WO2023108953A1 - 一种空调机组预测维护方法、装置、设备及介质 - Google Patents

一种空调机组预测维护方法、装置、设备及介质 Download PDF

Info

Publication number
WO2023108953A1
WO2023108953A1 PCT/CN2022/084760 CN2022084760W WO2023108953A1 WO 2023108953 A1 WO2023108953 A1 WO 2023108953A1 CN 2022084760 W CN2022084760 W CN 2022084760W WO 2023108953 A1 WO2023108953 A1 WO 2023108953A1
Authority
WO
WIPO (PCT)
Prior art keywords
performance index
index data
running time
data
time
Prior art date
Application number
PCT/CN2022/084760
Other languages
English (en)
French (fr)
Inventor
陈军
崔景利
廖革文
詹明臻
Original Assignee
维谛技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 维谛技术有限公司 filed Critical 维谛技术有限公司
Publication of WO2023108953A1 publication Critical patent/WO2023108953A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00Administration; Management
    • G06Q10/04Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00Administration; Management
    • G06Q10/06Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063Operations research, analysis or management
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00Administration; Management
    • G06Q10/06Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063Operations research, analysis or management
    • G06Q10/0639Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06393Score-carding, benchmarking or key performance indicator [KPI] analysis
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00Administration; Management
    • G06Q10/20Administration of product repair or maintenance

Definitions

  • the present disclosure relates to the technical field of air conditioning, and in particular to a predictive maintenance method, device, equipment and medium for an air conditioning unit.
  • Air conditioning is a device that "transfers" energy from one side with a lower temperature to the other side with a higher temperature. This process is involuntary. According to the second law of thermodynamics, energy consumption is required to realize this process, and the outdoor unit is responsible for transferring indoor energy and consumed energy to the outside atmosphere. If the heat exchange efficiency of the outdoor unit is reduced, the energy consumption of the whole unit will be greatly increased.
  • the outdoor unit of the air conditioner is exposed to the outdoor environment and is affected by sunlight, rain, dust and other factors. These factors may affect the heat exchange efficiency of the outdoor unit. As for the attenuation of the heat exchanger efficiency of the outdoor unit, what is the specific situation? and other issues have not yet been resolved.
  • the disclosure provides a method, device, equipment and medium for predictive maintenance of an air-conditioning unit to realize intelligent maintenance supported by data analysis.
  • a method for predictive maintenance of an air conditioning unit comprising:
  • Data collection is carried out during the operation of the air conditioning unit, and the collected data includes at least one actual performance index data and actual non-performance index data;
  • the actual non-performance index data is input into the benchmark model to obtain the predicted benchmark performance index data output by the benchmark model, and the benchmark model uses the non-performance index data in the training samples as input to output the
  • the model obtained by training the corresponding benchmark performance index data, the benchmark performance index data is data determined only according to the running time of the air conditioning unit;
  • the training samples are acquired in the following manner:
  • each piece of historical data includes at least one actual performance index data, actual non-performance index data, and predicted benchmark performance index data;
  • Data cleaning is performed on the multiple pieces of historical data according to data rules, and training samples are obtained after cleaning.
  • the data cleaning of the multiple pieces of historical data according to data rules includes at least one of the following steps:
  • the flag bit of the state of the air conditioner unit select the flag bit to be a plurality of historical data of the air conditioner operating state
  • the set time interval select a plurality of pieces of historical data collected corresponding to the time interval
  • the relationship between the pre-fitting performance index data difference and running time includes:
  • each piece of historical data includes at least one actual performance index data, actual non-performance index data, forecast benchmark performance index data, and running time;
  • the collected data also includes the actual running time, after fitting the relationship between the performance index data difference and the running time, it also includes:
  • each piece of real-time data includes the actual performance index data, the predicted benchmark performance index data and the actual running time;
  • the relationship between the fitted performance index data difference and the running time is updated.
  • the performance index data difference includes a high-pressure pressure difference ⁇ HP and a power difference ⁇ P, and according to the pre-fitted relationship between the performance index data difference and the running time, determine the Maintenance programs, including:
  • the final maintenance interval is determined according to the smaller value of the first interval number of days and the second interval number of days.
  • the determining the first number of days between the current time and the next maintenance time according to the first relationship curve between the pre-fitted ⁇ HP and the running time and the set ⁇ HP threshold includes:
  • the first number of days between the current time and the next maintenance time is determined.
  • the determining the second interval days between the current time and the next maintenance time according to the pre-fitted second relationship curve between ⁇ P and running time and the predetermined target maintenance interval includes:
  • the target maintenance interval is the time interval corresponding to the minimum sum of maintenance costs and electric energy consumption costs within a preset period of time, wherein, according to the pre-fitted second relationship curve between ⁇ P and running time, and The third relationship curve between the benchmark P and the running time is used to determine the power consumption cost of the preset period when maintenance is performed according to different time intervals; Electric energy consumption costs and maintenance costs determine said target maintenance intervals;
  • a second interval of days between the current time and the next maintenance time is determined.
  • Electricity consumption costs including:
  • the electricity consumption cost is determined.
  • an air conditioning unit predictive maintenance device comprising:
  • the data acquisition module is used for data acquisition during the operation of the air-conditioning unit, and the collected data includes at least one actual performance index data and actual non-performance index data;
  • the prediction benchmark performance index data acquisition module is used to input the actual non-performance index data into the benchmark model to obtain the forecast benchmark performance index data output by the benchmark model, and the benchmark model uses the non-performance index data in the training samples As an input, a model trained by outputting the corresponding benchmark performance index data in the training sample, the benchmark performance index data is data determined only according to the running time of the air conditioning unit;
  • a maintenance plan determination module configured to determine the difference between the actual performance index data and the corresponding predicted benchmark performance index data, and determine the maintenance plan for the air-conditioning unit according to the relationship between the pre-fitted performance index data difference and the running time .
  • an electronic device including: a processor; a memory for storing executable instructions of the processor; wherein, the processor executes the executable instructions to realize the prediction of the above-mentioned air-conditioning unit The steps of the maintenance method.
  • a computer-readable storage medium on which computer instructions are stored, and when the instructions are executed by a processor, the steps of the above method for predictive maintenance of air-conditioning units are implemented.
