Air conditioning control method and device, electrical equipment, and medium

A predictive deep neural network model optimizes air conditioning operation periods in base stations based on outdoor temperature and load, addressing inefficient power consumption and overheating issues.

JP7583072B2Active Publication Date: 2024-11-13ZTE CORP
View PDF 7 Cites 0 Cited by

Patent Information

Application Number
JP2022580480
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Priority Date
2020-06-28
Filing Date
2021-06-28
Publication Date
2024-11-13
Estimated Expiration
2041-06-28

AI Technical Summary

Technical Problem

Existing air conditioning control methods for base stations struggle to determine optimal startup and shutdown temperatures, leading to inefficient power consumption and potential overheating due to frequent activation and deactivation.

Method used

A predictive method using a deep neural network model to determine a predicted operation period based on outdoor temperature and indoor load, combined with additional rules, to control air conditioning efficiently and reduce power consumption while preventing overheating.

Benefits of technology

The method effectively reduces air conditioning power consumption and prevents overheating by optimizing operation periods, ensuring efficient energy use and equipment safety.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007583072000003
    Figure 0007583072000003
  • Figure 0007583072000004
    Figure 0007583072000004
  • Figure 0007583072000005
    Figure 0007583072000005
Patent Text Reader

Abstract

The present disclosure provides an air conditioning control method that includes determining a predicted outdoor temperature for a future target time, determining a predicted operation period for the air conditioning at the target time based on the target time and the predicted outdoor temperature, the predicted operation period being an operation period for adjusting the total power consumption of the air conditioning to a second predetermined standard if the indoor temperature does not exceed a first predetermined standard, and controlling the air conditioning at the target time based on the predicted operation period. Embodiments of the present disclosure also provide an air conditioning control device, an electronic device, and a computer-readable medium.
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The embodiments disclosed herein relate to the field of communications technology, and in particular to an air conditioning control method and device, an electronic device, and a computer-readable medium. [Background technology]

[0002] In a communication network, about 80% of the power consumption is in base stations which are widely distributed, and the air conditioning power consumption of base stations accounts for 46% of the total power consumption of base stations.

[0003] Start-up temperature and stop temperature control By installing the air conditioning of the base station, the start-up temperature and stop temperature can be freely set and adjusted, but it is often difficult to determine truly reasonable start-up and stop temperatures. Inappropriate start-up and stop temperatures may result in frequent start-up and stop of the air conditioning, and conversely, may result in high power consumption of the air conditioning of the base station. Summary of the Invention

[0004] The disclosed embodiment comprises: determining a predicted outdoor temperature for a future target time; determining a predicted operation period of the air conditioning during the target time based on the target time and the predicted outdoor temperature, the predicted operation period being an operation period for causing the total power consumption of the air conditioning to conform to a second predetermined standard when the indoor temperature does not exceed a first predetermined standard; Controlling the air conditioning based on the predicted operation period at the target time. A method for controlling air conditioning is provided.

[0005] The disclosed embodiment comprises: a determination module arranged to determine a predicted outdoor temperature for a future target time; a prediction module that determines a predicted operation period of the air conditioning at the target time based on the target time and the predicted outdoor temperature, and arranges the predicted operation period to be an operation period that makes the total power consumption of the air conditioning conform to a second predetermined standard when the indoor temperature does not exceed a first predetermined standard; and a control module arranged to control the air conditioning based on the predicted operation period at the target time. A climate control device is further provided.

[0006] The disclosed embodiment comprises: one or more processors; A memory having one or more programs stored therein, the one or more processors executing the one or more programs causing the one or more processors to realize the air conditioning control method disclosed herein; Electronic equipment is further provided.

[0007] The disclosed embodiment comprises: A computer program is stored in the computer, and when the computer program is executed by a processor, the processor realizes the air conditioning control method disclosed herein. A computer readable medium is further provided. [Brief description of the drawings]

[0008] About the drawings published in this publication [Figure 1] FIG. 2 is a flow diagram of an air conditioning control method provided by the present disclosure. [Diagram 2] FIG. 4 is another flow chart of the air conditioning control method provided by the present disclosure. [Diagram 3] FIG. 1 is a schematic diagram of the logical structure of a deep neural network used in the air conditioning control method provided by the present disclosure. [Figure 4] 1 is a configuration block diagram of an air conditioning control device provided by the present disclosure. [Diagram 5] FIG. 1 is a configuration block diagram of an electronic device provided by the present disclosure. [Figure 6] FIG. 1 is a block diagram of a computer-readable medium provided by the present disclosure. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] In order to enable those skilled in the art to better understand the technical solutions of the embodiments of the present disclosure, the air conditioning control method and device, electronic device, and computer-readable medium provided in the present disclosure will be described in detail below in combination with the drawings.

[0010] Hereinafter, the embodiments of the present disclosure will be described more fully with reference to the drawings, but the embodiments may be embodied in different forms and should not be construed as being limited to the embodiments of the present disclosure. On the contrary, the purpose of providing these embodiments is to make the present disclosure more thorough and complete, and to allow those skilled in the art to fully understand the scope of the present disclosure.

[0011] The drawings of the embodiments of the present disclosure are for providing a further understanding of the embodiments of the present disclosure, constitute a part of the specification, are used to explain the present disclosure together with the embodiments of the present disclosure, and do not constitute limitations on the present disclosure. The above and other features and advantages will become more apparent to those skilled in the art by describing the detailed exemplary embodiments with reference to the drawings.

[0012] The embodiments of the present disclosure can be described with reference to plan views and / or cross-sectional views using ideal schematic diagrams of the present disclosure. Therefore, the exemplary drawings can be modified based on manufacturing techniques and / or limitations.

