A method and device for predicting cleaning time of refrigeration unit

Abstract

The present disclosure relates to the technical field of air conditioning engineering, and provides a method and device for predicting the cleaning time of a refrigeration unit. The method includes obtaining the operating data of the refrigeration unit according to a preset refrigeration cycle; preprocessing the operating data of the refrigeration unit and extracting the time characteristics of the operating data; calculating the daily comprehensive approach temperature true value; taking the time characteristics as input and the comprehensive approach temperature true value as output, training the initial comprehensive approach temperature model to obtain a target comprehensive approach temperature prediction model; based on the predicted time of the next refrigeration cycle, extracting the predicted time characteristics, inputting them into the target comprehensive approach temperature prediction model, and obtaining the comprehensive approach temperature prediction value; determining the cleaning time of the refrigeration unit based on the comprehensive approach temperature prediction value and the comprehensive approach temperature threshold. Therefore, predicting the cleaning time of the refrigeration unit can reduce energy loss, save production costs, and help ensure the long-term operation of the refrigeration unit.

Description

Technical Field

[0001] The present disclosure relates to the technical field of air-conditioning engineering, and in particular to a method and device for predicting the cleaning time of a refrigeration unit. Background Art

[0002] The heat exchanger in the refrigeration unit is an important component of the air conditioning refrigeration system. It uses cooling water to convert high-temperature and high-pressure gas into liquid. As the refrigeration unit runs for a long time, dirt will appear inside the evaporator and condenser. The increase in dirt will increase the heat transfer resistance, reduce the heat transfer efficiency, and affect the overall performance of the unit.

[0003] Currently, whether the dirt in the refrigeration unit needs to be cleaned is mainly judged based on manual experience or regular cleaning. However, due to the large difference in the dirt accumulation rate under different working conditions, excessive dirt accumulation often occurs without cleaning or the cleaning intervals are too frequent, making it impossible to determine the appropriate cleaning time. Summary of the invention

[0004] In view of this, the embodiments of the present disclosure provide a method and device for predicting the cleaning time of a refrigeration unit to solve the problem in the prior art that excessive dirt accumulation is not performed for cleaning or the cleaning intervals are too frequent, making it impossible to determine the appropriate cleaning time.

[0005] According to a first aspect of an embodiment of the present disclosure, a method for predicting a refrigeration unit cleaning time is provided, comprising:

[0006] According to the preset refrigeration cycle, the operation data of the refrigeration unit is obtained, and the operation data at least includes historical approach temperature data, load rate data and on / off state data;

[0007] Preprocess the operating data of the refrigeration unit and extract the time characteristics of the operating data;

[0008] Divide the load rate data into preset load rate intervals and calculate the daily comprehensive approach temperature true value;

[0009] Taking the time feature as input and the real value of the comprehensive approaching temperature as output, the initial comprehensive approaching temperature model is trained to obtain the target comprehensive approaching temperature prediction model;

[0010] Based on the predicted time of the next refrigeration cycle, the time feature is extracted, and the predicted time feature is input into the target comprehensive approach temperature prediction model to obtain the comprehensive approach temperature prediction value;

[0011] The cleaning time of the refrigeration unit is determined based on the comprehensive approach temperature prediction value and the comprehensive approach temperature threshold.

[0012] A second aspect of the embodiments of the present disclosure provides a device for predicting the cleaning time of a refrigeration unit, comprising:

[0013] The data acquisition module is configured to acquire the operation data of the refrigeration unit according to a preset refrigeration cycle, and the operation data at least includes historical approach temperature data, load rate data and on / off state data;

[0014] A feature extraction module is configured to pre-process the operation data of the refrigeration unit and extract the time features of the operation data;

[0015] A data calculation module is configured to divide the load rate data into preset load rate intervals and calculate a daily comprehensive approach temperature true value;

[0016] The training model module is configured to take the time feature as input and the real value of the comprehensive approaching temperature as output, train the initial comprehensive approaching temperature model, and obtain the target comprehensive approaching temperature prediction model;

[0017] A predicted temperature acquisition module is configured to extract predicted time features based on the predicted time of the next refrigeration cycle, and input the predicted time features into a target comprehensive approach temperature prediction model to obtain a predicted comprehensive approach temperature value;

[0018] The time determination module is configured to determine the cleaning time of the refrigeration unit based on the comprehensive approach temperature prediction value and the comprehensive approach temperature threshold.