  • the base value calculation model was created to eliminate the influence of non-performance indicators, and realized the relationship between the performance index data of key analysis points and the running time, and calculated the actual performance index data and corresponding prediction in real time
  • the difference between the benchmark performance index data and the running time, fitting the relationship between the difference of the performance index data and the running time, and determining the maintenance plan of the air-conditioning unit has changed the current situation that the maintenance of the air-conditioning unit is mainly based on experience, and realized digitization and intelligence. Thereby improving the accuracy of maintenance and optimizing maintenance measures.
  • Fig. 1 is a schematic diagram of an application scenario according to an exemplary embodiment
  • Fig. 2 is a flow chart showing a method for predictive maintenance of an air conditioning unit according to an exemplary embodiment
  • Fig. 3 is a schematic diagram of a module structure of a predictive maintenance device for an air-conditioning unit according to an exemplary embodiment
  • Fig. 4 is a schematic diagram of electronic equipment showing a predictive maintenance method for an air conditioning unit according to an exemplary embodiment
  • Fig. 5 is a schematic diagram of a program product showing a method for predictive maintenance of an air-conditioning unit according to an exemplary embodiment.
  • the outdoor unit of the air conditioner is exposed to the outdoor environment and is affected by sunlight, rain, dust and other factors; these may affect the heat exchange efficiency of the outdoor unit.
  • the attenuation of the heat exchanger efficiency of the outdoor unit how is it? Maintenance, how long is the optimal maintenance interval, no one has been tracking and researching it, and it is a blank in the industry.
  • the present disclosure provides an air conditioning unit predictive maintenance method, device, equipment and media, which realize digitalization and intelligence, thereby improving maintenance accuracy, optimizing maintenance measures, and creating greater economic benefits.
  • FIG. 1 is a schematic diagram of an application scenario of an embodiment of the present disclosure.
  • This disclosure is applied to a plurality of air-conditioning units 10, and the historical data of the air-conditioning units during operation is provided to the database 11 through a wired/wireless communication network, and the central server 12 has the function of calculating benchmark data using a simulation system, which can be input into the database 11 to provide Multiple pieces of historical data of each air-conditioning unit, and obtain the corresponding benchmark performance index data through simulation systems and other methods.
  • the historical data includes non-performance index data, such as outdoor ambient temperature, compressor output, return air temperature and humidity, indoor fan speed and the speed of the outdoor fan.
  • the loss of the air conditioning unit during operation is not only affected by the running time, but also by the output of the air conditioning unit itself, such as outdoor ambient temperature, compressor output, return air temperature and humidity, indoor fan speed, outdoor fan speed, etc.
  • the central server eliminates the influence of factors such as the output of the air-conditioning unit itself, and only considers the loss of the air-conditioning unit under the influence of the running time.
  • the method provided by the present disclosure is applied to the server 13 independent of the central server, and the server 13 is mainly used for determining the maintenance plan by collecting data from the database 11 and the central server 12 .
  • the server 13 obtains the data collected during the operation of the air-conditioning unit provided by the database 11, and the collected data includes at least one actual performance index data and actual non-performance index data; the actual non-performance index data Input the reference model in the central server 12 to obtain the predicted reference performance index data output by the reference model, the reference model uses the non-performance index data in the training samples as input to output the corresponding reference in the training samples
  • the model obtained by training the performance index data, the benchmark performance index data is the data determined only according to the running time of the air-conditioning unit; determine the difference between the actual performance index data and the corresponding predicted benchmark performance index data, and according to the pre-fitting
  • the relationship between the performance index data difference and the running time is used to determine the maintenance plan of the air conditioning unit.
  • a method for predictive maintenance of an air-conditioning unit is provided. Based on the same idea, an apparatus for predictive maintenance of an air-conditioning unit, an electronic device, and a computer readable and writable storage medium are also provided.
  • FIG. 2 A method for predictive maintenance of an air-conditioning unit provided by the present disclosure will be described below through specific embodiments, as shown in FIG. 2 , including:
  • Step 201 collecting data during the operation of the air-conditioning unit, and the collected data includes at least one actual performance index data and actual non-performance index data;
  • data collection is carried out according to the time point, and one piece of data is collected at each time point, and multiple pieces of data collected can be stored in a table according to the time point in rows, and each piece of data contains the required actual non-performance Index data and at least one actual performance index data, wherein the non-performance index data includes outdoor ambient temperature, compressor output, return air temperature and humidity, indoor fan speed and outdoor fan speed, etc.
  • Performance index data include HP (High Pressure, high pressure), P (Power, power), low pressure, exhaust temperature and suction temperature, etc.
  • Step 202 Input the actual non-performance index data into the benchmark model to obtain the predicted benchmark performance index data output by the benchmark model.
  • the benchmark model uses the non-performance index data in the training samples as input to output the The model obtained by training the corresponding benchmark performance index data in the training sample, and the benchmark performance index data is data determined only according to the running time of the air conditioning unit;
  • the above benchmark performance index data is in an ideal environment, and the performance index data in the running state of the air conditioning unit is only affected by the running time. But in real life, the running state of the air conditioning unit is not only affected by the running time, but also by other factors, such as outdoor ambient temperature, compressor output, return air temperature and humidity, indoor fan speed, outdoor fan speed, etc. Input the actual non-performance index data at a certain moment into the benchmark model, and the benchmark model eliminates the influence of these non-performance index data on the air-conditioning unit, so as to determine the benchmark performance index data at this moment when the air-conditioning unit is only affected by the running time in its operating state .
  • the training samples are obtained by obtaining multiple pieces of historical data of the air-conditioning unit within a preset period of time from the central server, and performing data cleaning on these historical data according to data rules, wherein each piece of historical data includes at least one performance index data, non-performance indicator data, and benchmark performance indicator data.
  • the benchmark performance index data is to input the actual non-performance index data into the benchmark model of the central server to obtain the corresponding predicted benchmark performance index data.
  • the central server builds a benchmark model by using a simulation system or model selection software based on the historical data of the air-conditioning unit within a preset period of time, for example, multiple pieces of data of the air-conditioning unit collected in the last month.
  • the above-mentioned data cleaning of the multiple pieces of historical data according to data rules may include at least one of the following steps:
  • the selected flag bits are multiple pieces of historical data of the air conditioner operating status.