[0013] Where not in conflict, each embodiment and each feature in the embodiments of the present disclosure may be combined with each other.

[0014] The terms used in this disclosure are used only to describe certain embodiments and are not intended to limit this disclosure. The term "and / or" as used in this disclosure includes any and all combinations of one or more of the associated listed items. As used in this disclosure, the singular forms "a" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. As used in this disclosure, the terms "comprising" and "comprising" specify the presence of features, wholes, steps, operations, elements, and / or assemblies, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, assemblies, and / or groups thereof.

[0015] Unless otherwise specified, the meaning of all terms (including technical and scientific terms) used in this disclosure is the same as that commonly understood by those skilled in the art. It will also be understood that terms as defined in common dictionaries are to be interpreted as having a meaning consistent with the meaning in the context of the relevant art and this disclosure, and are not to be interpreted as having an ideal or excessively formal meaning unless expressly limited by this disclosure.

[0016] The embodiments of the present disclosure are not limited to the embodiments shown in the drawings, and include configuration changes formed based on the manufacturing process. Therefore, the regions illustrated in the drawings have schematic attributes, and the shapes of the regions illustrated in the drawings are illustrative of the specific shapes of the elements, but are not intended to be limiting.

[0017] In the related art, a temperature control on / off method can be used to control the air conditioning of a base station. Specifically, in the temperature control on / off method, the temperature parameters for on / off of the air conditioning are set based on artificial experience, for example, 35°C / 25°C, that is, when the indoor temperature (room temperature) of the base station exceeds 35°C, the air conditioning is controlled to be on, and when it is below 25°C, the air conditioning is controlled to be off.

[0018] However, all of the above temperature parameters are "bidirectional", that is, they should not be too high or too low. For example, if the start-up temperature is set too high, the air conditioning may not start up in a timely manner, which may cause the base station equipment to overheat and cause accidents. On the other hand, if the start-up temperature is set too low, the air conditioning will start up too frequently, which will increase unnecessary power consumption.

[0019] Although the temperature control start / stop method is simple, it is often difficult to implement in practice because it controls the air conditioning based only on the indoor temperature of the base station and does not take other factors into consideration. That is, it is not possible to determine the start / stop temperature parameters for air conditioning start / stop, such as whether 35°C / 25°C, 33°C / 23°C, or 37°C / 23°C is preferable.

[0020] For example, if the indoor temperature of a certain base station is below 35°C for a long time, but at a certain point in time, the peak of business and the peak of temperature overlap, and the indoor temperature temporarily exceeds 35°C, the air conditioning will be turned on. However, in reality, the indoor temperature will drop in a short time without turning on the air conditioning, so there is no need to turn on the air conditioning at this time. This is because the operating temperature range of the base station equipment is 40°C for a long time and can reach 50°C in a short time. On the other hand, if we only use the indoor temperature as a criterion, it is impossible to determine whether the indoor temperature exceeds 35°C for a long time or only for a short time, so it is impossible to determine whether the parameter value of 35°C is reasonable.

[0021] FIG. 1 is a flow diagram of an air conditioning control method provided by this disclosure.

[0022] Referring to FIG. 1, the air-conditioning control method disclosed herein includes steps S101 to S103.

[0023] In step S101, a predicted outdoor temperature at a future target time is determined.

[0024] For example, a UME (Unified Management Expert), which is a unified management expert of a cloud management system, can determine a future time or period (e.g., "one day") when air conditioning needs to be controlled as a target time, and obtain the predicted outdoor temperature at the base station location at the target time.

[0025] Because the target time is a future time (or time period), the predicted outdoor temperature needs to be obtained using a prediction method such as a weather forecast, and this step obtains the outdoor temperature (i.e., the environmental temperature) at the time when it is desired to control the air conditioning in the future.

[0026] In step S102, a predicted operation period of the air conditioning at the target time is determined based on the target time and the predicted outdoor temperature.

[0027] The predicted operation period is an operation period during which the total power consumption of the air conditioning is adapted to a second predetermined standard when the indoor temperature does not exceed a first predetermined standard.

[0028] For example, the UME can determine a predicted operation period of the air conditioning of the base station in the target time based on the target time and the acquired preset outdoor temperature. That is, when determining which period the air conditioning will operate based on the target time and the preset outdoor temperature, the UME prediction can ensure that the indoor temperature of the base station does not exceed a first predetermined standard, and also ensure that the total power consumption of the air conditioning during the target time meets a second predetermined standard.

[0029] If the indoor temperature of the base station meets the above-mentioned "first predetermined criterion", it can ensure that the equipment of the base station does not overheat, that is, the "first predetermined criterion" can ensure that the equipment of the base station does not overheat. Of course, it is feasible if the "first predetermined criterion" can also make the equipment of the base station better avoid overheating (for example, if there is a certain margin before overheating occurs).

[0030] The "second predetermined standard" means a standard that can make the total power consumption of the air conditioner relatively low. For example, the "second predetermined standard" may be a standard that guarantees the total power consumption of the air conditioner to be "minimum" when the indoor temperature does not exceed the first predetermined standard, or the "second predetermined standard" may be a standard that guarantees the total power consumption of the air conditioner to not exceed a "predetermined value" that is set in advance when the indoor temperature does not exceed the first predetermined standard.

[0031] In other words, the predicted operation period actually embodies the period during which the air conditioning "should" theoretically be operated (activated), or the "preferred operation period" of the air conditioning, and the specific method of displaying this may vary.

[0032] For example, the predicted operation period may include multiple groups (e.g., 12 groups) of "start times" and "operation periods" where the air conditioning should be turned on (powered on) when the start time is reached and turned off (stopped) after the operation period continues.