[0019] According to a third aspect of an embodiment of the present disclosure, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above method when executing the computer program.

[0020] According to a fourth aspect of the embodiments of the present disclosure, a computer-readable storage medium is provided, which stores a computer program, and when the computer program is executed by a processor, the steps of the above method are implemented.

[0021] Compared with the prior art, the embodiments of the present disclosure have at least the following beneficial effects: the present disclosure obtains the operation data of the refrigeration unit according to the preset refrigeration cycle, and the operation data at least includes historical approach temperature data, load rate data and on / off state data; pre-processes the operation data of the refrigeration unit and extracts the time characteristics of the operation data; divides the load rate data into each preset load rate interval, and calculates the daily comprehensive approach temperature true value; takes the time characteristics as input and the comprehensive approach temperature true value as output, trains the initial comprehensive approach temperature model, and obtains the target comprehensive approach temperature prediction model; extracts the predicted time characteristics based on the predicted time of the next refrigeration cycle, and inputs the predicted time characteristics into the target comprehensive approach temperature prediction model to obtain the comprehensive approach temperature prediction value; determines the cleaning time of the refrigeration unit based on the comprehensive approach temperature prediction value and the comprehensive approach temperature threshold. Therefore, the embodiments of the present disclosure use the comprehensive approach temperature prediction model to predict the cleaning time according to the dirt accumulation situation, so as to clean the refrigeration unit in time, which can ensure the heat exchange efficiency of the refrigeration unit, reduce energy loss, save production costs, and help to ensure the long-term operation of the refrigeration unit. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0023] Figure 1 It is a flow chart of a method for predicting the cleaning time of a refrigeration unit provided by an embodiment of the present disclosure;

[0024] Figure 2 It is a flow chart of calculating the real value of the comprehensive approaching temperature in a method for predicting the cleaning time of a refrigeration unit provided by an embodiment of the present disclosure;

[0025] Figure 3 is a schematic diagram of a device for predicting cleaning time of a refrigeration unit provided by an embodiment of the present disclosure;

[0026] Figure 4 It is a structural schematic diagram of an electronic device provided by an embodiment of the present disclosure. DETAILED DESCRIPTION

[0027] In the following description, specific details such as specific system structures and technologies are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present disclosure. However, it should be clear to those skilled in the art that the present disclosure may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid obstructing the description of the present disclosure with unnecessary details.

[0028] A method and device for predicting the cleaning time of a refrigeration unit according to an embodiment of the present disclosure will be described in detail below in conjunction with the accompanying drawings. In this embodiment, there is one refrigeration unit. In practical applications, there may be one or more refrigeration units, and the present invention does not limit this embodiment.

[0029] In the refrigeration system, the refrigeration unit is composed of a compressor, an evaporator, a condenser and an expansion valve, of which the evaporator and the condenser are both heat exchangers. The heat exchanger is used to achieve energy exchange, and the heat in the hot fluid is transferred to the cooling water through the heat exchanger wall to complete the transfer of heat energy. After the refrigeration unit has been running for a period of time, since the cooling water needs to be recycled, it is easy to produce dirt after the water quality deteriorates, and the heat transfer resistance increases, resulting in a significant decrease in heat exchange efficiency, which has an adverse effect on the performance of the refrigeration unit.

[0030] In the prior art, whether it is necessary to clean the dirt in the heat exchanger depends on manual experience or regular cleaning to deal with the accumulated dirt. However, due to limited manual experience and the lack of a quantifiable judgment system, it is impossible to determine when to clean the dirt in the heat exchanger, resulting in cleaning only after a large area of ​​dirt has accumulated or become clogged. Alternatively, due to the lack of experience of the operation and maintenance personnel, cleaning is performed too frequently, resulting in a waste of resources.

[0031] The disclosed embodiment provides a new prediction method for the cleaning time of a refrigeration unit. The prediction method obtains the operation data of the refrigeration unit according to a preset refrigeration cycle, and the operation data at least includes historical approach temperature data. The operation data of the refrigeration unit is preprocessed, the time characteristics of the operation data are extracted, the load rate data is divided into each preset load rate interval, and the daily comprehensive approach temperature true value is calculated. The time characteristics are used as input, and the comprehensive approach temperature true value is used as output. The initial comprehensive approach temperature model is trained to obtain a target comprehensive approach temperature prediction model. Based on the predicted time of the next refrigeration cycle, the predicted time characteristics are extracted, and the predicted time characteristics are input into the target comprehensive approach temperature prediction model to obtain the comprehensive approach temperature prediction value. Based on the comprehensive approach temperature prediction value and the comprehensive approach temperature threshold, the cleaning time of the refrigeration unit is determined. The disclosed embodiment uses the comprehensive approach temperature prediction model to predict the cleaning time to avoid the reduction of heat transfer efficiency due to untimely cleaning of dirt. Based on the provided predicted cleaning time, the operation and maintenance personnel can prepare in advance, clean the dirt according to the predicted time, reduce energy loss, save production and maintenance costs, and optimize the operation efficiency of the refrigeration unit.