  • the end state of the unit operation the selection flag bit is a plurality of historical data of the air conditioner operation state;
  • the set time interval select a plurality of pieces of historical data collected corresponding to the time interval, for example, considering that the outdoor environment temperature will not change too quickly, you can select a plurality of pieces of historical data collected at a corresponding interval of 1 hour;
  • multiple pieces of historical data with adjacent outdoor ambient temperature differences greater than the set outdoor ambient temperature difference are selected, for example, multiple pieces of historical data with adjacent outdoor ambient temperature differences greater than 1 degree Celsius are selected.
  • Step 203 Determine the difference between the actual performance index data and the corresponding predicted benchmark performance index data, and determine the maintenance plan for the air conditioning unit according to the relationship between the pre-fitted performance index data difference and running time.
  • the number of days between the current time and the next maintenance time can be determined according to the relationship between the pre-fitted performance index data difference and the running time.
  • This disclosure changes the current situation of relying on experience to realize the maintenance of air-conditioning units. It mainly uses the revised theoretical model as a benchmark to make real-time comparisons with the operating status of online operating units in the market, and continuously compares the operating status of online units with the benchmark. Then use the AI (Artificial Intelligence, artificial intelligence) machine learning method to predict the operating status of the air conditioning unit in the future, and give a reasonable maintenance time according to the set reliability or economic conditions, and realize the intelligent maintenance supported by data analysis. Optimize maintenance, thereby improving maintenance accuracy and optimizing maintenance measures.
  • AI Artificial Intelligence, artificial intelligence
  • At least one actual performance index data and actual non-performance index data of the air-conditioning unit may be collected by means of local collection and storage or cloud collection and storage.
  • the collected actual non-performance index data may be outdoor ambient temperature, compressor output, return air temperature and humidity, indoor fan speed and outdoor fan speed, and the actual performance index data may be HP and P.
  • the actual non-performance index data is input into the benchmark model, and the predicted benchmark performance index data output by the benchmark model is obtained.
  • each piece of historical data includes outdoor ambient temperature, compressor output, return air temperature and humidity, indoor fan speed, outdoor fan speed, and baseline HP.
  • Data cleaning is performed on these data to obtain training samples, and based on the training samples, a benchmark model that can be used to calculate continuous data is created through a machine learning method.
  • each piece of historical data includes outdoor ambient temperature, compressor output, return air temperature and humidity, indoor fan speed, outdoor fan speed, and benchmark P
  • Data cleaning is performed on these data to obtain training samples, and based on the training samples, a benchmark model that can be used to calculate continuous data is created through a machine learning method.
  • the aforementioned benchmark model is a model trained by using the non-performance index data in the training sample as input to output the corresponding benchmark performance index data in the training sample, and the benchmark performance index data is determined only according to the operating hours of the air-conditioning unit The data.
  • the purpose of creating a benchmark model is that the operating state of the air conditioning unit is not only affected by the operating time, but also the output of the unit itself, such as outdoor ambient temperature, compressor output, return air temperature and humidity, and indoor and outdoor fan speed.
  • create a benchmark model The main purpose is to disassociate the output from the unit itself, and only analyze the relationship with the running time.
  • the training samples can be obtained in the following ways:
  • each piece of historical data includes at least one actual performance index data, actual non-performance index data, and predicted benchmark performance index data;
  • the efficiency will be low.
  • the multiple pieces of historical data are cleaned according to the data rules, and the training samples are obtained after cleaning.
  • the above-mentioned data cleaning of the multiple pieces of historical data according to data rules includes at least one of the following steps:
  • the selected flags are a plurality of historical data of the operating state of the air conditioner.
  • the flags of the state of the air conditioner include the non-operating state of the air conditioner and the open state of the air conditioner. and air conditioning unit operation end status;
  • select a plurality of pieces of historical data collected corresponding to the time interval for example, select a plurality of pieces of historical data collected at a corresponding interval of 1 hour;
  • multiple pieces of historical data with adjacent outdoor ambient temperature differences greater than the set outdoor ambient temperature difference are selected, for example, multiple pieces of historical data with adjacent outdoor ambient temperature differences greater than 1 degree Celsius are selected.
  • the relationship between the performance index data difference and the running time can be pre-fitted by the following methods:
  • each piece of historical data includes at least one actual performance index data, actual non-performance index data, forecast benchmark performance index data, and running time;
  • the relationship between the performance index data difference and the running time can be fitted.
  • Machine learning and other methods can be used to fit the relationship curve between the performance index data difference and the running time, for example Use XGBoost (eXtreme Gradient Boosting, extreme gradient boosting) algorithm to establish a model of performance index data difference and running time.
  • XGBoost eXtreme Gradient Boosting, extreme gradient boosting
  • the difference between the actual HP and the collected corresponding predicted benchmark HP can be used to obtain the high pressure difference ( ⁇ HP); according to the ⁇ HP and running time corresponding to each piece of historical data, the fitting ⁇ HP and running time relationship.
  • the difference between the actual P and the collected prediction reference P can be obtained to obtain the power difference ( ⁇ P); according to the ⁇ P and running time corresponding to each piece of historical data, the fitting ⁇ P and the running time time relationship.
  • the specific steps include:
  • each piece of real-time data includes the actual performance index data, the predicted benchmark performance index data and the actual running time;
  • the relationship between the fitted performance index data difference and the running time is updated.
  • the maintenance plan of the air conditioning unit can be determined by the following methods:
  • the final maintenance interval is determined according to the smaller value of the first interval number of days and the second interval number of days.
  • the first interval number of days or the second interval number of days can be directly used as the final maintenance interval; or when the first interval number of days is greater than the second interval number of days, from the perspective of reliability, the first interval number of days can be used as the final maintenance interval Maintenance interval, or from an economic point of view, the second interval of days as the final maintenance interval; also when the first interval of days is less than the second interval of days, from the perspective of reliability, the first interval of days as the final maintenance interval maintenance interval.
  • the high HP will trigger the automatic protection of the air conditioning unit, that is, the air conditioning unit will automatically shut down. If it is triggered multiple times within a period of time, the air conditioning unit will be locked and cannot operate normally. Therefore, according to the pre-fitting relationship between ⁇ HP and running time, determine the first interval between the current time and the next maintenance time, so as to realize the function of early warning.
