施例对此不做限制。
空调器可以是中央空调系统，也可以是家用空调器等。
当前室内环境参数可以为当前某刻采集的室内干球温度、室内湿球温度等，当前设备运行参数可以为冷冻水供水温度及冷冻水回水温度等，当前室外环境参数可以为当前某刻采集的室外湿球温度和室外干球温度和太阳辐射强度等，本实施例并不加以限制。
也就是说，在某一时刻同时采集室内干球温度、室内湿球温度、冷冻水供水温度、冷冻水回水温度、室外湿球温度和室外干球温度和太阳辐射强度，并判断上述采集的数据是否异常或缺失，在上述某刻采集的数据出现异常或缺失时，重新采集下一时刻的室内干球温度、室内湿球温度、冷冻水供水温度、冷冻水回水温度、室外湿球温度和室外干球温度和太阳辐射强度，直至某一时刻采集的数据完整。
可以理解为，假设6:00时，室内干球温度为30、室内湿球温度为31、冷冻水供水温度为23、冷冻水回水温度为24、室外湿球温度为0和室外干球温度为1和太阳辐射强度17，则室外湿球温度和室外干球温度为异常数据，6:00采集室内环境参数和室外环境参数无效，需要重新采集室内环境参数和室外环境参数等。
步骤S20：根据所述当前室内环境参数、所述当前室外环境参数及所述当前设备运行参数，通过预设时间预测模型进行时间预测，以确定所述空调器的目标预启动时间，所述预设时间预测模型通过对初始神经网络模型进行训练获得。
获取室内环境参数训练子集、室外环境参数训练子集及设备运行参数训练子集，根据室内环境参数训练子集、室外环境参数训练子集及设备运行参数训练子集构建预测模型参数训练集，预测模型参数训练集包括若干组预测模型参数，获取各组预测模型参数对应的训练时长标准值即空调器启动时长标准值，根据各组预测模型参数和各组预测模型参数对应的训练时长标准值对初始神经网络进行训练，获得初始时间预测神经网络，将初始时间预测神经网络作为预设时间预测模型。
其中，室内环境参数可以为室内干球温度、室内湿球温度、设备运行参数可以为冷冻水供水温度及冷冻水回水温度，室外环境参数可以为室外湿球温度和室外干球温度和太阳辐射强度等，其中一组预测模型参数为室内干球温度、室内湿球温度、冷冻水供水温度、冷冻水回水温度、室外湿球温度、室外干球温度和太阳辐射强度。
获取室内环境参数训练子集、室外环境参数训练子集及设备运行参数训练子集的步骤，在预设时间阈值内采集多个样本室内环境参数、多个样本室外环境参数及多个样本设备运行参数，判断多个所述样本室内环境参数、多个所述样本室外环境参数及多个所述样本设备运行参数是否符合预设数据连续条件，在多个所述样本室内环境参数、多个所述样本室外环境参数及多个所述样本设备运行参数符合所述预设数据连续条件时，分别对多个所述样本室内环境参数、多个所述样本室外环境参数及多个所述样本设备运行参数进行数据筛选，以确定多个室内环境参数、多个室外环境参数及多个设备运行参数，根据多个所述室内环境参数构建室内环境参数训练子集，根据多个所述室外环境参数构建室外环境参数训练子集，并根据多个所述设备运行参数构建设备运行参数训练子集。
预设数据连续条件为采集室内干球温度、室内湿球温度、冷冻水供水温度、冷冻水回水温度、室外湿球温度、室外干球温度和太阳辐射强度等参数时，数据不中断，可以完整采集一组数据，之后判断采集的数据是否为属于数据异常或数据缺失，在出现数据异常或数据缺失时，将数据异常或数据缺失对应的一组数据进行筛选，以去除该组数据，最后将筛选符合要求的参数分别构建室内环境参数训练子集、室外环境参数训练子集及设备运行参数训练子集。
将所述初始预测时间神经网络作为预设时间预测模型的步骤之前，获取室内环境测试参数、室外环境测试参数及设备运行测试参数，将室内环境测试参数、室外环境测试参数及设备运行测试参数输入至初始时间预测神经网络中，以获得空调器的启动时长预测值，根据启动时长预测值和预测时长标准值确定启动时长差值，判断启动时长差值是否小于预设时长阈值，在启动时长差值小于预设时长阈值时，执行将初始时间预测神经网络作为预设时间预测模型的步骤。
也就是说，对预设时间预测模型进行检测，判断预设时间预测模型是否符合标准模型条件，在某刻实时采集室内环境测试参数、室外环境测试参数及设备运行测试参数，并获取室内环境测试参数、室外环境测试参数及设备运行测试参数对应的预测时长标准值，之后，将室内环境测试参数和室外环境测试参数及设备运行测试参数输入至初始时间预测神经网络中，以获得空调器的启动时长预测值，并根据启动时长预测值和预测时长标准值确定启动时长差值，假设启动时长差值为10min，预设时长阈值为30min，则启动时长差值10min小于预测时长阈值30min，将训练后的初始神经网络作为预设时间预测模型，若启动时长差值为40min大于预测时间阈值30min，则需要根据采集的预测模型参数重新训练初始神经网络。
基于人工神经网络模型建立中央空调的最优启动时间预测模型即预设时间预测模型，由于建筑内的温度受室外温湿度、太阳辐射、室内设备、照明、人员的散热以及维护结构蓄热的影响，使得整个建筑成为一个非常复杂的热力系统，很难用精确的数学模型来描述建筑内温度的变化过程。因此，空调系统最优启动时间也无法用普通函数关系式来表示。人工神经网络算法具有非线性、适应性信息处理能力，弥补了传统算法存在的缺陷，能无限逼近任意多变量非线性关系，因此利用神经网络算法预测空调系统的最优启动时间具有很强的可行性。
人工神经网络从结构上分为输入层、隐含层和输出层，有时候还特别设置结构层，该层可以把隐含层上一个时刻中的信息储存起来，结构层与输入层按照一定的算法及连接权值的修正传递到隐含层，经隐含层处理结束后将此时的输出反馈传播到结构层，并记忆到下一训练时刻才释放该单元值。人工神经网络通过增加内部反馈信号提升了动态辨识的处理能力，适用于解决空调系统的动态仿真预测的问题。
预设时间预测模型的构建步骤可以划分为：
步骤一：从预测模型参数的历史数据库中筛选满足条件的预测模型参数，并进行数据清洗，去除零值、异常值、缺失值等数据；
步骤二：输入输出参数筛选，其中输入参数包括：t0时刻室外干球温度，t0时刻室外湿球温度，t0时刻室内干球温度，t0时刻室内湿球温度，t0时刻冷冻水供水温度，t0时刻冷冻水回水温度，t0时太阳辐射强度。t0时刻是每天主机开机前的某一时刻，例如6:00。输出参数为中央空调系统最优启动时间tpre：tpre = t2-t1，其中t1为中央空调实际开始制冷的时刻，t2为室内温度达到设定值的时刻。在过渡季节，有时早上中央空调系统即使不开启，室内温度也能达到设定值，此时中央空调系统最优启动时间tpre默认为0。