[0032] Figure 1 1 is a flow chart of a method for predicting the cleaning time of a refrigeration unit provided by an embodiment of the present disclosure. The method for predicting the cleaning time of a refrigeration unit includes:

[0033] S101, according to a preset refrigeration cycle, obtaining operation data of the refrigeration unit, the operation data at least including historical approach temperature data, load rate data and on / off state data.

[0034] S102, pre-processing the operating data of the refrigeration unit to extract the time characteristics of the operating data.

[0035] S103, dividing the load rate data into preset load rate intervals, and calculating the actual value of the daily comprehensive approach temperature.

[0036] S104, taking the time feature as input and the real value of the comprehensive approaching temperature as output, training the initial comprehensive approaching temperature model to obtain a target comprehensive approaching temperature prediction model.

[0037] S105, based on the predicted time of the next refrigeration cycle, extract the time feature, and input the predicted time feature into the target comprehensive approach temperature prediction model to obtain the comprehensive approach temperature prediction value.

[0038] S106, determining a refrigeration unit cleaning time based on the comprehensive approach temperature prediction value and the comprehensive approach temperature threshold.

[0039] In the above step S101, the time range of data acquisition is at least two complete refrigeration cycles, and a complete refrigeration cycle is usually divided into years. The operation data of the refrigeration unit from June to October in the refrigeration cycle is obtained, and the refrigeration cycle data in other time ranges can also be obtained, which is not specifically limited here.

[0040] The operation data of the refrigeration unit includes at least historical approach temperature data, load rate data and on / off status data.

[0041] Obtain historical approach temperature data. The specific steps include: determining the equipment type of the refrigeration unit, and then inputting the relevant parameters of the refrigeration unit into the calculation formula according to the equipment type to obtain the historical approach temperature. If the equipment is an evaporator, obtain the chilled water outlet temperature t f , refrigerant saturated evaporation temperature t eva , using the approximate temperature calculation formula t = t f -t eva , calculate the historical approach temperature; if the device is a condenser, obtain the refrigerant saturated condensing temperature t cond , Cooling water outlet temperature t c , using the approximate temperature calculation formula t = t cond -t c , and the historical approach temperature is obtained. It should be noted that the units of the historical approach temperatures obtained through the above calculations are all degrees Celsius.

[0042] Secondly, for the load rate of the refrigeration unit, if it can be directly obtained, the load rate of the refrigeration unit can be obtained by calculating the percentage of the average load of the refrigeration unit during the maximum load period of the refrigeration system and the maximum load of the refrigeration unit during this period. For the load rate that cannot be obtained directly, it can be obtained indirectly by other means. For example, for centrifugal refrigeration units, the guide vane opening can be used instead of the load rate, and for spiral refrigeration units, the slide valve position can be used instead of the load rate.

[0043] Finally, obtain the on / off status data of the refrigeration unit. Usually, the start / stop status of the refrigeration unit is displayed as 1 for on and 0 for off. For example, when the time is 5:00-6:00, the operating status of the refrigeration unit is displayed as 1, and the device is in the on state. Based on the obtained on / off status record, the 24-hour operating status information of the refrigeration unit is determined.

[0044] In the above step S102, first, clustering processing is performed on the operation data of the refrigeration unit.

[0045] Specifically, the operation data of the refrigeration unit within the preset time length is obtained, and clustering processing is performed based on the target time granularity within the preset time length. The target time granularity can be set to 1 minute, 5 minutes, 15 minutes, 30 minutes, 1 hour, etc., and is determined according to the specific application situation. In this embodiment, the historical approach temperature data, load rate data and power on / off status data within the preset time length are obtained, the target time granularity is set to 1 hour, and the data are divided according to the target time granularity, and clustering processing is performed respectively. Specifically, for the historical approach temperature data, the average of multiple data within the same hour is taken as the final approach temperature of the hour; for the load rate data, the load rate in percentage form is uniformly converted into a decimal form between 0 and 1, and the average of multiple data within the same hour is taken as the final load rate of the hour; for the power on / off status data, according to the rule that the power on is 1 and the power off is 0, the power on / off status data is converted into a numerical form, and the maximum value of multiple data within the same hour is taken as the final power on / off status of the hour. After the clustering processing is completed, the clustered operation data is obtained.