  • the specific steps are as follows:
  • the collected current non-performance index data is brought into the reference model to obtain the corresponding current prediction reference HP, and the difference between the current HP and the current prediction reference HP is the current ⁇ HP.
  • the first number of days between the current time and the next maintenance time is determined.
  • the first running time corresponding to the current ⁇ HP is 3 days
  • the second running time corresponding to the ⁇ HP threshold determined according to the first relationship curve is 7 days
  • the first interval between the current time and the next maintenance time is 4 days .
  • the target running time corresponding to the current ⁇ P in the second relationship curve and the target maintenance interval that is, subtracting the target maintenance interval from the target running time corresponding to the current ⁇ P in the second relationship curve to determine the current time distance
  • the second interval of days between the next maintenance time
  • the collected current non-performance index data is brought into the reference model to obtain the corresponding current prediction reference P, then the difference between the current P and the current prediction reference P is the current ⁇ P.
  • the above target maintenance intervals can be determined according to the following methods:
  • the time interval corresponding to the minimum sum of the maintenance fee and the electric energy consumption fee within the preset period is determined as the target maintenance interval.
  • the electric energy consumption fee for the preset time period can be determined by the following method:
  • the maintenance interval is 2 days
  • the maintenance is performed on the 2nd and 4th day out of 4 days. Since the running time is reset to 0 after each maintenance, the first maintenance interval, that is, the 1st to 4th day
  • the actual P for day 2 is the same as the actual P for the second maintenance interval, day 3 to 4.
  • the corresponding prediction benchmark P By inputting multiple pieces of actual non-performance index data collected during the actual operation of the air-conditioning unit into the benchmark model, the corresponding prediction benchmark P can be obtained. According to the obtained predicted benchmark P and its corresponding running time, a third relationship curve between the benchmark P and running time can be determined.
  • the above-mentioned preset period includes a period of time in the future, it is necessary to determine the reference P within the preset period according to the third relationship curve between the reference P and the running time, and according to the second relationship between the pre-fitted ⁇ P and the running time curve to determine the ⁇ P within the preset time period.
  • the reference P and ⁇ P within the preset period are added to obtain the actual P corresponding to the same running time.
  • the electricity consumption cost is determined.
  • an embodiment of the present disclosure also provides a device for predictive maintenance of an air-conditioning unit. Since the device is the device in the method in the embodiment of the present disclosure, and the problem-solving principle of the device is similar to the method, the For the implementation of the device, reference may be made to the implementation of the method, and repeated descriptions will not be repeated.
  • the above-mentioned device includes the following modules:
  • the data collection module 301 is used to collect data during the operation of the air conditioning unit, and the collected data includes at least one actual performance index data and actual non-performance index data;
  • the predicted benchmark performance index data obtaining module 302 is used to input the actual non-performance index data into the benchmark model to obtain the predicted benchmark performance index data output by the benchmark model, and the benchmark model uses the non-performance index data in the training samples
  • the data is used as an input to output the model obtained by training the corresponding benchmark performance index data in the training sample, and the benchmark performance index data is data determined only according to the running time of the air-conditioning unit;
  • a maintenance plan determination module 303 configured to determine the difference between the actual performance index data and the corresponding predicted benchmark performance index data, and determine the maintenance of the air conditioning unit according to the relationship between the pre-fitted performance index data difference and the running time plan.
  • the prediction benchmark performance index data acquisition module acquires the training samples in the following manner:
  • each piece of historical data includes at least one actual performance index data, actual non-performance index data, and predicted benchmark performance index data;
  • Data cleaning is performed on the multiple pieces of historical data according to data rules, and training samples are obtained after cleaning.
  • the prediction benchmark performance index data acquisition module is used to perform data cleaning on the multiple pieces of historical data according to data rules, including at least one of the following steps:
  • the flag bit of the state of the air conditioner unit select the flag bit to be a plurality of historical data of the air conditioner operating state
  • the set time interval select a plurality of pieces of historical data collected corresponding to the time interval
  • the maintenance scheme determination module is used to pre-fit the relationship between the performance index data difference and the running time, including:
  • each piece of historical data includes at least one actual performance index data, actual non-performance index data, forecast benchmark performance index data, and running time;
  • the collected data also includes actual running time
  • the device also includes:
  • a real-time data acquisition module configured to acquire multiple pieces of real-time data, each piece of real-time data including said actual performance index data, predicted benchmark performance index data, and actual running time;
  • a difference determination module configured to determine the difference between the actual performance index data and the corresponding predicted benchmark performance index data
  • the relationship update module is used to update the relationship between the fitted performance index data difference and the running time according to the difference corresponding to each piece of real-time data and the actual running time.
  • the performance index data difference includes ⁇ HP and ⁇ P
  • the maintenance plan determination module is used to determine the air conditioning unit's Maintenance programs, including:
  • the final maintenance interval is determined according to the smaller value of the first interval number of days and the second interval number of days.
  • the maintenance plan determination module is configured to determine the first time interval between the current time and the next maintenance time according to the first relationship curve between the pre-fitted ⁇ HP and the running time and the set ⁇ HP threshold. Number of days between, including:
  • the first number of days between the current time and the next maintenance time is determined.
  • the maintenance scheme determination module is configured to determine the distance between the current time and the next maintenance time according to the second relationship curve between the pre-fitted ⁇ P and the running time and the predetermined target maintenance interval.
  • the number of days between two days including:
  • the target maintenance interval is the time interval corresponding to the minimum sum of maintenance costs and electric energy consumption costs within a preset period of time, wherein, according to the pre-fitted second relationship curve between ⁇ P and running time, and The third relationship curve between the benchmark P and the running time is used to determine the power consumption cost of the preset period when maintenance is performed according to different time intervals; Electric energy consumption costs and maintenance costs determine said target maintenance intervals;
  • a second interval of days between the current time and the next maintenance time is determined.
  • the maintenance plan determination module is configured to determine the maintenance plan at different time intervals according to the second relationship curve between the pre-fitted ⁇ P and the running time, and the third relationship curve between the reference P and the running time.
  • Electricity consumption charges for preset periods when maintenance is performed including:
  • the electricity consumption cost is determined.