[0033] Also, the predicted travel period may include multiple interval travel periods, during which the air conditioning should be on, while the time intervals between the travel periods are off periods, during which the air conditioning should be off.

[0034] In step S103, air conditioning is controlled at the target time based on the predicted operation period.

[0035] When the target time is reached, the UME sends the predicted operation period to the air conditioning control device FSU (Field Supervision Unit), so that the UME controls the air conditioning of the base station via the FSU based on the predicted operation period, i.e., the air conditioning of the base station can be kept on during the predicted operation period.

[0036] In the embodiment disclosed herein, the predicted operation period for the target time, i.e., the preferred operation mode of the air conditioning at the target time, is predicted based on a specific future time (target time) and the outdoor temperature at the target time (predicted outdoor temperature), and the air conditioning is controlled to be turned on or off at the target time based on the preferred operation mode, thereby reducing power consumption as much as possible while ensuring that the base station equipment does not overheat.

[0037] FIG. 2 is another flow diagram of the air conditioning control method provided by the present disclosure.

[0038] Referring to FIG. 2, the step of determining a predicted outdoor temperature at a future target time (ie, step S101) may include step S011.

[0039] In step S1011, the actual outdoor temperature and the forecast temperature of the weather forecast for the target time are acquired, and the predicted outdoor temperature for the target time is calculated from the actual outdoor temperature and the forecast temperature.

[0040] Based on the actual outdoor temperature of the base station and the forecast temperature at the target time of the weather forecast, the predicted outdoor temperature of the base station at the future target time can be calculated comprehensively. For example, the predicted outdoor temperature can be a weighted average value of the weather forecast temperature and the actual outdoor temperature in the previous hour.

[0041] Referring to FIG. 2, the step of controlling air conditioning based on a predicted operation period at a target time (ie, step S103) may include step S1031.

[0042] In step S1031, at the target time, air conditioning is controlled based on the real-time indoor temperature, the preset additional rules, and the predicted operation period.

[0043] The predicted operation period is a prediction result and cannot absolutely match the actual situation. For example, the actual temperature at the base station location at the target time may be higher than the predicted outdoor temperature, and in this case, if the air conditioning of the base station is controlled based only on the predicted operation period, the base station equipment may overheat and be damaged.

[0044] To reduce the occurrence of such situations, additional rules can be placed on the UME, so that at a target time, the actual operating status of the air conditioning can be constantly "adjusted" based on the base station's real-time indoor temperature and the pre-set additional rules.

[0045] The additional rules may include controlling the air conditioning to be on when the real-time indoor temperature exceeds a preset ultra-high temperature threshold and the air conditioning is not operating, controlling the air conditioning to be off when the real-time indoor temperature is below a preset ultra-low temperature threshold and the air conditioning is operating, and controlling the air conditioning to be operating when the real-time indoor temperature exceeds a preset operating high temperature threshold and is within a predicted operating period.

[0046] When the real-time indoor temperature of the base station is detected to exceed the ultra-high temperature threshold (high set temperature value), it indicates that if the temperature is not lowered promptly, the base station equipment may become abnormal due to overheating. Therefore, if the air conditioning of the base station is not started at this time (for example, it is not during the predicted operation period), it is necessary to forcibly control the start of the air conditioning of the base station to cool the base station equipment and avoid abnormality due to overheating.

[0047] When the real-time indoor temperature of the base station is detected to be lower than the ultra-low temperature threshold (low set temperature value), the temperature of the base station equipment is within a very safe range and there is a high probability that it will not overheat for a long time. Therefore, if the air conditioning of the base station is activated at this time (e.g., during the predicted operation period), the air conditioning of the base station can be forcibly controlled to be turned off to save power.

[0048] Theoretically, the air conditioning should be turned on when it is within the predicted operation period, but if the real-time indoor temperature of the base station is not high (does not exceed the operating high temperature threshold), there is no need to actually turn on the air conditioning. The air conditioning can be controlled to be turned on only when the real-time indoor temperature of the base station exceeds the operating high temperature threshold and is within the predicted operation period.

[0049] Additional rules may also include other parameters such as a minimum stop length, a maximum trip length, etc.

[0050] For example, when the air conditioning is turned on (e.g., entering a forecast operating period), it must be ensured that a minimum amount of downtime (e.g., 0.5 hours) has elapsed since the last time it was turned off, so that the air conditioning is not turned on too frequently.

[0051] Also, if the air conditioning continues to operate for longer than the maximum operating time (for example, 12 hours), the air conditioning can be forcibly turned off to allow it to rest.

[0052] The specific values ​​of the above parameters such as the ultra-high temperature threshold, ultra-low temperature threshold, high temperature threshold, minimum off time length, and maximum operation time can be set as needed (however, it should be ensured that the ultra-high temperature threshold is higher than the high temperature threshold, and the high temperature threshold is higher than the ultra-low temperature threshold). For example, if the equipment of a certain base station is sensitive to temperature, its ultra-high temperature threshold and high temperature threshold should be set to low parameters.

[0053] The number of parameters to be used, such as the ultra-high temperature threshold, ultra-low temperature threshold, high temperature threshold, minimum off time length, and maximum operation time length, as well as the priority relationships between each parameter and rule, can be set as necessary. For example, if the minimum off time length is not met, the air conditioning may not be started even if the temperature exceeds the ultra-high temperature threshold, and if the temperature exceeds the ultra-high temperature threshold, the air conditioning may be started regardless of whether the minimum off time length is met.

[0054] Referring to FIG. 2, the step of determining a predicted operation period of air conditioning at the target time based on the target time and the predicted outdoor temperature (ie, step S102) may include step S1021.

[0055] In step S1021, the target time and the predicted outdoor temperature are input to a preset deep neural network model, and the predicted operation period output by the deep neural network model is obtained.