步骤三：人工神经网络模型根据输入输出参数进行训练，对模型中的权重系数进行自动辨识。为了保证预测时间的准确性，神经网络模型每隔一段时间根据新增历史数据重新自适应训练，对模型权重系数进行更新。
步骤四：在完成最优启动时间预测模型训练的基础上，建筑自动化系统每天在t0时刻将输入参数发送给预测模型，获取最优启动时间预测值tpre，并根据最优启动时间tpre和上班时刻t3（建筑自动化系统上设定）计算开机时刻t4：t4=t3-tpre。建筑自动化系统判断当前时刻是否到达开机时刻t4，若未到达则等待，若到达开机时刻t4，则向冷水机组发送开机指令，也就是说判断人员上班时间与当前时间之差是否小于等于预测时长即空调开机时刻，如小于等于预测时长则发送开机指令，否则不动作等待。
根据所述当前室内环境参数、所述当前室外环境参数及所述当前设备运行参数，通过预设时间预测模型进行时间预测，以确定所述空调器的目标预启动时间的步骤，将当前室内环境参数和当前室外关键参数输入至预设时间预测模型，以获得空调器的启动时长编码，对启动时长编码进行处理，获得空调器的启动时长，获取室内舒适温度对应的期许时间，根据启动时长和所述期许时间确定所述空调器的目标预启动时间。
对所述启动时长编码进行处理，获得所述空调器的启动时长的步骤，对启动时长编码进行处理，以获取对应的时间浮点，将时间浮点与预设时间映射关系表中的样本时间浮点进行匹配，若匹配成功，则将匹配成功的样本时间浮点对应的启动时长样本作为空调器的启动时长。
对启动时间编码进行处理，以获取对应的时间浮点的步骤之前，获取多个样本时间浮点，根据多个样本时间浮点分别确定对应的样本启动时长，根据多个样本时间浮点和样本启动时长建立预设时间映射关系表。
也就是说，随着云计算技术的发展，越来越多的建筑自动化系统开始接入共有云或私有云，将楼宇系统的数据上传至云计算平台，进行可视化的分析展示，从而有效提升了楼宇的能源管理水平、节约了运维成本。因此，为实现云计算平台最优启动时间预测控制，本实施例在本地设置智能楼宇管理系统，该系统不仅具备常规建筑自动化系统的功能，还能通过特定接口与云计算平台进行实时数据交互，同时利用云技术将中央空调系统最优启动时间预测模型部署在云计算平台，建立智能楼宇管理系统远程调用服务。智能楼宇管理系统定时发送数据给云端预测模型，经模型计算后返回预测时间值，实现对中央空调系统的开机时间的动态优化。其中，云计算平台远程最优启动时间预测控制步骤如下：
（1）在上位机部署智能楼宇管理系统、云计算平台部署最优启动时间预测模型，建立云计算平台与智能楼宇管理系统之间的通讯连接及接口调用服务；
（2）智能楼宇管理系统每天在开机前某时刻（例如6:00）向云计算平台发送数据请求，并将该时刻的室内温湿度、室外温湿度、太阳辐射强度、冷水机组供回水温度等参数发送给云计算平台预测模型，数据处理成特定的编码；
云计算平台模型经过计算后，向智能楼宇管理系统返回包含最优启动时间的编码，智能楼宇管理系统将编码自动解析成时间值，并修改当天的中央空调系统开机时间设定值；
（3）当时钟到达开机时间设定值，智能楼宇管理系统向中央空调系统发出开机指令，完成最优启动时间预测控制。
步骤S30：控制所述空调器按照所述目标预启动时间实现定时启动。
需要说明的是，将当前室内环境参数、当前室外关键参数及当前设备运行参数输入至预设时间预测模型，以获得空调器的启动时长编码，对启动时长编码进行处理，获得空调器的启动时长，获取室内舒适温度对应的期许时间，根据启动时长和所述期许时间确定所述空调器的目标预启动时间，之后根据目标预启动时间控制空调器实现定时启动。
本实施例，首先获取空调器所处的当前室内环境参数、当前室外环境参数及当前设备运行参数，然后根据当前室内环境参数、当前室外环境参数及当前设备运行参数，通过预设时间预测模型进行时间预测，以确定空调器的目标预启动时间，最后控制空调器按照目标预启动时间实现定时启动。由于相关技术通过手动控制空调器启动，但空调器定时开启时间过长或过短时，会造成资源浪费，而本实施例根据室内环境参数、室外环境参数及设备运行参数，通过预设时间预测模型进行时间预测确定空调器的目标预启动时间，从而控制空调器按照精准的预启动时间定时启动，进而降低空调器的能耗。
此外，参照图3，图3为基于上述空调的预启动时间控制方法第一实施例，提出本申请空调的预启动时间控制方法第二实施例。
在第二实施例中，空调的预启动时间控制方法中所述步骤S20，包括：
步骤S201：将所述当前室内环境参数、所述当前室外环境参数、所述当前设备运行参数输入至预设时间预测模型，以获得所述空调器的启动时长编码。
当前室内环境参数可以为当前某刻采集的室内干球温度、室内湿球温度等，设备运行参数可以为冷冻水供水温度及冷冻水回水温度等，当前室外环境参数可以为当前某刻采集的室外湿球温度和室外干球温度和太阳辐射强度等，本实施例并不加以限制。
进一步地，将室内干球温度、室内湿球温度、冷冻水供水温度、冷冻水回水温度、室外湿球温度、室外干球温度和太阳辐射强度输入至预设时间预测模型，以获取所述空调器的启动时长编码。
可以理解的是，每组室内干球温度、室内湿球温度、冷冻水供水温度、冷冻水回水温度、室外湿球温度、室外干球温度和太阳辐射强度对应的一组启动时长编码，也就是说每组预测模型参数不同，对应的启动时长编码也不同。
步骤S202：对所述启动时长编码进行处理，获得所述空调器的启动时长。
对启动时长编码进行处理，以获取对应的时间浮点，将时间浮点与预设时间映射关系表中的样本时间浮点进行匹配，若匹配成功，则将匹配成功的样本时间浮点对应的启动时长样本作为空调器的启动时长。还可以对启动时长编码进行编码处理，获得对应的时间浮点，之后根据时间浮点按照一定的数据转换规则，以转换成对应的空调器的启动时长等，本实施例并不加以限制。
对启动时间编码进行处理，以获取对应的时间浮点的步骤之前，获取多个样本时间浮点，根据多个样本时间浮点分别确定对应的样本启动时长，根据多个样本时间浮点和样本启动时长建立预设时间映射关系表。
启动时长可以理解为空调器的预先制冷或制热的启动时长。
步骤S203：获取室内舒适温度对应的期许时间。
期许时间可以为用户需要在某刻时间使室内温度达到舒适温度的时刻，可以为用户上班时间8:00，也可以为休息时间17:00等，本实施例并不加以限制。