  • the embodiments of the present disclosure also provide an electronic device for predictive maintenance of air-conditioning units, because the electronic device is the electronic device in the method in the embodiment of the present disclosure, and the principle of solving problems of the electronic device is the same as The method is similar, so the implementation of the electronic device can refer to the implementation of the method, and the repetition will not be repeated.
  • FIG. 4 An electronic device 40 according to this embodiment of the present disclosure is described below with reference to FIG. 4 .
  • the electronic device 40 shown in FIG. 4 is only an example, and should not limit the functions and scope of use of the embodiments of the present disclosure.
  • the electronic device 40 may be in the form of a general-purpose computing device, for example, it may be a terminal device.
  • Components of the electronic device 40 may include, but are not limited to: at least one processor 41 , at least one memory 42 storing processor-executable instructions, and a bus 43 connecting different system components (including the memory 42 and the processor 41 ).
  • the processor implements the following steps by running the executable instructions:
  • Data collection is carried out during the operation of the air conditioning unit, and the collected data includes at least one actual performance index data and actual non-performance index data;
  • the actual non-performance index data is input into the benchmark model to obtain the predicted benchmark performance index data output by the benchmark model, and the benchmark model uses the non-performance index data in the training samples as input to output the
  • the model obtained by training the corresponding benchmark performance index data, the benchmark performance index data is data determined only according to the running time of the air conditioning unit;
  • the processor also implements the method for predictive maintenance of the air-conditioning unit in Embodiment 1 by running the executable instructions, and repeated descriptions will not be repeated here.
  • Memory 42 may include readable media in the form of volatile memory, such as random access memory (RAM) 421 and/or cache memory 422 , and may further include read only memory (ROM) 423 .
  • RAM random access memory
  • ROM read only memory
  • Memory 42 may also include programs/utilities 425 having a set (at least one) of program modules 424 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, which Each or some combination of the examples may include the implementation of a network environment.
  • programs/utilities 425 having a set (at least one) of program modules 424 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, which Each or some combination of the examples may include the implementation of a network environment.
  • the electronic device 40 can also communicate with one or more external devices 44 (such as keyboards, pointing devices, etc.), and can also communicate with one or more devices that enable the user to interact with the electronic device 40, and/or communicate with the electronic device 40. Any device (eg, router, modem, etc.) capable of communicating with one or more other computing devices communicates. Such communication may occur through input/output (I/O) interface 45 .
  • the electronic device 40 can also communicate with one or more networks (such as a local area network (LAN), a wide area network (WAN) and/or a public network such as the Internet) through the network adapter 46 . As shown, network adapter 46 communicates with other modules of electronic device 40 via bus 43 . It should be appreciated that although not shown, other hardware and/or software modules may be used in conjunction with electronic device 40, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives And data backup storage system, etc.
  • various aspects of the present disclosure can also be implemented in the form of a program product, which includes program code.
  • the program code is used to make the terminal device execute the above-mentioned
  • the terminal device can be used to collect data during the operation of the air conditioning unit, and the collected data Including at least one actual performance index data and actual non-performance index data; the actual non-performance index data is input into the benchmark model, and the predicted benchmark performance index data output by the benchmark model is obtained, and the benchmark model uses the training sample
  • the non-performance index data is used as input to output the model obtained by training the corresponding benchmark performance index data in the training sample, and the benchmark performance index data is data determined only according to the running time of the air conditioning unit; determine the actual performance index The difference between the data and the corresponding predicted benchmark performance index data, and according to the relationship between the pre-fitted performance index data
  • a program product may take the form of any combination of one or more readable media.
  • the readable medium may be a readable signal medium or a readable storage medium.
  • a readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any combination thereof. More specific examples (non-exhaustive list) of readable storage media include: electrical connection with one or more conductors, portable disk, hard disk, random access memory (RAM), read only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.
  • a program product 50 for predictive maintenance of air-conditioning units according to an embodiment of the present disclosure is described, which can adopt a portable compact disk read-only memory (CD-ROM) and include program codes, and can be installed on a terminal device , such as running on a personal computer.
  • a program product of the present disclosure is not limited thereto.
  • a readable storage medium may be any tangible medium containing or storing a program, and the program may be used by or in combination with an instruction execution system, apparatus or device.
  • a readable signal medium may include a data signal carrying readable program code in baseband or as part of a carrier wave. Such propagated data signals may take many forms, including - but not limited to - electromagnetic signals, optical signals, or any suitable combination of the foregoing.
  • a readable signal medium may also be any readable medium other than a readable storage medium that can transmit, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device.
  • Program code embodied on a readable medium may be transmitted using any appropriate medium, including - but not limited to - wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
  • Program code for performing the operations of the present disclosure may be written in any combination of one or more programming languages, including object-oriented programming languages—such as Java, C++, etc., as well as conventional procedural programming Language - such as "C" or similar programming language.
  • the program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device, or entirely on the remote computing device or server to execute.
  • the remote computing device may be connected to the user computing device through any kind of network, including a local area network (LAN) or a wide area network (WAN), or, alternatively, may be connected to an external computing device (e.g., using an Internet service Provider via Internet connection).
  • LAN local area network
  • WAN wide area network
  • an external computing device e.g., using an Internet service Provider via Internet connection.
  • the embodiments of the present disclosure may be provided as methods, systems, or computer program products. Accordingly, the present disclosure can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present disclosure may take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, optical storage, etc.) having computer-usable program code embodied therein.
  • These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, such that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, the instructions The device realizes the function specified in one or more procedures of the flowchart and/or one or more blocks of the block diagram.