[0046] Furthermore, the clustered operation data is processed, and the processing method at least includes valid data screening, elimination of abnormal values, merging of identical values, and completion of missing values. Specifically, the valid data screening is first screened according to the power-on / off status value, wherein the power-on / off status value is converted into a numerical form according to the rule that the power-on status value is 1 and the power-off status value is 0. The data with the power-on / off status value of 0 is deleted, that is, only the data with the power-on status value of 1 is retained, and then the data with the load rate less than 0.5 or greater than 0.9 is filtered out.

[0047] Eliminating abnormal values ​​means eliminating values ​​that are too large or too small. In this embodiment, if it is approaching temperature data, values ​​greater than 10 or less than 0 are eliminated. If it is load rate, the load rate in percentage is first uniformly converted into a decimal form between 0 and 1, and values ​​greater than 1 or less than 0 are eliminated. The same values ​​are merged, and the approaching temperature data, load rate data, and the on / off status data of the refrigeration unit are spliced ​​according to the timestamp records for the same period of time.

[0048] The time information of each preprocessed operation data is processed to extract the time features of each operation data, which at least include year, month, day, and week. For example, if the time information is "2021-12-20", the corresponding time information is extracted to obtain four features of "2021", "12", "20" and "52", which represent 2021, December, 20th and the 52nd week respectively.

[0049] In the above step S103, the real value of the comprehensive approach temperature is obtained, which needs to be calculated using the obtained operating data. Figure 2 : is a flow chart of calculating the real value of the comprehensive approaching temperature in a method for predicting the cleaning time of a refrigeration unit provided by an embodiment of the present disclosure, such as Figure 2 As shown, in this embodiment, calculating the comprehensive approach temperature true value includes the following steps:

[0050] S201, obtaining a load rate, dividing the load rate into intervals, and determining the number of load rate intervals.

[0051] According to the preset load rate interval division rule, the load rate is divided into corresponding load rate intervals, and the load rate is divided at intervals of 0.1 or 0.05, and the number of load rate intervals is m. In this embodiment, based on expert experience, the load rate of the refrigeration unit is operated between 0.5-0.9, and the load rate intervals are divided at intervals of 0.1, and the load rate intervals of the refrigeration unit are [0.5, 0.6), [0.6, 0.7), [0.7, 0.8), [0.8, 0.9], and the number of load rate intervals m is 4.

[0052] S202, based on the load rate interval, calculating the data volume within the load rate interval and the total data volume, and then determining the proportion of the data volume within the load rate interval to the total data volume.

[0053] Based on the load rate interval, calculate the amount of data in the load rate interval, and add up the amount of data in each interval to get the total amount of data. [a,b] Divide by the total data volume n, and get the proportion of the data volume in each load rate interval to the total data volume f [a,b] .

[0054] n=n [0.5,0.6) +n [0.6,0.7) +n [0.7,0.8) +n [0.8,0.9]

[0055]

[0056]

[0057]

[0058]

[0059] S203, based on the amount of data in each load factor interval, calculate the average value of the approach temperature in each load factor interval.

[0060] Get the historical approximate temperature data of the i-th data within a certain day's load rate interval Based on the amount of data n in the load rate interval [a,b] [a,b] , the average temperature is calculated by the following formula.

[0061]

[0062] S204, calculating the actual value of the daily comprehensive approach temperature based on the number of each load rate interval, the ratio of the data volume in the interval to the total data volume, and the average approach temperature.

[0063] The historical approach temperature data in each load rate interval on the same day are processed according to the above steps to obtain the average approach temperature in each load rate interval.

[0064] The number of load rate intervals m and the proportion of data in the load rate interval to the total data volume f i , the average temperature approaching in each load rate range The daily comprehensive approach temperature true value t is calculated by weighted summation of the following formula: cp .

[0065]

[0066] The historical approach temperature data of all dates are processed according to the above steps to obtain the actual value of the daily comprehensive approach temperature under the dynamic load rate.