步骤S204：根据所述启动时长和所述期许时间确定所述空调器的目标预启动时间。
也就是说，假设启动时长为tpre，期许时间为t2，则空调器的目标预启动时间t1为输出参数为期许时间减去启动时长，也就是t1=
t2-tpre。可以理解的是，在过渡季节有时早上中央空调系统即使不开启，室内温度也能达到设定值，此时中央空调系统最优启动时间tpre默认为0。
在本实施例中，首先将当前室内环境参数、当前室外关键参数及当前设备运行参数输入至预设时间预测模型，以获得空调器的启动时长编码，对启动时长编码进行处理，获得空调器的启动时长，获取室内舒适温度对应的期许时间，根据启动时长和期许时间确定空调器的目标预启动时间，由于相关技术中，手动设置空调器的预启动时间，会导致忘记预先设置预启动时间，而本实施例对空调器的启动时长编码进行处理，获得空调器的启动时长，获取室内舒适温度对应的期许时间，根据启动时长和期许时间确定空调器的目标预启动时间，从而实现自动控制空调器的预启动时间。
此外，本申请实施例还提出一种非易失性存储介质，所述非易失性存储介质上存储有空调的预启动时间控制程序，所述空调的预启动时间控制程序被处理器执行时实现如上文所述的空调的预启动时间控制方法的步骤。
此外，参照图4，本申请实施例还提出一种空调的预启动时间控制装置，所述空调的预启动时间控制装置包括：
采集模块4001，用于获取空调器所处的当前室内环境参数和当前室外环境参数、当前设备运行参数。
需要说明的是，本实施例的执行主体是空调的预启动时间控制设备，其中，该设备是具有数据处理，数据通信及程序运行等功能的空调的预启动时间控制设备，也可为其他设备，本实施例对此不做限制。
空调器可以是中央空调系统，也可以是家用空调器等。
当前室内环境参数可以为当前某刻采集的室内干球温度、室内湿球温度等，当前设备运行参数可以为冷冻水供水温度及冷冻水回水温度等，当前室外环境参数可以为当前某刻采集的室外湿球温度和室外干球温度和太阳辐射强度等，本实施例并不加以限制。
也就是说，在某一时刻同时采集室内干球温度、室内湿球温度、冷冻水供水温度、冷冻水回水温度、室外湿球温度和室外干球温度和太阳辐射强度，并判断上述采集的数据是否异常或缺失，在上述某刻采集的数据出现异常或缺失时，重新采集下一时刻的室内干球温度、室内湿球温度、冷冻水供水温度、冷冻水回水温度、室外湿球温度和室外干球温度和太阳辐射强度，直至某一时刻采集的数据完整。
可以理解为，假设6:00时，室内干球温度为30、室内湿球温度为31、冷冻水供水温度为23、冷冻水回水温度为24、室外湿球温度为0和室外干球温度为1和太阳辐射强度17，则室外湿球温度和室外干球温度为异常数据，6:00采集室内环境参数和室外环境参数无效，需要重新采集室内环境参数和室外环境参数等。
算法模块4002，用于根据所述当前室内环境参数、所述当前室外环境参数及所述当前设备运行参数，通过预设时间预测模型进行时间预测，以确定所述空调器的目标预启动时间，所述预设时间预测模型通过对初始神经网络模型进行训练获得。
获取室内环境参数训练子集、室外环境参数训练子集及设备运行参数训练子集，根据室内环境参数训练子集、室外环境参数训练子集及设备运行参数训练子集构建预测模型参数训练集，预测模型参数训练集包括若干组预测模型参数，获取各组预测模型参数对应的训练时长标准值即空调器启动时长标准值，根据各组预测模型参数和各组预测模型参数对应的训练时长标准值对初始神经网络进行训练，获得初始时间预测神经网络，将初始时间预测神经网络作为预设时间预测模型。
其中，室内环境参数可以为室内干球温度、室内湿球温度、设备运行参数可以为冷冻水供水温度及冷冻水回水温度，室外环境参数可以为室外湿球温度和室外干球温度和太阳辐射强度等，其中一组预测模型参数为室内干球温度、室内湿球温度、冷冻水供水温度、冷冻水回水温度、室外湿球温度、室外干球温度和太阳辐射强度。
获取室内环境参数训练子集、室外环境参数训练子集及设备运行参数训练子集的步骤，在预设时间阈值内采集多个样本室内环境参数、多个样本室外环境参数及多个样本设备运行参数，判断多个所述样本室内环境参数、多个所述样本室外环境参数及多个所述样本设备运行参数是否符合预设数据连续条件，在多个所述样本室内环境参数、多个所述样本室外环境参数及多个所述样本设备运行参数符合所述预设数据连续条件时，分别对多个所述样本室内环境参数、多个所述样本室外环境参数及多个所述样本设备运行参数进行数据筛选，以确定多个室内环境参数、多个室外环境参数及多个设备运行参数，根据多个所述室内环境参数构建室内环境参数训练子集，根据多个所述室外环境参数构建室外环境参数训练子集，并根据多个所述设备运行参数构建设备运行参数训练子集。
预设数据连续条件为采集室内干球温度、室内湿球温度、冷冻水供水温度、冷冻水回水温度、室外湿球温度、室外干球温度和太阳辐射强度等参数时，数据不中断，可以完整采集一组数据，之后判断采集的数据是否为属于数据异常或数据缺失，在出现数据异常或数据缺失时，将数据异常或数据缺失对应的一组数据进行筛选，以去除该组数据，最后将筛选符合要求的参数分别构建室内环境参数训练子集、室外环境参数训练子集及设备运行参数训练子集。
将所述初始预测时间神经网络作为预设时间预测模型的步骤之前，获取室内环境测试参数、室外环境测试参数及设备运行测试参数，将室内环境测试参数、室外环境测试参数及设备运行测试参数输入至初始时间预测神经网络中，以获得空调器的启动时长预测值，根据启动时长预测值和预测时长标准值确定启动时长差值，判断启动时长差值是否小于预设时长阈值，在启动时长差值小于预设时长阈值时，执行将初始时间预测神经网络作为预设时间预测模型的步骤。
也就是说，对预设时间预测模型进行检测，判断预设时间预测模型是否符合标准模型条件，在某刻实时采集室内环境测试参数、室外环境测试参数及设备运行测试参数，并获取室内环境测试参数、室外环境测试参数及设备运行测试参数对应的预测时长标准值，之后，将室内环境测试参数和室外环境测试参数及设备运行测试参数输入至初始时间预测神经网络中，以获得空调器的启动时长预测值，并根据启动时长预测值和预测时长标准值确定启动时长差值，假设启动时长差值为10min，预设时长阈值为30min，则启动时长差值10min小于预测时长阈值30min，将训练后的初始神经网络作为预设时间预测模型，若启动时长差值为40min大于预测时间阈值30min，则需要根据采集的预测模型参数重新训练初始神经网络。