Landscapes

  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Engineering & Computer Science (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Physics & Mathematics (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Development Economics (AREA)
  • Game Theory and Decision Science (AREA)
  • Educational Administration (AREA)
  • Air Conditioning Control Device (AREA)

Abstract

一种空调机组预测维护方法、装置、设备及介质,涉及空调技术领域。该方法包括:在空调机组运行过程中进行数据采集,所采集的数据包括至少一个实际性能指标数据及实际非性能指标数据(201);将实际非性能指标数据输入到基准模型,得到基准模型输出的预测基准性能指标数据,所述基准模型为利用训练样本中的非性能指标数据作为输入,以输出所述训练样本中对应的基准性能指标数据而训练得到的模型,所述基准性能指标数据为仅根据空调机组的运行时长确定的数据(202);确定实际性能指标数据与对应预测基准性能指标数据的差值,并根据预先拟合的性能指标数据差值与运行时间的关系,确定所述空调机组的维护方案(203)。该方法可以进行维护时间预测,提高维护的准确度。

Description

一种空调机组预测维护方法、装置、设备及介质
本申请要求于2021年12月16日提交中国专利局、申请号为202111545585.1、发明名称为“一种空调机组预测维护方法、装置、设备及介质”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
技术领域
本公开涉及空调技术领域,尤其涉及一种空调机组预测维护方法、装置、设备及介质。
背景技术
空调是实现把能量从温度较低的一侧“搬运”到温度较高的另一侧的设备,这个过程是非自发的。根据热力学第二定律,实现这个过程需要消耗能量,室外机负责把室内能量和消耗的能量传递到外界大气中。若室外机的换热效率降低,将大大增加整机的能耗。
空调外机暴露在室外环境中,受阳光、雨水、灰尘以及其他因素的影响,这些因素都有可能会影响室外机的换热效率,至于室外机换热器效率的衰减情况具体是怎么样的等问题,目前尚未解决。
因此,为了解决上述问题,需要提出一种预测室外机换热器效率的方法。
发明内容
本公开提供一种空调机组预测维护方法、装置、设备及介质,实现以数据分析为支持的智能化维护。
根据本公开实施例的第一方面,提供一种空调机组预测维护方法,该方法包括:
在空调机组运行过程中进行数据采集,所采集的数据包括至少一个实际性能指标数据及实际非性能指标数据;
将所述实际非性能指标数据输入到基准模型,得到所述基准模型输出的预测基准性能指标数据,所述基准模型为利用训练样本中的非性能指标数据作为输入,以输出所述训练样本中对应的基准性能指标数据而训练得到的模型,所述基准性能指标数据为仅根据空调机组的运行时长确定的数据;
确定所述实际性能指标数据与对应预测基准性能指标数据的差值,并根据预先拟合的性能指标数据差值与运行时间的关系,确定所述空调机组的维护方案。
在一种可能的实现方式中,通过如下方式获取所述训练样本:
从中心服务器获取预设时间段内所述空调机组的多条历史数据,每条历史数据包括至少一个实际性能指标数据、实际非性能指标数据及预测基准性能指标数据;
对所述多条历史数据按照数据规则进行数据清洗,清洗后得到训练样本。
在一种可能的实现方式中,所述对所述多条历史数据按照数据规则进行数据清洗,包括如下至少一个步骤:
根据空调机组状态的标志位,选择标志位是空调运行状态的多条历史数据;
根据设定的时间间隔,选取对应所述时间间隔采集的多条历史数据;
根据设定的室外环境温度差值,选取相邻室外环境温差大于所述设定的室外环境温度差值的多条历史数据。
在一种可能的实现方式中,所述预先拟合的性能指标数据差值与运行时间的关系,包括:
从中心服务器获取预设时间段内所述空调机组的多条历史数据,每条历史数据包括至少一个实际性能指标数据、实际非性能指标数据、预测基准性能指标数据及运行时间;
确定所述实际性能指标数据与对应预测基准性能指标数据的差值;
根据各条历史数据对应的差值及运行时间,拟合性能指标数据差值与运行时间的关系。
在一种可能的实现方式中,所采集的数据还包括实际运行时间,拟合性能 指标数据差值与运行时间的关系之后,还包括:
获取多条实时数据,每条实时数据包括所述实际性能指标数据、预测基准性能指标数据及实际运行时间;
确定所述实际性能指标数据与对应的预测基准性能指标数据的差值;
根据各条实时数据对应的差值及实际运行时间,更新拟合的性能指标数据差值与运行时间的关系。
在一种可能的实现方式中,所述性能指标数据差值包括高压压力差值ΔHP和功率差值ΔP,根据预先拟合的性能指标数据差值与运行时间的关系,确定所述空调机组的维护方案,包括:
根据预先拟合的ΔHP与运行时间的第一关系曲线和设定的ΔHP阈值,确定当前时间距离下次维护时间的第一间隔天数;
根据预先拟合的ΔP与运行时间的第二关系曲线和预先确定的目标维护间隔,确定当前时间距离下次维护时间的第二间隔天数;
根据第一间隔天数和第二间隔天数的较小值确定最终的维护间隔。
在一种可能的实现方式中,所述根据预先拟合的ΔHP与运行时间的第一关系曲线和设定的ΔHP阈值,确定当前时间距离下次维护时间的第一间隔天数,包括:
根据预先拟合的ΔHP与运行时间的第一关系曲线,确定当前ΔHP对应的第一运行时间;
根据所述第一关系曲线确定ΔHP阈值对应的第二运行时间;
根据所述第一运行时间和第二运行时间的差值,确定当前时间距离下次维护时间的第一间隔天数。
在一种可能的实现方式中,所述根据预先拟合的ΔP与运行时间的第二关系曲线和预先确定的目标维护间隔,确定当前时间距离下次维护时间的第二间隔天数,包括:
获取预先确定的目标维护间隔,所述目标维护间隔为预设时段内维护费用与电能消耗费用总和最小对应的时间间隔,其中,预先根据预先拟合的ΔP与运 行时间的第二关系曲线,及基准P与运行时间的第三关系曲线,确定按照不同时间间隔进行维护时预设时段的电能消耗费用;确定按照不同时间间隔维护时,在预设时段的维护费用,根据所述预设时段的电能消耗费用和维护费用确定所述目标维护间隔;
根据当前ΔP在所述第二关系曲线对应的目标运行时间和所述目标维护间隔,确定当前时间距离下次维护时间的第二间隔天数。
在一种可能的实现方式中,所述根据预先拟合的ΔP与运行时间的第二关系曲线,及基准P与运行时间的第三关系曲线,确定按照不同时间间隔进行维护时预设时段的电能消耗费用,包括:
在预设时段内,假设每次维护后运行时长重置为零,确定不同维护间隔对应的各运行时长;
根据预先拟合的ΔP与运行时间的第二关系曲线,及基准P与运行时间的第三关系曲线,确定同一运行时长对应的实际P;
根据各运行时长和对应的实际P及电价,确定电能消耗费用。
根据本公开实施例的第二方面,提供一种空调机组预测维护装置,该装置包括:
数据采集模块,用于在空调机组运行过程中进行数据采集,所采集的数据包括至少一个实际性能指标数据及实际非性能指标数据;
预测基准性能指标数据获得模块,用于将所述实际非性能指标数据输入到基准模型,得到所述基准模型输出的预测基准性能指标数据,所述基准模型为利用训练样本中的非性能指标数据作为输入,以输出所述训练样本中对应的基准性能指标数据而训练得到的模型,所述基准性能指标数据为仅根据空调机组的运行时长确定的数据;
维护方案确定模块,用于确定所述实际性能指标数据与对应预测基准性能指标数据的差值,并根据预先拟合的性能指标数据差值与运行时间的关系,确定所述空调机组的维护方案。
根据本公开实施例的第三方面,提供一种电子设备包括:处理器;用于存 储处理器可执行指令的存储器;其中,所述处理器通过运行所述可执行指令以实现上述空调机组预测维护方法的步骤。
根据本公开实施例的第四方面,提供一种计算机可读存储介质,其上存储有计算机指令,该指令被处理器执行时实现上述空调机组预测维护方法的步骤。
另外,第二方面至第四方面中任一种实现方式所带来的技术效果可参见第一方面中不同实现方式所带来的技术效果,此处不再赘述。
本公开的实施例提供的技术方案至少带来以下有益效果:
通过对采集的空调机组运行数据的统计分析,创建的基准值计算模型,排除非性能指标的影响,实现了关键分析点性能指标数据与运行时间的关系,通过实时计算实际性能指标数据与对应预测基准性能指标数据的差值和运行时间,拟合性能指标数据差值与运行时间的关系,确定空调机组的维护方案,改变了目前主要以经验实现空调机组维护的现状,实现数字化和智能化,从而提高维护的准确度,优化维护的措施。
附图说明
为了更清楚地说明本公开实施例中的技术方案,下面将对实施例描述中所需要使用的附图作简要介绍,显而易见地,下面描述中的附图仅仅是本公开的一些实施例,对于本领域的普通技术人员来讲,在不付出创造性劳动性的前提下,还可以根据这些附图获得其他的附图。
图1是根据一示例性实施例示出的应用场景示意图;
图2是根据一示例性实施例示出的一种空调机组预测维护方法的流程图;
图3是根据一示例性实施例示出的一种空调机组预测维护装置的模块结构示意图;
图4是根据一示例性实施例示出的一种空调机组预测维护方法的电子设备示意图;
图5是根据一示例性实施例示出的一种空调机组预测维护方法的程序产品示意图。
具体实施方式
为了使本公开的目的、技术方案和优点更加清楚,下面将结合附图对本公开作进一步地详细描述,显然,所描述的实施例仅仅是本公开一部分实施例,而不是全部的实施例。基于本公开中的实施例,本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其它实施例,都属于本公开保护的范围。
1、本公开实施例中术语“和/或”,描述关联对象的关联关系,表示可以存在三种关系,例如,A和/或B,可以表示:单独存在A,同时存在A和B,单独存在B这三种情况。字符“/”一般表示前后关联对象是一种“或”的关系。
2、本公开的说明书和权利要求书及上述附图中的术语“第一”、“第二”等是用于区别类似的对象,而不必用于描述特定的顺序或先后次序。应该理解这样使用的数据在适当情况下可以互换,以便这里描述的本公开的实施例能够以除了在这里图示或描述的那些以外的顺序实施。
本公开实施例描述的应用场景是为了更加清楚的说明本公开实施例的技术方案,并不构成对于本公开实施例提供的技术方案的限定,本领域普通技术人员可知,随着新应用场景的出现,本公开实施例提供的技术方案对于类似的技术问题,同样适用。其中,在本公开的描述中,除非另有说明,“多个”的含义是两个或两个以上。
空调外机暴露在室外环境中,受阳光、雨水、灰尘以及其他因素的影响;这些都有可能会影响室外机的换热效率,至于室外机换热器效率的衰减情况具体是怎么样,如何维护,最佳维护间隔是多久,目前没有人持续跟踪研究,在业界属于空白。
为了解决上述问题,本公开提供了一种空调机组预测维护方法、装置、设备及介质,实现数字化,智能化,从而提高维护的准确度,优化维护的措施,创造更大的经济效益。
首先参考图1,其为本公开实施例的应用场景示意图。本公开应用于多个 空调机组10,通过有线/无线通信网络,将空调机组在运行过程中的历史数据提供给数据库11,中心服务器12具有利用仿真系统计算基准数据的功能,可以输入数据库11提供的每个空调机组的多条历史数据,通过仿真系统等方法获得对应的基准性能指标数据,所述历史数据包括非性能指标数据,例如室外环境温度、压缩机输出、回风温湿度、室内风机转速和室外风机转速。空调机组在运行过程中的损耗除了受运行时间的影响,还会受空调机组本身输出等因素影响,例如室外环境温度、压缩机输出、回风温湿度、室内风机转速、室外风机转速等。中心服务器剔除了空调机组本身输出等因素的影响,仅考虑空调机组在运行时间的影响下的损耗。
在此基础上,本公开提供的方法应用于独立于中心服务器的服务器13,该服务器13主要用于通过从数据库11和中心服务器12采集数据进行维护方案的确定。本公开实施例中,服务器13、获得数据库11提供的在空调机组运行过程中采集的数据,所采集的数据包括至少一个实际性能指标数据及实际非性能指标数据;将所述实际非性能指标数据输入到中心服务器12中的基准模型,得到所述基准模型输出的预测基准性能指标数据,所述基准模型为利用训练样本中的非性能指标数据作为输入,以输出所述训练样本中对应的基准性能指标数据而训练得到的模型,所述基准性能指标数据为仅根据空调机组的运行时长确定的数据;确定所述实际性能指标数据与对应预测基准性能指标数据的差值,并根据预先拟合的性能指标数据差值与运行时间的关系,确定所述空调机组的 维护方案。