[0067] In the above step S104, based on the time characteristics of the operating data and the real value of the comprehensive approach temperature, a regression model is used to tune the parameters and construct an initial comprehensive approach temperature model.

[0068] y=Model(x 1 ,x 2 ,x 3 ,x 4 ),

[0069] Among them, x 1 ,x 2 ,x 3 ,x 4 represents the time characteristic data year, month, day, and week obtained after data processing, and y represents the comprehensive approach temperature value t obtained after data processing cp .

[0070] The pre-acquired time characteristics of the refrigeration unit are input as samples into the comprehensive approaching temperature prediction model, and the output value is the comprehensive approaching temperature prediction value. According to the comprehensive approaching temperature prediction value and the comprehensive approaching temperature true value, a loss function is constructed.

[0071]

[0072] Among them, k represents the number of training samples, y i represents the comprehensive approach to the true value of the temperature of the i-th sample, represents the comprehensive approach temperature prediction value calculated by the model for the i-th sample.

[0073] The comprehensive approximate temperature prediction model is iteratively trained using the loss function until the preset number of rounds is reached and the loss function reaches the threshold. When the value of the loss function reaches the threshold, the training is terminated to obtain the target comprehensive approximate temperature prediction model.

[0074] In the above step S105, the prediction time of the next refrigeration cycle is set, and the set prediction time is not more than 90 days. The prediction time feature is extracted, and the prediction time feature is input into the target comprehensive approach temperature prediction model to obtain the comprehensive approach temperature prediction value of each day within the prediction time.

[0075] As an example, take the forecast of the comprehensive approach temperature of the refrigeration unit in the next 60 days as an example to forecast the comprehensive approach temperature forecast value of the next refrigeration cycle. Process the date data of 60 days to obtain the time features of year, month, day and week. Set the time feature as x e Input into the trained comprehensive approach temperature prediction model to obtain the daily comprehensive approach temperature prediction value

[0076]

[0077] In the above step S106, the comprehensive approach temperature prediction value is compared with the comprehensive approach temperature threshold, wherein the comprehensive approach temperature threshold is obtained by calculation.

[0078] The steps for calculating the comprehensive approaching temperature threshold are as follows: In this embodiment, first, the load rate is divided into [0.5, 0.6), [0.6, 0.7), [0.7, 0.8) and [0.8, 0.9] with a load rate interval of 0.1, and the number of intervals m is determined to be 4. Then, based on the divided intervals, the proportion of the data volume in the load rate interval to the total data volume is calculated. i The approaching temperature threshold T in each load rate interval i According to expert experience, the approach temperature thresholds of the evaporator are 1.8, 2.1, 2.4 and 2.7, respectively, and the approach temperature thresholds of the condenser corresponding to each interval are 2.4, 2.8, 3.2 and 3.6, respectively.

[0079] The comprehensive approach temperature threshold value T is obtained by the comprehensive approach temperature threshold value calculation formula as shown below.

[0080]

[0081] It is determined in turn whether the daily predicted comprehensive approach temperature value within the prediction time range is greater than the comprehensive approach temperature threshold. If the predicted comprehensive approach temperature value of a certain day is greater than the comprehensive approach temperature threshold, the day is determined to be the latest cleaning time for the refrigeration unit. For example, if the predicted comprehensive approach temperature value is 3.5°C and the comprehensive approach temperature threshold is 3°C, and the predicted comprehensive approach temperature value is greater than the comprehensive approach temperature threshold, then when the predicted comprehensive approach temperature value is 3.5°C, the day is determined to be the latest cleaning time for the refrigeration unit.

[0082] A method for predicting the cleaning time of a refrigeration unit provided by an embodiment of the present disclosure comprises the following steps: obtaining operating data of the refrigeration unit according to a preset refrigeration cycle, wherein the operating data at least includes historical approach temperature data, load rate data and on / off state data; preprocessing the operating data of the refrigeration unit to extract time characteristics of the operating data; dividing the load rate data into preset load rate intervals, and calculating the daily comprehensive approach temperature true value; taking the time characteristics as input and the comprehensive approach temperature true value as output, training an initial comprehensive approach temperature model to obtain a target comprehensive approach temperature prediction model; extracting time characteristics based on the predicted time of the next refrigeration cycle, and inputting the predicted time characteristics into the target comprehensive approach temperature prediction model to obtain a comprehensive approach temperature prediction value; determining the cleaning time of the refrigeration unit based on the comprehensive approach temperature prediction value and the comprehensive approach temperature threshold. Therefore, the disclosed embodiment can predict the dirt cleaning time of the refrigeration unit by combining the comprehensive approach temperature prediction model with the power on / off state data without relying on manual experience, so that the operation and maintenance personnel can clean the dirt generated in the heat exchanger in time, which is conducive to extending the service life of the refrigeration unit, reducing the energy waste of the compressor, saving production costs, and providing normal cooling capacity. At the same time, the disclosed embodiment continuously trains the prediction model based on the operating status of the refrigeration unit, so that the accuracy of the predicted cleaning time is continuously improved, which helps to achieve accurate operation and maintenance of the equipment and improve the production efficiency of the equipment.