基于人工神经网络模型建立中央空调的最优启动时间预测模型即预设时间预测模型，由于建筑内的温度受室外温湿度、太阳辐射、室内设备、照明、人员的散热以及维护结构蓄热的影响，使得整个建筑成为一个非常复杂的热力系统，很难用精确的数学模型来描述建筑内温度的变化过程。因此，空调系统最优启动时间也无法用普通函数关系式来表示。人工神经网络算法具有非线性、适应性信息处理能力，弥补了传统算法存在的缺陷，能无限逼近任意多变量非线性关系，因此利用神经网络算法预测空调系统的最优启动时间具有很强的可行性。
人工神经网络从结构上分为输入层、隐含层和输出层，有时候还特别设置结构层，该层可以把隐含层上一个时刻中的信息储存起来，结构层与输入层按照一定的算法及连接权值的修正传递到隐含层，经隐含层处理结束后将此时的输出反馈传播到结构层，并记忆到下一训练时刻才释放该单元值。人工神经网络通过增加内部反馈信号提升了动态辨识的处理能力，适用于解决空调系统的动态仿真预测的问题。
预设时间预测模型的构建步骤可以划分为：
步骤一：从预测模型参数的历史数据库中筛选满足条件的预测模型参数，并进行数据清洗，去除零值、异常值、缺失值等数据；
步骤二：输入输出参数筛选，其中输入参数包括：t0时刻室外干球温度，t0时刻室外湿球温度，t0时刻室内干球温度，t0时刻室内湿球温度，t0时刻冷冻水供水温度，t0时刻冷冻水回水温度，t0时太阳辐射强度。t0时刻是每天主机开机前的某一时刻，例如6:00。输出参数为中央空调系统最优启动时间tpre：tpre = t2-t1，其中t1为中央空调实际开始制冷的时刻，t2为室内温度达到设定值的时刻。在过渡季节，有时早上中央空调系统即使不开启，室内温度也能达到设定值，此时中央空调系统最优启动时间tpre默认为0。
步骤三：人工神经网络模型根据输入输出参数进行训练，对模型中的权重系数进行自动辨识。为了保证预测时间的准确性，神经网络模型每隔一段时间根据新增历史数据重新自适应训练，对模型权重系数进行更新。
步骤四：在完成最优启动时间预测模型训练的基础上，建筑自动化系统每天在t0时刻将输入参数发送给预测模型，获取最优启动时间预测值tpre，并根据最优启动时间tpre和上班时刻t3（建筑自动化系统上设定）计算开机时刻t4：t4=t3-tpre。建筑自动化系统判断当前时刻是否到达开机时刻t4，若未到达则等待，若到达开机时刻t4，则向冷水机组发送开机指令，也就是说判断人员上班时间与当前时间之差是否小于等于预测时长即空调开机时刻，如小于等于预测时长则发送开机指令，否则不动作等待。
根据所述当前室内环境参数、所述当前室外环境参数及所述当前设备运行参数，通过预设时间预测模型进行时间预测，以确定所述空调器的目标预启动时间的步骤，将当前室内环境参数和当前室外关键参数输入至预设时间预测模型，以获得空调器的启动时长编码，对启动时长编码进行处理，获得空调器的启动时长，获取室内舒适温度对应的期许时间，根据启动时长和所述期许时间确定所述空调器的目标预启动时间。
对所述启动时长编码进行处理，获得所述空调器的启动时长的步骤，对启动时长编码进行处理，以获取对应的时间浮点，将时间浮点与预设时间映射关系表中的样本时间浮点进行匹配，若匹配成功，则将匹配成功的样本时间浮点对应的启动时长样本作为空调器的启动时长。
对启动时间编码进行处理，以获取对应的时间浮点的步骤之前，获取多个样本时间浮点，根据多个样本时间浮点分别确定对应的样本启动时长，根据多个样本时间浮点和样本启动时长建立预设时间映射关系表。
也就是说，随着云计算技术的发展，越来越多的建筑自动化系统开始接入共有云或私有云，将楼宇系统的数据上传至云计算平台，进行可视化的分析展示，从而有效提升了楼宇的能源管理水平、节约了运维成本。因此，为实现云计算平台最优启动时间预测控制，本实施例在本地设置智能楼宇管理系统，该系统不仅具备常规建筑自动化系统的功能，还能通过特定接口与云计算平台进行实时数据交互，同时利用云技术将中央空调系统最优启动时间预测模型部署在云计算平台，建立智能楼宇管理系统远程调用服务。智能楼宇管理系统定时发送数据给云端预测模型，经模型计算后返回预测时间值，实现对中央空调系统的开机时间的动态优化。其中，云计算平台远程最优启动时间预测控制步骤如下：
（1）在上位机部署智能楼宇管理系统、云计算平台部署最优启动时间预测模型，建立云计算平台与智能楼宇管理系统之间的通讯连接及接口调用服务；
（2）智能楼宇管理系统每天在开机前某时刻（例如6:00）向云计算平台发送数据请求，并将该时刻的室内温湿度、室外温湿度、太阳辐射强度、冷水机组供回水温度等参数发送给云计算平台预测模型，数据处理成特定的编码；
云计算平台模型经过计算后，向智能楼宇管理系统返回包含最优启动时间的编码，智能楼宇管理系统将编码自动解析成时间值，并修改当天的中央空调系统开机时间设定值；
（3）当时钟到达开机时间设定值，智能楼宇管理系统向中央空调系统发出开机指令，完成最优启动时间预测控制。
控制模块4003，用于控制所述空调器按照所述目标预启动时间实现定时启动。
需要说明的是，将当前室内环境参数、当前室外关键参数及当前设备运行参数输入至预设时间预测模型，以获得空调器的启动时长编码，对启动时长编码进行处理，获得空调器的启动时长，获取室内舒适温度对应的期许时间，根据启动时长和所述期许时间确定所述空调器的目标预启动时间，之后根据目标预启动时间控制空调器实现定时启动。
本实施例，首先获取空调器所处的当前室内环境参数、当前室外环境参数及当前设备运行参数，然后根据当前室内环境参数、当前室外环境参数及当前设备运行参数，通过预设时间预测模型进行时间预测，以确定空调器的目标预启动时间，最后控制空调器按照目标预启动时间实现定时启动。由于相关技术通过手动控制空调器启动，但空调器定时开启时间过长或过短时，会造成资源浪费，而本实施例根据室内环境参数、室外环境参数及设备运行参数，通过预设时间预测模型进行时间预测确定空调器的目标预启动时间，从而控制空调器按照精准的预启动时间定时启动，进而降低空调器的能耗。
本申请空调的预启动时间控制装置的其他实施例或具体实现方式可参照上述各方法实施例，此处不再赘述。