[0056] In a specific embodiment of this disclosure, a predicted operation period can be obtained using a pre-configured deep neural network (NN) model.

[0057] Specifically, the deep neural network model can be placed on the UME, so that the UME can obtain predicted operation periods based on the deep neural network model, and control the air conditioning by the FSU in combination with additional rules placed inside.

[0058] FIG. 3 is a schematic diagram of the logical structure of a deep neural network used in the air conditioning control method provided in this disclosure.

[0059] Referring to FIG. 3, the deep neural network model may include a first sub-model, a second sub-model and a third sub-model.

[0060] The first sub-model is arranged to determine a predicted load of the base station at a target time and input the predicted load into the second sub-model.

[0061] The second sub-model is configured to determine a predicted indoor temperature of the base station when the air conditioning is not operated based on the predicted load and the predicted outdoor temperature, and input the predicted indoor temperature to the third sub-model.

[0062] A third sub-model is arranged to determine a predicted operation duration based on a predicted indoor temperature and a cooling parameter of the air conditioning.

[0063] The step of inputting the target time and the predicted outdoor temperature into the pre-defined deep neural network model (i.e., step S1021) may include inputting the target time into a first sub-model and inputting the predicted outdoor temperature into a second sub-model.

[0064] The deep neural network model may include three sub-models, and any of the three sub-models may also be a deep neural network model.

[0065] Naturally, the load of a base station may have a certain relationship with time characteristics (i.e., time parameters). For example, whether a particular day (e.g., weekend), belongs to a holiday, busy / slow season, whether there is a local event (e.g., large gathering), etc. will affect the load of the base station. Therefore, after the first sub-model is trained, the predicted load of the base station in the target time (e.g., 1 day) can be predicted based on the actual load of the historical base station and the corresponding time parameters (e.g., holidays, busy / slow seasons, local events, etc.), and the target time and the corresponding time parameters.

[0066] When the air conditioning is not turned on, the indoor temperature of the base station is mainly determined by the load of the base station (related to the heat generation of the base station equipment) and the outdoor temperature. Therefore, after training, the second sub-model can predict the target time with the predicted load and predicted outdoor temperature obtained by the first sub-model, and predict the predicted indoor temperature of the base station when the air conditioning is not turned on.

[0067] Assuming that other conditions remain unchanged, the third sub-model can calculate the cooling effect of turning on the air conditioner at a certain indoor temperature. Therefore, the third sub-model can calculate the indoor temperature of the base station (i.e., the indoor temperature after the air conditioner is turned on) in various air conditioner operation modes (i.e., various air conditioner operation periods) using the predicted indoor temperature obtained by the second sub-model, and determine in which operation mode the power consumption of the air conditioner meets a second predetermined standard (e.g., the power consumption is minimal) and is output as the predicted operation period among various operation modes in which the indoor temperature can meet a first predetermined standard (e.g., not overheating the base station equipment).

[0068] The "cooling parameter of the air conditioner" refers to the actual cooling capacity (or the capacity to lower the indoor temperature) of the air conditioner in the current actual situation of the base station and the air conditioner, which can be expressed in the form of a "cooling efficiency factor" or the like.

[0069] Specifically, the "air conditioning cooling parameters (cooling efficiency factors)" may be determined based on the layout of the base station (e.g., building area, residential type, etc.), the layout of the equipment in the base station (e.g., the type and number of equipment in the base station, the position of the equipment in the base station, etc.), the air conditioning performance (e.g., air conditioning power, model number, parameter settings, etc.), the air conditioning arrangement method (e.g., the position of the air conditioning in the base station, the position of the air conditioning air duct in the base station, etc.), etc., or may be derived theoretically, or may be obtained by actual testing of the base station.

[0070] Naturally, if the location of the base station and the air conditioning does not change, the "air conditioning cooling parameters (cooling efficiency coefficient)" are constant values ​​that do not change, so the air conditioning cooling parameters (cooling efficiency coefficient) in the third sub-model can be set in advance and reset only when the base station is changed or the location of the equipment and air conditioning at the base station changes.

[0071] The above three sub-model divisions do not limit the scope of the disclosed embodiments, but are intended to more accurately derive the predicted operation period, and the deep neural network model disclosed herein may have other different structures.

[0072] Referring to FIG. 2, before the step of inputting the target time and the predicted outdoor temperature into a pre-established deep neural network model (ie, step S1021), step S100 may be included.

[0073] In step S100, a deep neural network model is trained.

[0074] Deep neural network models typically need to be trained before they can be used.

[0075] The basic training process of a deep neural network model is to input training data whose actual results are known into the model, obtain the predicted results output by the model, and based on the difference between the predicted results and the actual results, adjust various parameters in the deep neural network model to gradually optimize the performance of the deep neural network model.

[0076] The training of the deep neural network model may be "temporary" in that the training is not continued after intensive training on a large amount of training data has allowed the deep neural network model to reach a desired performance.

[0077] The deep neural network model may be a continuous one, in which, during the actual use process of the deep neural network model, the deep neural network model is continuously trained or the deep neural network model is constantly optimized based on new data accumulated in the actual use.

[0078] The three sub-models in the deep neural network model disclosed herein are relatively independent, so that the three sub-models can be trained independently, that is, in practical application, the output of the previous sub-model is input to the subsequent sub-model, but in training, the measured data can be directly input to the subsequent sub-model, so that the training process can be more accurate and efficient.

[0079] Specifically, the training process of the deep neural network model disclosed in this disclosure may include the following steps A1 to A8.

[0080] In step S1, a thermal distribution map of the machine room environment, heat generating equipment and air conditioning is constructed through computer simulation, and the cooling parameters (cooling efficiency factor) of the air conditioning are obtained based on it.