[0083] All the above optional technical solutions can be arbitrarily combined to form optional embodiments of the present application, which will not be described one by one here.

[0084] The following are embodiments of the device disclosed herein, which can be used to execute the method embodiments disclosed herein. For details not disclosed in the device embodiments disclosed herein, please refer to the method embodiments disclosed herein.

[0085] Figure 3 is a schematic diagram of a device for predicting the cleaning time of a refrigeration unit provided by an embodiment of the present disclosure, such as Figure 3 As shown, the prediction device for the cleaning time of the refrigeration unit includes:

[0086] The data acquisition module 301 is configured to acquire the operation data of the refrigeration unit according to a preset refrigeration cycle, and the operation data at least includes historical approach temperature data, load rate data and on / off state data;

[0087] The feature extraction module 302 is configured to pre-process the operation data of the refrigeration unit and extract the time features of the operation data;

[0088] The data calculation module 303 is configured to divide the load rate data into preset load rate intervals and calculate the daily comprehensive approach temperature true value;

[0089] The training model module 304 is configured to take the time feature as input and the real value of the comprehensive approaching temperature as output, train the initial comprehensive approaching temperature model, and obtain the target comprehensive approaching temperature prediction model;

[0090] The predicted temperature module 305 is configured to extract the predicted time feature based on the predicted time of the next refrigeration cycle, and input the predicted time feature into the target comprehensive approach temperature prediction model to obtain the comprehensive approach temperature prediction value;

[0091] The time determination module 306 is configured to determine the refrigeration unit cleaning time based on the comprehensive approach temperature prediction value and the comprehensive approach temperature threshold.

[0092] In some embodiments, the data acquisition module 301 is specifically configured to determine the equipment type of the refrigeration unit; if the equipment type is an evaporator, the chilled water outlet temperature and the refrigerant saturated evaporation temperature of the refrigeration unit are obtained; based on the chilled water outlet temperature and the refrigerant saturated evaporation temperature, the historical approach temperature is obtained; if the equipment type is a condenser, the refrigerant saturated condensation temperature and the cooling water outlet temperature of the refrigeration unit are obtained; based on the refrigerant saturated condensation temperature and the cooling water outlet temperature, the historical approach temperature is obtained.

[0093] In some embodiments, the feature extraction module 302 is specifically configured to perform clustering processing on the operation data based on the target time granularity to obtain clustered operation data; perform at least one of effective data screening, elimination of outliers, merging of identical values, and missing value completion processing on the clustered operation data to obtain preprocessed operation data; process the time information of each preprocessed operation data to obtain the time feature of each operation data, and the time feature includes at least year, month, day, and week.

[0094] In some embodiments, the data calculation module 303 is specifically configured to divide the load rate into corresponding load rate intervals according to a preset load rate interval division rule, and calculate the proportion of the data volume in each load rate interval to the total data volume; based on the data volume in each load rate interval and the historical approach temperature data, calculate the average approach temperature in each load rate interval; based on the proportion of the data volume in each load rate interval to the total data volume, the average approach temperature, and the number of intervals, calculate the comprehensive approach temperature to obtain the actual value of the comprehensive approach temperature for each day.

[0095] In some embodiments, the training model module 304 is specifically configured to construct an initial comprehensive approach temperature model based on the time characteristics of the operating data and the comprehensive approach temperature using a regression model; input the time characteristics into the comprehensive approach temperature prediction model to obtain a comprehensive approach temperature prediction value; construct a loss function based on the comprehensive approach temperature prediction value and the comprehensive approach temperature true value; use the loss function to iteratively train the comprehensive approach temperature prediction model until a preset number of rounds is reached or the loss function reaches a preset threshold, thereby obtaining a target comprehensive approach temperature prediction model.