需要说明的是，在本文中，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者系统不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者系统所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括该要素的过程、方法、物品或者系统中还存在另外的相同要素。
上述本申请实施例序号仅仅为了描述，不代表实施例的优劣。在列举了若干装置的单元权利要求中，这些装置中的若干个可以是通过同一个硬件项来具体体现。词语第一、第二、以及第三等的使用不表示任何顺序，可将这些词语解释为名称。
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到上述实施例方法可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件，但很多情况下前者是更佳的实施方式。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个非易失性存储介质(如只读存储器镜像（Read Only Memory image，ROM）/随机存取存储器（Random Access
Memory，RAM）、磁碟、光盘)中，包括若干指令用以使得一台终端设备(可以是手机，计算机，服务器，空调器，或者网络设备等)执行本申请各个实施例所述的方法。
以上仅为本申请的优选实施例，并非因此限制本申请的专利范围，凡是利用本申请说明书及附图内容所作的等效结构或等效流程变换，或直接或间接运用在其他相关的技术领域，均同理包括在本申请的专利保护范围内。
Claims (17)
- 一种空调的预启动时间控制方法，其中，所述方法包括以下步骤：获取空调器所处的当前室内环境参数、当前室外环境参数及当前设备运行参数；根据所述当前室内环境参数、所述当前室外环境参数及所述当前设备运行参数，通过预设时间预测模型进行时间预测，以确定所述空调器的目标预启动时间，所述预设时间预测模型通过对初始神经网络模型进行训练获得；以及控制所述空调器按照所述目标预启动时间实现定时启动。
- 如权利要求1所述的方法，其中，所述获取空调器所处的当前室内环境参数、当前室外环境参数及当前设备运行参数的步骤之前，还包括：获取室内环境参数训练子集、室外环境参数训练子集及设备运行参数训练子集；根据所述室内环境参数训练子集、所述室外环境参数训练子集及所述设备运行参数训练子集构建预测模型参数训练集，所述预测模型参数训练集包括若干组预测模型参数；获取各组预测模型参数对应的训练时长标准值；根据各组预测模型参数和各组预测模型参数对应的训练时长标准值对初始神经网络进行训练，获得初始时间预测神经网络；以及将所述初始时间预测神经网络作为预设时间预测模型。
- 如权利要求2所述的方法，其中，所述获取室内环境参数训练子集、室外环境参数训练子集及设备运行参数训练子集的步骤，包括：在预设时间阈值内采集多个样本室内环境参数、多个样本室外环境参数及多个样本设备运行参数；判断多个所述样本室内环境参数、多个所述样本室外环境参数及多个所述样本设备运行参数是否符合预设数据连续条件；在多个所述样本室内环境参数、多个所述样本室外环境参数及多个所述样本设备运行参数符合所述预设数据连续条件时，分别对多个所述样本室内环境参数、多个所述样本室外环境参数及多个所述样本设备运行参数进行数据筛选，以确定多个室内环境参数、多个室外环境参数及多个设备运行参数；以及根据多个所述室内环境参数构建室内环境参数训练子集，根据多个所述室外环境参数构建室外环境参数训练子集，并根据多个所述设备运行参数构建设备运行参数训练子集。
- 如权利要求3所述的方法，其中，所述预设数据连续条件为：采集室内干球温度、室内湿球温度、冷冻水供水温度、冷冻水回水温度、室外湿球温度、室外干球温度和太阳辐射强度等参数时，数据不中断且完整采集一组数据。
- 如权利要求3所述的方法，其中，所述分别对多个所述样本室内环境参数、多个所述样本室外环境参数及多个所述样本设备运行参数进行数据筛选，以确定多个室内环境参数、多个室外环境参数及多个设备运行参数的步骤包括：判断采集的数据是否为属于数据异常或数据缺失，若数据异常或数据缺失，将数据异常或数据缺失对应的一组数据进行筛选，以去除该组数据，以及将筛选符合要求的参数分别构建室内环境参数训练子集、室外环境参数训练子集及设备运行参数训练子集。
- 如权利要求2至-3任一项所述的方法，其中，所述将所述初始预测时间神经网络作为预设时间预测模型的步骤之前，还包括：获取室内环境测试参数、室外环境测试参数及设备运行测试参数；将所述室内环境测试参数、所述室外环境测试参数及所述设备运行测试参数输入至所述初始时间预测神经网络中，以获得空调器的启动时长预测值；根据所述启动时长预测值和预测时长标准值确定启动时长差值；判断所述启动时长差值是否小于预设时长阈值；以及在所述启动时长差值小于所述预设时长阈值时，执行所述将所述初始时间预测神经网络作为预设时间预测模型的步骤。
- 如权利要求6所述的方法，其中，所述方法还包括：在所述启动时长差值大于所述预设时长阈值时，根据采集的预测模型参数重新训练初始神经网络的步骤。
- 如权利要求1所述的方法，其中，所述根据所述当前室内环境参数、所述当前室外环境参数及所述当前设备运行参数，通过预设时间预测模型进行时间预测，以确定所述空调器的目标预启动时间的步骤，包括：将所述当前室内环境参数、所述当前室外环境参数、所述当前设备运行参数输入至预设时间预测模型，以获得所述空调器的启动时长编码；对所述启动时长编码进行处理，获得所述空调器的启动时长；获取室内舒适温度对应的期许时间；以及根据所述启动时长和所述期许时间确定所述空调器的目标预启动时间。
- 如权利要求8所述方法，其中，所述对所述启动时长编码进行处理，获得所述空调器的启动时长的步骤，包括：对所述启动时长编码进行处理，以获取对应的时间浮点；将所述时间浮点与预设时间映射关系表中的样本时间浮点进行匹配；以及若匹配成功，则将匹配成功的所述样本时间浮点对应的启动时长样本作为所述空调器的启动时长。
- 如权利要求9所述的方法，其中，所述对所述启动时间编码进行处理，以获取对应的时间浮点的步骤之前，还包括：获取多个样本时间浮点；根据多个所述样本时间浮点分别确定对应的样本启动时长；以及根据多个所述样本时间浮点和所述样本启动时长建立预设时间映射关系表。