[0081] In step S2, a large amount of sample data such as the outdoor temperature, the indoor temperature (without air conditioning), and the load of the base station at different historical times is collected.

[0082] In step S3, an air-conditioning control optimal solution vector (i.e., a preferable operation period of the air-conditioning) is artificially calculated based on the indoor temperature and the cooling parameters of the air-conditioning.

[0083] For example, each air conditioning control optimal solution vector may include multiple sets of air conditioning start times and corresponding start durations.

[0084] In step A4, all sample data are normalized according to the following formula so that each sample data is between 0 and 1.

[0085]

number

[0086] Wherein, X* is the normalized sample data, Xreal is the true value of the sample data, Xmax is the maximum or upper limit value of the sample data, and Xmin is the minimum or lower limit value of the sample data.

[0087] Normalization is just to simplify the data and make it easier to process; it is not a required step.

[0088] In step A5, sample data from different times are divided into a training set, a validation set, and a test set and added.

[0089] The training set is used to train the model (or is used for the first half of training), the validation set is used to verify that the model is trained (or is used for the second half of training), and the test set is used to test the model once training is complete (or is used to test the results of training).

[0090] In step A6, a first sub-model is constructed and trained.

[0091] The load amount for a certain historical time and corresponding time parameters (such as holidays, peak / low seasons, local events, etc.) and the measured waiting time (which is also a historical time) and corresponding time parameters are used as input parameters, and the first sub-model is used to output a predicted load amount for the measured waiting time, and the predicted load amount is compared with the actual load amount for the response time to train the first sub-model.

[0092] In step A7, a second sub-model is constructed and trained.

[0093] The actual outdoor temperature and load amount for a certain historical time period collected periodically are used as input parameters, and the second sub-model is used to output the predicted indoor temperature for that historical time period, and the second sub-model is trained by comparing it with the actual indoor temperature for that historical time period.

[0094] In step A8, a third sub-model is constructed and trained.

[0095] The actual indoor temperature and cooling efficiency factor for a certain historical time period periodically collected are used as input parameters, and the air conditioning control optimal solution vector spaced apart from this historical time period is output as output using the third sub-model, and compared with the air conditioning control optimal solution vector for the corresponding time period obtained in step A3 above, to train the third sub-model.

[0096] The air conditioning control method disclosed herein may specifically include the following steps B01 to B14.

[0097] In step B01, the additional rules are preset.

[0098] The above additional rules are preset based on the experience of normal base station operation.

[0099] For example, if the air conditioning of a base station typically turns on when the indoor temperature exceeds 35°C and turns off when the temperature drops to approximately 25°C, the following parameters can be placed on the FSU or UME:

[0100] 1) The high temperature threshold HT (high temperature) allows the air conditioning to be started when the indoor temperature exceeds this threshold, and the default value is 35°C.

[0101] 2) The very high temperature threshold VHT (very high temperature) is a threshold at which the air conditioning will start unconditionally if the indoor temperature exceeds this threshold, and the default value is 40°C.

[0102] 3) The low temperature threshold LT (low temperature) is a threshold below which the air conditioning can be turned off when the air conditioning is in operation. The default value is 25°C.

[0103] 4) Very low temperature threshold VLT (very low temperature): if the indoor temperature falls below this threshold, the air conditioning will be turned off unconditionally. The default value is 15°C.

[0104] 5) The maximum operation time, MAXCOT (Maximum Continuous operation time), indicates the maximum continuous operation time that the air conditioning system allows. The default value is 12 hours.

[0105] 6) Minimum off time MINCST (Minimum Continuous Shutdown time) indicates the minimum continuous off time that the air conditioning allows, and the default value is 0.5 hours.

[0106] In step B02, data is collected to obtain sample data.

[0107] A large amount of external characteristic parameters such as the machine room outside temperature TRout, the machine room inside temperature TRin, and the load amount LR are collected.

[0108] The collection cycles of the above parameters can be determined based on the rate of change of the general parameters, for example, the collection cycle of TRout is 10 minutes, and the collection cycles of TRin and LR are 5 minutes.

[0109] The above data may be actual measurement data that needs to be recorded by simulating the equipment with a dummy load and collecting real-time data such as TRout, TRin, and LR when TRin is high and air conditioning operation is required.

[0110] The above data can also utilize historical data such as when TRin is low and the air conditioning is turned off for a long time (e.g., during seasons or at night when TRout is low), allowing a large amount of existing historical data to be used.

[0111] In step B03, the data is marked up and a sample tag is obtained.

[0112] Calculate fluid dynamics CFD (Computational Fluid Dynamics) software (e.g., FloTHERM) to build a simulation model of the machine room environment, heat generation equipment, and air conditioning, and obtain the cooling parameters of the air conditioning (cooling efficiency factor).

[0113] The sample data (TRin, LR) is simulated to obtain an optimal solution vector for air conditioning control (i.e., a preferred operation period for air conditioning). For example, the optimal solution vector for air conditioning control includes multiple air conditioning start times Tmoment (hh:mm:ss) and corresponding start times Thours, which can be stored in the big data as corresponding sample tags.

[0114] From the simulation results and daily experience, we can see that the air conditioning should not be started frequently every day. For example, the maximum number of times of starting in a day can be set to 12 times, so if there are two valid values ​​in the Tmoment / Thours tag set, it means that the air conditioning should be started and operated twice, and should be started every time Tmoment is reached and run for the corresponding time length Thours.

[0115] In step B04, all sample data is normalized according to the following formula so that each sample data is between 0 and 1.

[0116]

number

[0117] Wherein, X* is the normalized sample data, Xreal is the true value of the sample data, Xmax is the maximum or upper limit value of the sample data, and Xmin is the minimum or lower limit value of the sample data.