[0096] In some embodiments, the predicted temperature module 305 is specifically configured to set the predicted time for the next refrigeration cycle, the predicted time is no more than 90 days, extract the predicted time information, and obtain the predicted time characteristics; the time characteristics corresponding to the predicted time are input into the target comprehensive approach temperature prediction model to obtain the daily comprehensive approach temperature prediction value within the predicted time.

[0097] In some embodiments, the determination time module 306 is specifically configured to determine in turn whether the comprehensive approach temperature prediction value of each day within the prediction time range is greater than the comprehensive approach temperature threshold; if the comprehensive approach temperature prediction value of a certain day is greater than the comprehensive approach temperature threshold, then the day is determined to be the latest cleaning time of the refrigeration unit.

[0098] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present disclosure.

[0099] Figure 4 Schematic diagram of an electronic device 4 provided in an embodiment of the present disclosure. Figure 4 As shown, the electronic device 4 of this embodiment includes: a processor 401, a memory 402, and a computer program 403 stored in the memory 402 and executable on the processor 401. When the processor 401 executes the computer program 403, the steps in the above-mentioned method embodiments are implemented. Alternatively, when the processor 401 executes the computer program 403, the functions of the modules / units in the above-mentioned device embodiments are implemented.

[0100] The electronic device 4 may be a desktop computer, a notebook, a PDA, a cloud server, or other electronic device. The electronic device 4 may include, but is not limited to, a processor 401 and a memory 402. Those skilled in the art will appreciate that Figure 4 The electronic device 4 is merely an example and does not limit the electronic device 4 , and may include more or less components than those shown in the figure, or different components.

[0101] The processor 401 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0102] The memory 402 may be an internal storage unit of the electronic device 4, for example, a hard disk or memory of the electronic device 4. The memory 402 may also be an external storage device of the electronic device 4, for example, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the electronic device 4. The memory 402 may also include both an internal storage unit and an external storage device of the electronic device 4. The memory 402 is used to store computer programs and other programs and data required by the electronic device.

[0103] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units.

[0104] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present disclosure implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. The computer program may include computer program code, and the computer program code may be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electric carrier signals and telecommunication signals.

[0105] The above embodiments are only used to illustrate the technical solutions of the present disclosure, rather than to limit them. Although the present disclosure has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present disclosure, and should all be included in the protection scope of the present disclosure.

Claims

1. A method for predicting the cleaning time of a refrigeration unit. It is characterized in that include: According to a preset refrigeration cycle, the operation data of the refrigeration unit is obtained, wherein the operation data at least includes historical approach temperature data, load rate data and on / off state data; Preprocessing the operating data of the refrigeration unit to extract the time characteristics of the operating data; Dividing the load rate data into preset load rate intervals, and calculating the daily comprehensive approach temperature true value; Taking the time feature as input and the real value of the comprehensive approaching temperature as output, an initial comprehensive approaching temperature model is trained to obtain a target comprehensive approaching temperature prediction model; Based on the predicted time of the next refrigeration cycle, extract the predicted time feature, and input the predicted time feature into the target comprehensive approach temperature prediction model to obtain the comprehensive approach temperature prediction value; Determining the cleaning time of the refrigeration unit based on the comprehensive approach temperature prediction value and the comprehensive approach temperature threshold; The step of acquiring the historical approach temperature data comprises: Determine the equipment type of the refrigeration unit; If the device type is an evaporator, the chilled water outlet temperature and the refrigerant saturated evaporation temperature of the refrigeration unit are obtained; based on the chilled water outlet temperature and the refrigerant saturated evaporation temperature, the historical approach temperature is obtained; If the device type is a condenser, the refrigerant saturated condensing temperature and the cooling water outlet temperature of the refrigeration unit are obtained; based on the refrigerant saturated condensing temperature and the cooling water outlet temperature, the historical approach temperature is obtained; The calculation of the daily comprehensive approach to the true value of temperature includes: According to a preset load rate interval division rule, the load rate is divided into corresponding load rate intervals, the number of intervals is determined, and the proportion of the data volume in each load rate interval to the total data volume is calculated; Based on the data volume and historical approach temperature data in each load rate interval, the approach temperature average value in each load rate interval is calculated; Based on the number of intervals, the proportion of the data volume in each load rate interval to the total data volume, and the average approach temperature, the comprehensive approach temperature is calculated to obtain the actual value of the daily comprehensive approach temperature.