- 如权利要求8所述的方法，其中，所述对所述启动时长编码进行处理，获得所述空调器的启动时长的步骤，包括：对启动时长编码进行编码处理，获得对应的时间浮点，根据时间浮点按照一定的数据转换规则，将时间浮点转换成对应的空调器的启动时长。
- 如权利要求1所述的方法，其中，所述当前室内环境参数包括：当前时刻采集的室内干球温度、室内湿球温度；所述当前设备运行参数包括：冷冻水供水温度及冷冻水回水温度；所述当前室外环境参数包括：当前某刻采集的室外湿球温度、室外干球温度以及太阳辐射强度。
- 如权利要求1至9任一项所述的方法，其中，该启动时间控制方法借助于云计算平台控制并预测。
- 一种空调的预启动时间控制装置，其中，所述装置包括：采集模块，用于获取空调器所处的当前室内环境参数、当前室外环境参数及当前设备运行参数；算法模块，用于根据所述当前室内环境参数、所述当前室外环境参数及所述当前设备运行参数，通过预设时间预测模型进行时间预测，以确定所述空调器的目标预启动时间，所述预设时间预测模型通过对初始神经网络模型进行训练获得；以及控制模块，用于控制所述空调器按照所述目标预启动时间实现定时启动。
- 如权利要求14所述的预启动时间控制装置，其中，所述当前室内环境参数包括：当前时刻采集的室内干球温度、室内湿球温度；所述当前设备运行参数包括：冷冻水供水温度及冷冻水回水温度；所述当前室外环境参数包括：当前某刻采集的室外湿球温度、室外干球温度以及太阳辐射强度。
- 一种空调器，其中，所述空调器包括：存储器、处理器及存储在所述存储器上并可在所述处理器上运行的空调的预启动时间控制程序，所述空调的预启动时间控制程序配置为实现如权利要求1至13中任一项所述的空调的预启动时间控制方法的步骤。
- 一种非易失性存储介质，其中，所述非易失性存储介质上存储有空调的预启动时间控制程序，所述空调的预启动时间控制程序被处理器执行时实现如权利要求1至13中任一项所述的空调的预启动时间控制方法的步骤。
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011082502.5A CN112283889A (zh) | 2020-10-10 | 2020-10-10 | 空调的预启动时间控制方法、装置、设备及存储介质 |
| CN202011082502.5 | 2020-10-10 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2022073307A1 true WO2022073307A1 (zh) | 2022-04-14 |
Family
ID=74496100
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2020/141404 WO2022073307A1 (zh) | 2020-10-10 | 2020-12-30 | 空调的预启动时间控制方法、装置、设备及非易失性存储介质 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN112283889A (zh) |
| WO (1) | WO2022073307A1 (zh) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114754477A (zh) * | 2022-04-28 | 2022-07-15 | 广州奥特信息科技股份有限公司 | 空调控制系统、方法、中央空调系统、装置及存储介质 |
| CN114815939A (zh) * | 2022-06-28 | 2022-07-29 | 中成空间（深圳）智能技术有限公司 | 一种微负压气膜实验室内环境质量可控可调的方法及系统 |
| CN115629637A (zh) * | 2022-11-04 | 2023-01-20 | 广州旭杰电子有限公司 | 一种用于5g机房的冷热场温度智能控制系统 |
| CN115758869A (zh) * | 2022-11-07 | 2023-03-07 | 浙江迈新科技股份有限公司 | 一种基于bim的智能楼宇管控系统 |
| CN116085937A (zh) * | 2023-04-11 | 2023-05-09 | 湖南禾自能源科技有限公司 | 智能中央空调节能控制方法及系统 |
| CN117308288A (zh) * | 2023-09-21 | 2023-12-29 | 中国工业互联网研究院 | 一种制丝空调启停预测控制方法以及装置 |
| CN117724353A (zh) * | 2024-01-22 | 2024-03-19 | 江苏谷峰电力科技股份有限公司 | 基于物联网通信的智能家居电力管理系统 |
| WO2024073937A1 (zh) * | 2022-10-08 | 2024-04-11 | 佳都科技集团股份有限公司 | 一种空调控制算法评估方法、装置、设备及存储介质 |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112944620A (zh) * | 2021-02-22 | 2021-06-11 | 珠海格力电器股份有限公司 | 一种空调控制方法、装置、存储介质及空调 |
| CN113283646B (zh) * | 2021-05-14 | 2023-02-03 | 青岛海尔科技有限公司 | 控制设备运行的方法、装置、存储介质及电子装置 |
| CN113339965A (zh) * | 2021-05-19 | 2021-09-03 | 青岛海尔空调器有限总公司 | 用于空调控制的方法、装置和空调 |
| CN113311892B (zh) * | 2021-05-27 | 2022-05-10 | 中通服咨询设计研究院有限公司 | 一种会展场馆中央空调能效最优预冷控制方法 |
| CN113375318A (zh) * | 2021-06-04 | 2021-09-10 | 日立楼宇技术（广州）有限公司 | 一种空调预启动的控制方法、系统、装置及存储介质 |
| CN113516188B (zh) * | 2021-07-16 | 2024-04-12 | 北京上格云智能技术有限公司 | 一种预测空调提前开机时长的方法、装置、设备及介质 |