[0118] The sum of Xmax in the normalization can be set as desired.

[0119] For example, in the case of TRout and TRin, Xmax can be determined to have an upper limit of 100°C, and Xmin can be determined to have a lower limit of -40°C.

[0120] Regarding LR, Xmax can be set to the value when the base station is fully loaded, and Xmin can be set to 0.

[0121] In addition, for Tmoment (hh:mm:ss), Xmax can be set to the upper limit of 1440 (1 day 24*60=1440 minutes) and Xmin can be set to 0.

[0122] Furthermore, for the air conditioning startup time length Thours, Xmax can be set to the upper limit value 24 (24 hours per day), and Xmin can be set to 0.

[0123] In step B05, sample data from different times is divided into a training set, a validation set, and a test set and added.

[0124] The number of samples in the training, validation, and test sets can be allocated in a 6:2:2 ratio.

[0125] In step B06, the first sub-model (load forecasting model) is constructed and trained.

[0126] The load amount for a certain historical time and corresponding time parameters (such as holidays, peak / low seasons, local events, etc.) and the measured waiting time (which is also a historical time) and corresponding time parameters are used as input parameters, and the first sub-model is used to output a predicted load amount for the measured waiting time, and the predicted load amount is compared with the actual load amount for the response time to train the first sub-model.

[0127] The specific format of each piece of data may vary.

[0128] For example, the load amount for a historical time period can be the average value of the load amounts for the corresponding time periods.

[0129] In addition, the holiday parameter Fholiday is a characteristic parameter between (0, 1), and for a certain housing complex, it can be determined based on artificial experience, such as 0 for normal weekdays, 0.1 for weekends, and 0.25 for Chinese New Year holidays.

[0130] Furthermore, the peak / low-peak time Ftide is a characteristic parameter between (0, 1), and for a certain industrial park, it can be promised based on artificial experience, such as 0.5 during working hours, 0.4 during overtime hours, and 0.3 during late night hours or non-working days.

[0131] In addition, the regional event parameter Fevent is a characteristic parameter between (0, 1), and for a certain region, it can be determined based on artificial experience, such as normal 0, commercial marketing activity 0.1, rally 0.2, concert 0.3, etc.

[0132] In step B07, a second sub-model (indoor temperature prediction model) is constructed and trained.

[0133] The actual outdoor temperature and load amount for a certain historical time period collected periodically are used as input parameters, and the second sub-model is used to output a predicted indoor temperature for that historical time period, which is then compared with the actual indoor temperature for that historical time period to train the second sub-model.

[0134] In step B08, a third sub-model (air-conditioning control prediction model) is constructed and trained.

[0135] The actual indoor temperature and cooling efficiency factor for a certain historical time period periodically collected are used as input parameters, and the air conditioning control optimal solution vector spaced apart from this historical time period is output as output using the third sub-model, and compared with the air conditioning control optimal solution vector for the corresponding time period obtained in step B03 above, to train the third sub-model.

[0136] The cooling efficiency factor may be a fixed value, such as 0.5.

[0137] Unless the placement of the base station or air conditioning is changed (for example, replacing a new air conditioning, changing the location of the air conditioning duct, replacing some equipment in the base station, etc.), the value of the cooling efficiency coefficient usually does not change.

[0138] As an example, assume that there are two valid values ​​for Tmoment / Thours in an air conditioning control optimal solution vector sample obtained in a certain time, for example, Tmoment 1 is 0.45, Thours 1 is 0.05, Tmoment 2 is 0.60, and Thours 2 is 0.10, and the meaning represented by the air conditioning control optimal solution vector is as follows:

[0139] 1) The air conditioning will be started up twice on the day.

[0140] 2) The first start time is 10:48 (0.45*24=10.8=10:48), and the operation time is 1.2 hours (0.05*24=1.2), that is, the operation period is from 10:48 to 12:00 (0.45*24+0.05*24=12).

[0141] 3) The second start time is 14:24 (0.60*24=14.4=14:24), and the operation time is 2.4 hours (0.10*24=2.4), that is, the operation period is from 14:24 to 16:48 (0.60*24+0.10*24=16.8=16:48).

[0142] In step B09, a deep neural network model is deployed.

[0143] After training and optimization, the deep neural network model is deployed based on the actual operating environment. For example, all three sub-models are deployed in the UME, making full use of the powerful computing resources of the cloud for real-time or online training.

[0144] If necessary, a deep neural network model can be placed on the edge side, such as the FSU, by adding a computation rod, for example.

[0145] In step B10, the FSU collects real-time information such as indoor temperature, outdoor temperature, load, etc., and uploads it to the UME.

[0146] During the actual operation of the base station, the FSU collects various parameters in real time and uploads them to the UME.

[0147] In step B11, the model is executed on the UME, and an optimal solution vector for air conditioning control (predicted operation period) is output.

[0148] Each sub-model operates according to its respective function and outputs a predicted operation period of the air conditioning for a future target time (e.g., 1 day).

[0149] The predicted outdoor temperature input to the second sub-model can be calculated based on the predicted temperature in the weather forecast and the actual outdoor temperature. For example, Predicted outdoor temperature = local weather forecast temperature * 0.8 + actual outdoor temperature one hour ago * 0.2.

[0150] In step B12, an air conditioning control command calculation is performed.

[0151] Based on the above air conditioning control optimal solution vector (predicted operation period), the additional rules previously set, and the real-time indoor temperature of the base station, a control plan for air conditioning is determined.

[0152] For example, the specific flow can be as follows:

[0153] 1) Initialize the air conditioning. The initial state of the air conditioning is off, the length of time the air conditioning is running (Ton) is cleared, and the length of time the air conditioning is off (Toff) is cleared.