2. The method according to claim 1, It is characterized in that The preprocessing of the operating data of the refrigeration unit to extract the time characteristics of the operating data includes: Clustering the operation data based on the target time granularity to obtain clustered operation data; Perform at least one of effective data screening, outlier removal, identical value merging, and missing value filling processing on the clustered operation data to obtain preprocessed operation data; The time information of each piece of preprocessed operation data is processed to obtain a time feature of each piece of operation data, wherein the time feature at least includes year, month, day, and week.

3. The method according to claim 1, It is characterized in that The method uses the time feature as input and the real value of the comprehensive approaching temperature as output to train the initial comprehensive approaching temperature model to obtain a target comprehensive approaching temperature prediction model, including: Based on the time characteristics of the operating data and the real value of the comprehensive approach temperature, a regression model is used to construct an initial comprehensive approach temperature prediction model; Inputting the time characteristics into the initial comprehensive approach temperature prediction model to obtain a comprehensive approach temperature prediction value; Constructing a loss function according to the predicted value of the comprehensive approach temperature and the real value of the comprehensive approach temperature; The initial comprehensive approach temperature prediction model is iteratively trained using the loss function until a preset number of rounds is reached or the loss function reaches a preset threshold, thereby obtaining a target comprehensive approach temperature prediction model.

4. The method according to claim 1, It is characterized in that The method of extracting a prediction time feature based on the prediction time of the next refrigeration cycle and inputting the prediction time feature into the target comprehensive approach temperature prediction model to obtain a comprehensive approach temperature prediction value includes: Set a prediction time for the next refrigeration cycle, the prediction time being no more than 90 days, and extract the prediction time feature; The predicted time characteristics are input into the target comprehensive approach temperature prediction model to obtain the daily comprehensive approach temperature prediction value within the prediction time.

5. The method according to claim 1, It is characterized in that The step of determining the refrigeration unit cleaning time based on the comprehensive approach temperature prediction value and the comprehensive approach temperature threshold value includes: Determine in sequence whether the daily comprehensive approach temperature forecast value within the forecast time range is greater than the comprehensive approach temperature threshold; If the comprehensive approach temperature prediction value of a certain day is greater than the comprehensive approach temperature threshold, then this day is determined as the latest cleaning time of the refrigeration unit.

6. A device for predicting the cleaning time of a refrigeration unit, It is characterized in that include: A data acquisition module is configured to acquire operation data of the refrigeration unit according to a preset refrigeration cycle, wherein the operation data at least includes historical approach temperature data, load rate data, and on / off state data; A feature extraction module is configured to pre-process the operation data of the refrigeration unit and extract the time feature of the operation data; A data calculation module is configured to divide the load rate data into preset load rate intervals and calculate the daily comprehensive approach temperature true value; A training model module is configured to take the time feature as input and the real value of the comprehensive approaching temperature as output, train the initial comprehensive approaching temperature model, and obtain a target comprehensive approaching temperature prediction model; A predicted temperature module is configured to extract predicted time features based on the predicted time of the next refrigeration cycle, and input the predicted time features into the target comprehensive approach temperature prediction model to obtain a predicted comprehensive approach temperature value; A time determination module, configured to determine a cleaning time of the refrigeration unit based on the comprehensive approach temperature prediction value and the comprehensive approach temperature threshold; The data acquisition module is specifically configured to determine the device type of the refrigeration unit; if the device type is an evaporator, the chilled water outlet temperature and the refrigerant saturated evaporation temperature of the refrigeration unit are obtained; based on the chilled water outlet temperature and the refrigerant saturated evaporation temperature, the historical approach temperature is obtained; if the device type is a condenser, the refrigerant saturated condensation temperature and the cooling water outlet temperature of the refrigeration unit are obtained; based on the refrigerant saturated condensation temperature and the cooling water outlet temperature, the historical approach temperature is obtained; The data calculation module is specifically configured to divide the load rate into corresponding load rate intervals according to a preset load rate interval division rule, determine the number of intervals, and calculate the proportion of the data volume in each load rate interval to the total data volume; based on the data volume in each load rate interval and the historical approach temperature data, calculate the average approach temperature in each load rate interval; based on the number of intervals, the proportion of the data volume in each load rate interval to the total data volume, and the average approach temperature, calculate the comprehensive approach temperature to obtain the actual value of the comprehensive approach temperature for each day.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, It is characterized in that When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium storing a computer program. It is characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images