| CN113449443B (zh) * | 2021-09-01 | 2021-11-30 | 深圳百胜扬工业电子商务平台发展有限公司 | 一种数据加载建模调度方法及装置 |
| CN113739390B (zh) * | 2021-09-30 | 2023-01-24 | 上海美控智慧建筑有限公司 | 空调器控制方法、装置和电子设备 |
| CN114090091A (zh) * | 2021-11-26 | 2022-02-25 | 联影(常州)医疗科技有限公司 | 设备自动开机方法、装置、电子设备及存储介质 |
| CN115796066B (zh) * | 2022-09-06 | 2024-01-19 | 东南大学 | 一种基于预制冷的数据中心空调系统节能方法 |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH09229449A (ja) * | 1996-02-20 | 1997-09-05 | Nippon Metsukusu Kk | 空調装置における最適起動時刻の算定方法 |
| CN103123460A (zh) * | 2011-11-21 | 2013-05-29 | 才秀君 | 温度控制系统和温度控制方法 |
| CN104807135A (zh) * | 2014-01-27 | 2015-07-29 | 海尔集团公司 | 一种基于位置定位的智能空调控制方法及智能空调系统 |
| CN106225172A (zh) * | 2016-08-17 | 2016-12-14 | 珠海格力电器股份有限公司 | 空调控制装置、方法及系统 |
| CN107576002A (zh) * | 2017-08-17 | 2018-01-12 | 青岛海尔空调器有限总公司 | 空调器及其控制方法 |
| CN110736232A (zh) * | 2019-10-29 | 2020-01-31 | 珠海格力电器股份有限公司 | 空调控制方法及装置 |
| CN111649443A (zh) * | 2020-06-17 | 2020-09-11 | 宁波奥克斯电气股份有限公司 | 空调器控制方法、装置、空调器及存储介质 |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR102424689B1 (ko) * | 2015-05-15 | 2022-07-26 | 삼성전자 주식회사 | 공조 기기의 기동 제어 방법 및 장치 |
| CN104913441B (zh) * | 2015-05-29 | 2017-06-27 | 广东美的制冷设备有限公司 | 空调器的温度调节时间预测方法、控制器及空调器 |
| CN111179108A (zh) * | 2018-11-12 | 2020-05-19 | 珠海格力电器股份有限公司 | 用电能耗的预测方法和装置 |
| CN109359727B (zh) * | 2018-12-07 | 2022-01-11 | 北京字节跳动网络技术有限公司 | 神经网络的结构确定方法、装置、设备及可读介质 |
| CN110598336B (zh) * | 2019-09-17 | 2021-08-31 | 美的集团股份有限公司 | 热水器用水量预测方法及装置、热水器及电子设备 |
| CN110686382B (zh) * | 2019-10-16 | 2021-02-09 | 广东美的暖通设备有限公司 | 空调控制方法、装置及计算机可读存储介质 |
-
2020
- 2020-10-10 CN CN202011082502.5A patent/CN112283889A/zh active Pending
- 2020-12-30 WO PCT/CN2020/141404 patent/WO2022073307A1/zh active Application Filing
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH09229449A (ja) * | 1996-02-20 | 1997-09-05 | Nippon Metsukusu Kk | 空調装置における最適起動時刻の算定方法 |
| CN103123460A (zh) * | 2011-11-21 | 2013-05-29 | 才秀君 | 温度控制系统和温度控制方法 |
| CN104807135A (zh) * | 2014-01-27 | 2015-07-29 | 海尔集团公司 | 一种基于位置定位的智能空调控制方法及智能空调系统 |
| CN106225172A (zh) * | 2016-08-17 | 2016-12-14 | 珠海格力电器股份有限公司 | 空调控制装置、方法及系统 |
| CN107576002A (zh) * | 2017-08-17 | 2018-01-12 | 青岛海尔空调器有限总公司 | 空调器及其控制方法 |
| CN110736232A (zh) * | 2019-10-29 | 2020-01-31 | 珠海格力电器股份有限公司 | 空调控制方法及装置 |
| CN111649443A (zh) * | 2020-06-17 | 2020-09-11 | 宁波奥克斯电气股份有限公司 | 空调器控制方法、装置、空调器及存储介质 |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114754477A (zh) * | 2022-04-28 | 2022-07-15 | 广州奥特信息科技股份有限公司 | 空调控制系统、方法、中央空调系统、装置及存储介质 |
| CN114754477B (zh) * | 2022-04-28 | 2023-11-28 | 广州奥特信息科技股份有限公司 | 空调控制系统、方法、中央空调系统、装置及存储介质 |
| CN114815939A (zh) * | 2022-06-28 | 2022-07-29 | 中成空间（深圳）智能技术有限公司 | 一种微负压气膜实验室内环境质量可控可调的方法及系统 |