[0154] 2) Start timing the length of time the air conditioning is off, Toff.

[0155] 3) Get the current time.

[0156] 4) Get the current real-time indoor temperature (Temp).

[0157] 5) Determine whether to start the high temperature abnormal start flow. If Tmep is greater than VHT and Toff is greater than MINCST, first set the maximum air conditioning operation time Ton-max value to MAXCOT, and then execute the following step 8). If not, proceed to the next step 6).

[0158] 6) Determine whether to initiate a low temperature abnormality shutdown flow. If Tmep is less than VLT, execute step 10) below. If not, proceed to step 7).

[0159] 7) Determine whether to start the high temperature pre-start flow. If the current time reaches Tmoment, Tmep is greater than HT, and Toff is greater than MINCST, set Ton-max to the smaller value of Thours and MAXCOT, and proceed to step 8). If not, return to step 2).

[0160] 8) Air conditioning startup operation flow: Execute the air conditioning startup operation, clear Toff, and start counting Ton.

[0161] 9) Determine whether the air conditioning operation has timed out. If Ton is greater than the Ton-max value, proceed to the next step 10). Get the current time and continue to determine whether the air conditioning operation has timed out.

[0162] 10) Air conditioning off flow: Execute the air conditioning off operation, clear Ton, start timing Toff, and return to step 3).

[0163] In step B13, the UME sends the air-conditioning control command obtained in step B12 to the FSRU, and the FSRU actually controls the air-conditioning on or off based on the air-conditioning control command.

[0164] The FSU can also have a built-in response program for the conventional temperature control start-up / stop method, and when the air conditioning control command from the UME cannot be received in a timely manner (e.g., the communication network is interrupted for a long period of time), the FSU will automatically execute the built-in response program to temporarily control the air conditioning using the conventional temperature control start-up / stop method.

[0165] In step B14, real-time training is performed.

[0166] If the environmental conditions are relatively good (e.g., there is a high-speed Ethernet interconnection between the FSU and the UME and sufficient cloud computing resources) and support real-time or online training, when the air conditioning is not running for a relatively long time (e.g., in cool seasons or on cold nights), the deep neural network model can be continuously trained in real time with newly collected data based on the outdoor temperature, load, indoor temperature, etc. collected in real time to improve the prediction accuracy of the model.

[0167] The above steps B01 to B09 can be performed once in a concentrated manner before the start of the next step. Meanwhile, steps B10, B11, B12, B13, and B14 can be designed as tasks (or processes) that are executed independently, and each task can be executed simultaneously. Step B10 can be operated periodically (for example, the operation period is 5 minutes), step B11 can be executed once every day before midnight and outputs the air conditioning control optimal solution vector for that day, step B12 can be operated in real time, and step B13 can be executed immediately after receiving the sent air conditioning control command. If the actual air conditioning start time and operation time length in step B12 do not match the Tmoment and Thours output by step B11 (for example, the error exceeds 10 minutes), step B11 can be executed again to update the air conditioning control optimal solution vector for that day, thereby improving the accuracy of the prediction.

[0168] FIG. 4 is a configuration block diagram of the air conditioning control device provided in this disclosure.

[0169] Referring to FIG. 4, the air conditioning control device provided in this disclosure includes a determination module, a prediction module, and a control module.

[0170] The determination module is arranged to determine a predicted outdoor temperature for a future target time.

[0171] The prediction module is arranged to determine a predicted operation period at the target time of the air conditioning based on the target time and the predicted outdoor temperature such that the total power consumption of the air conditioning satisfies a second predetermined criterion when the indoor temperature does not exceed a first predetermined criterion.

[0172] The control module is arranged to control the air conditioning at the target time based on the predicted duration of operation.

[0173] In the embodiment disclosed herein, the predicted operation period for the target time, i.e., the preferred operation mode of the air conditioning at the target time, is predicted based on a specific future time (target time) and the outdoor temperature at the target time (predicted outdoor temperature), and the air conditioning is controlled to be turned on or off at the target time based on the preferred operation mode, thereby reducing power consumption as much as possible while ensuring that the base station equipment does not overheat.

[0174] FIG. 5 is a configuration block diagram of the electronic device provided in this disclosure.

[0175] Referring to FIG. 5, the electronic device provided by the present disclosure includes one or more processors and a memory in which one or more programs are stored, the one or more programs being executed by the one or more processors and causing the one or more processors to realize the air conditioning control method according to the present disclosure.

[0176] The electronic device may further include one or more I / O interfaces coupled between the processor and the memory for implementing communication between the processor and the memory.

[0177] The processor is a device having a data processing capability, including but not limited to a central processor (CPU), etc. The memory is a device having a data storage capability, including but not limited to a random access memory (RAM, more specifically SDRAM, DDR, etc.), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), a flash memory (FLASH), etc. The I / O interface (read / write interface), including but not limited to a data bus (Bus), etc., is connected between the processor and the memory to realize the exchange of information between the memory and the processor.

[0178] FIG. 6 is a block diagram of a computer-readable medium provided in this disclosure.

[0179] With reference to FIG. 6, the present disclosure provides a computer-readable medium storing a computer program that, when executed by a processor, causes the processor to implement an air conditioning control method according to the present disclosure.

[0180] Those skilled in the art can understand that all or some of the steps, systems, functional modules / units in the apparatus disclosed above can be implemented as software, firmware, hardware, and appropriate combinations thereof.

[0181] In hardware embodiments, the division between the functional modules / units described above does not necessarily correspond to a division of physical components, e.g. one physical component may have multiple functions, and one function or step may be performed by multiple physical components in cooperation.

[0182] Some or all of the physical components may be implemented as software executed by a processor, such as a central processor (CPU), digital signal processor, or microprocessor, or may be implemented as hardware, or may be implemented as an integrated circuit, such as a dedicated integrated circuit. Such software may be distributed on computer-readable media, which may include computer storage media (or non-transitory media) and communication media (or transitory media). As is well known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information, such as computer-readable instructions, data structures, program modules, or other data. Computer storage media may include, but is not limited to, random access memory (RAM, more specifically SDRAM, DDR, etc.), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory (FLASH) or other disk memory, read-only optical disk (CD-ROM), digital versatile disk (DVD) or other optical disk memory, magnetic cassette, magnetic tape, magnetic disk storage device or other magnetic storage device, and any other medium that stores the desired information and is accessible by a computer. Additionally, communication media typically includes computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transmission mechanism and may include any information delivery media as known to those skilled in the art.

[0183] Although exemplary embodiments are disclosed in this disclosure, and specific terms are used, they are used only in a general descriptive sense and should be interpreted as such only, and not for limiting purposes. It will be apparent to those skilled in the art that in some examples, unless otherwise specified, features, characteristics, and / or elements described in connection with a particular embodiment may be used alone, or features, characteristics, and / or elements described in connection with other embodiments may be used in combination. Thus, those skilled in the art will appreciate that various modifications in form and detail are possible without departing from the scope of this disclosure as defined by the appended claims.

Claims

1. determining a predicted outdoor temperature for a future target time; determining a predicted operation period of the air conditioning during the target time based on the target time and the predicted outdoor temperature, the predicted operation period being an operation period for causing a total power consumption of the air conditioning to conform to a second predetermined standard when the indoor temperature does not exceed a first predetermined standard; controlling the air conditioning based on the predicted operation period at the target time; The step of determining a predicted operation period of the air conditioning at the target time based on the target time and the predicted outdoor temperature includes: inputting the target time and the predicted outdoor temperature into a preset deep neural network model, and obtaining the predicted operation period output by the deep neural network model; the deep neural network model includes a first sub-model, a second sub-model, and a third sub-model; the first sub-model is arranged to determine a predicted load of a base station at the target time and input the predicted load into a second sub-model; the second sub-model is arranged to determine a predicted indoor temperature of the base station when the air conditioning is not operated based on the predicted load and the predicted outdoor temperature, and input the predicted indoor temperature to a third sub-model; the third sub-model is configured to determine the predicted operation duration based on the predicted indoor temperature and a cooling parameter of the air conditioning; An air-conditioning control method, in which the air-conditioning is applied to the base station.

2. The step of controlling the air conditioning based on the predicted operation period at the target time includes: and controlling the air conditioning based on a real-time indoor temperature, a preset additional rule, and the predicted operation period at the target time. The method of claim 1.

3. The additional rules are: In response to the real-time indoor temperature exceeding a preset ultra-high temperature threshold and the air conditioning is not operating, control the air conditioning to be on; In response to the real-time indoor temperature being less than a preset ultra-low temperature threshold and the air conditioning being in operation, control the air conditioning to be off; and controlling the air conditioning to be in operation in response to the real-time indoor temperature exceeding a preset high operating temperature threshold and being within the predicted operation period. The method of claim 2.

4. The step of inputting the target time and the predicted outdoor temperature into a preset deep neural network model includes: inputting the target time into a first sub-model and inputting the predicted outdoor temperature into a second sub-model. The method of claim 1.

5. before inputting the target time and the predicted outdoor temperature into a pre-defined deep neural network model; training the deep neural network model. The method of claim 1.

6. The step of determining a predicted outdoor temperature for a future target time includes: obtaining an actual outdoor temperature and a forecast temperature of a weather forecast for the target time, and calculating a predicted outdoor temperature for the target time based on the actual outdoor temperature and the forecast temperature; The method of claim 1.

7. a determination module arranged to determine a predicted outdoor temperature for a future target time; a prediction module that determines a predicted operation period of the air conditioning at the target time based on the target time and the predicted outdoor temperature, and arranges the predicted operation period to be an operation period that makes the total power consumption of the air conditioning conform to a second predetermined standard when the indoor temperature does not exceed a first predetermined standard; a control module arranged to control the air conditioning based on the predicted operation period at the target time; The prediction module further inputs the target time and the predicted outdoor temperature into a pre-defined deep neural network model, and obtains the predicted operation period output by the deep neural network model; the deep neural network model includes a first sub-model, a second sub-model, and a third sub-model; the first sub-model is arranged to determine a predicted load of a base station at the target time and input the predicted load into a second sub-model; the second sub-model is arranged to determine a predicted indoor temperature of the base station when the air conditioning is not operated based on the predicted load and the predicted outdoor temperature, and input the predicted indoor temperature to a third sub-model; the third sub-model is configured to determine the predicted operation duration based on the predicted indoor temperature and a cooling parameter of the air conditioning; An air conditioning control device in which the air conditioning is applied to the base station.

8. one or more processors; A memory having one or more programs stored therein, the one or more processors executing the one or more programs causing the one or more processors to realize the air conditioning control method according to any one of claims 1 to 6. electronic equipment.

9. A computer program is stored in the computer, and when the computer program is executed by a processor, the processor realizes the air conditioning control method according to any one of claims 1 to 6. Computer-readable medium.

Citation Information

Patent Citations

  • Energy saving method for communication machine room

    CN102821581A

  • Air-conditioning control system

    JP2002089929A

  • Air conditioning control device, cooling system, and air-conditioning control program

    JP2010169388A

  • Air conditioning energy saving control device

    JP2014081123A

  • Air conditioning control method, air conditioning control device and air conditioning control program

    JP2017067427A