| WO2024073937A1 (zh) * | 2022-10-08 | 2024-04-11 | 佳都科技集团股份有限公司 | 一种空调控制算法评估方法、装置、设备及存储介质 |
| CN115629637A (zh) * | 2022-11-04 | 2023-01-20 | 广州旭杰电子有限公司 | 一种用于5g机房的冷热场温度智能控制系统 |
| CN115758869A (zh) * | 2022-11-07 | 2023-03-07 | 浙江迈新科技股份有限公司 | 一种基于bim的智能楼宇管控系统 |
| CN116085937A (zh) * | 2023-04-11 | 2023-05-09 | 湖南禾自能源科技有限公司 | 智能中央空调节能控制方法及系统 |
| CN117308288A (zh) * | 2023-09-21 | 2023-12-29 | 中国工业互联网研究院 | 一种制丝空调启停预测控制方法以及装置 |
| CN117308288B (zh) * | 2023-09-21 | 2024-04-19 | 中国工业互联网研究院 | 一种制丝空调启停预测控制方法以及装置 |
| CN117724353A (zh) * | 2024-01-22 | 2024-03-19 | 江苏谷峰电力科技股份有限公司 | 基于物联网通信的智能家居电力管理系统 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112283889A (zh) | 2021-01-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2022073307A1 (zh) | 空调的预启动时间控制方法、装置、设备及非易失性存储介质 | |
| CN112577161B (zh) | 空调能耗模型训练方法与空调系统控制方法 | |
| CN110990159B (zh) | 一种基于历史数据分析的容器云平台资源配额预测方法 | |
| JP6807556B2 (ja) | 空調制御方法、空調制御装置及び空調制御プログラム | |
| WO2023093820A1 (zh) | 设备的控制优化方法、展示平台、云服务器及存储介质 | |
| WO2020103677A1 (zh) | 数值天气预报的气象要素数据处理方法及装置 | |
| CN111623497A (zh) | 一种辐射空调预冷预热方法、系统、存储介质及辐射空调 | |
| CN113325762B (zh) | 一种智能建筑个性化用能控制方法、系统、装置和设备 | |
| WO2019063079A1 (en) | SYSTEM, DEVICE AND METHOD FOR OPTIMIZING ENERGY AND COMFORT IN AN IMMOTIC ENVIRONMENT | |
| KR20180072468A (ko) | 빅데이터기반 사용자의 지능형 냉방 및 난방 서비스 제공 방법 | |
| CN115220351B (zh) | 一种基于云边端的建筑空调系统智能节能优化控制方法 | |
| CN114923265B (zh) | 一种基于物联网的中央空调节能控制系统 | |
| KR101727434B1 (ko) | 랜덤 포레스트 모델을 이용한 냉동기의 성능 판단 방법 | |
| KR102369350B1 (ko) | 인공신경망을 이용한 공기조화설비의 에너지 사용량 및 데이터 예측 시스템 및 그 방법 | |
| CN114897241A (zh) | 一种基于数字孪生的楼宇能效智能化监管预测方法 | |
| Seeam et al. | Evaluating the potential of simulation assisted energy management systems: A case for electrical heating optimisation | |
| CN111339641A (zh) | 制冷系统管理方法、装置、云平台及存储介质 | |
| Huang et al. | A sequential modelling approach for indoor temperature prediction and heating control in smart buildings | |
| Rizwan et al. | Optimal environment control mechanism based on OCF connectivity for efficient energy consumption in greenhouse | |
| CN116928984B (zh) | 一种恒温恒湿冷库内外双循环温湿调控系统及调控方法 | |
| CN113757922B (zh) | 一种基于深度学习的空调系统节能控制方法、装置、设备及计算机介质 | |
| CN116430748A (zh) | 一种基于智能设备的节能控制方法及系统 | |
| CN114818059B (zh) | 建筑节能策略优化控制方法、装置、设备及可读存储介质 | |
| CN113341792B (zh) | 一种基于bim的建筑数据采集方法及系统 | |
| US20220236704A1 (en) | Control system, server, apparatus and control method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 20956622 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 20956622 Country of ref document: EP Kind code of ref document: A1 |
|
| 32PN | Ep: public notification in the ep bulletin as address of the adressee cannot be established |
Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 02/10/2023) |