本公开实施例中,提供了一种空调机组预测维护方法,基于同一构思,还提供了一种空调机组预测维护装置、一种电子设备以及一种计算机可读写存储介质。
实施例1
下面通过具体的实施例对本公开提供的一种空调机组预测维护方法进行说明,如图2所示,包括:
步骤201,在空调机组运行过程中进行数据采集,所采集的数据包括至少一个实际性能指标数据及实际非性能指标数据;
在空调机组运行过程中,按照时间点进行数据采集,每个时间点采集一条数据,可以将采集的多条数据根据时间点按行排列存放到表格中,每条数据包含所需的实际非性能指标数据和至少一个实际性能指标数据,其中非性能指标数据包括室外环境温度、压缩机输出、回风温湿度、室内风机转速和室外风机转速等。性能指标数据包括HP(High Pressure,高压压力)、P(Power,功率)、低压压力、排气温度和吸气温度等。
步骤202,将所述实际非性能指标数据输入到基准模型,得到所述基准模型输出的预测基准性能指标数据,所述基准模型为利用训练样本中的非性能指标数据作为输入,以输出所述训练样本中对应的基准性能指标数据而训练得到的模型,所述基准性能指标数据为仅根据空调机组的运行时长确定的数据;
上述基准性能指标数据是在一种理想环境下,空调机组运行状态中的性能指标数据仅受运行时间的影响。但在实际生活中,空调机组的运行状态不仅受运行时间的影响,还会受其他因素的影响,例如室外环境温度、压缩机输出、回风温湿度、室内风机转速、室外风机转速等。将某一时刻的实际非性能指标数据输入到基准模型中,基准模型剔除这些非性能指标数据对空调机组的影响,从而确定空调机组在运行状态只受运行时间影响下此时刻的基准性能指标数 据。
所述训练样本是通过从中心服务器获取预设时间段内所述空调机组的多条历史数据,并且对这些历史数据按照数据规则进行数据清洗获得的,其中,每条历史数据包括至少一个性能指标数据、非性能指标数据及基准性能指标数据。
所述基准性能指标数据是将实际非性能指标数据输入到中心服务器的基准模型中,得到对应的预测基准性能指标数据。中心服务器根据预设时间段内所述空调机组的历史数据,例如最近1个月内所采集的空调机组的多条数据,利用仿真系统或选型软件,构建基准模型。上述对所述多条历史数据按照数据规则进行数据清洗,可以包括如下至少一个步骤:
根据空调机组状态的标志位,选择标志位是空调运行状态的多条历史数据,例如,在excel文本中有相关的标志位,所述标志位包括空调机组未运行状态、空调机组开启状态和空调机组运行结束状态,选择标志位是空调运行状态的多条历史数据;
根据设定的时间间隔,选取对应所述时间间隔采集的多条历史数据,例如,考虑室外环境温度变化不会太快,可以选取对应间隔1小时采集的多条历史数据;
根据设定的室外环境温度差值,选取相邻室外环境温差大于所述设定的室外环境温度差值的多条历史数据,例如,选取相邻室外环境温差大于1摄氏度的多条历史数据。
步骤203,确定所述实际性能指标数据与对应预测基准性能指标数据的差值,并根据预先拟合的性能指标数据差值与运行时间的关系,确定所述空调机组的维护方案。
可以根据预先拟合的性能指标数据差值与运行时间的关系,确定当前时间距离下次维护时间的隔天数。
本公开改变了目前依靠经验实现空调机组维护的现状,其主要是利用修正的理论模型作为基准,和市场在线运行机组的运行状态,做实时比较,并对在线机组运行状态和基准不断做比较,再使用AI(Artificial Intelligence,人工智 能)机器学习的方法,预测未来空调机组的运行状态,并根据设定的可靠性或经济性条件,给出合理的维护时间,实现以数据分析为支持的智能化维护,从而提高维护的准确度,优化维护的措施。
在空调机组运行过程中可以使用本地采集存储或云端采集存储等方式采集空调机组的至少一个实际性能指标数据和实际非性能指标数据。例如,采集的实际非性能指标数据可以是室外环境温度、压缩机输出、回风温湿度、室内风机转速和室外风机转速,实际性能指标数据可以是HP和P。
在得到采集的实际数据后,将所述实际非性能指标数据输入到基准模型,得到所述基准模型输出的预测基准性能指标数据。
例如,将上述采集的压缩机输出、室内风机转速、室外风机转速、回风温湿度和室外环境温度分别输入到用于计算基准HP和基准P的基准模型内,便可以得到对应的基准模型输出的预测基准HP和预测基准P。
为了得到HP的基准模型,获取中心服务器中最近3个月的多条历史数据,每条历史数据包括室外环境温度、压缩机输出、回风温湿度、室内风机转速、室外风机转速和基准HP,对这些数据进行数据清洗,得到训练样本,基于所述训练样本,通过机器学习的方法创建可以用于计算连续数据的基准模型。
为了得到P的基准模型,获取中心服务器中最近3个月的多条历史数据,每条历史数据包括室外环境温度、压缩机输出、回风温湿度、室内风机转速、室外风机转速和基准P,对这些数据进行数据清洗,得到训练样本,基于所述训练样本,通过机器学习的方法创建可以用于计算连续数据的基准模型。
上述基准模型为利用训练样本中的非性能指标数据作为输入,以输出所述训练样本中对应的基准性能指标数据而训练得到的模型,所述基准性能指标数据为仅根据空调机组的运行时长确定的数据。创建基准模型的目的是空调机组的运行状态除了受运行时长的影响外,还有机组本身的输出,例如室外环境温度、压缩机输出、回风温湿度、室内外风机转速相关,创建基准模型,主要是为解除和机组本身的输出关联,只分析和运行时长的关系。
所述训练样本可以通过以下方式获取:
从中心服务器获取预设时间段内所述空调机组的多条历史数据,每条历史数据包括至少一个实际性能指标数据、实际非性能指标数据及预测基准性能指标数据;
若将上述获取的多条历史数据都输入到基准模型中,势必效率低下,以此为了剔除不稳定的运行数据,对所述多条历史数据按照数据规则进行数据清洗,清洗后得到训练样本。
上述对所述多条历史数据按照数据规则进行数据清洗,包括如下至少一个步骤:
在excel文本中有相关的标志位,根据空调机组状态的标志位,选择标志位是空调运行状态的多条历史数据,所述空调机组状态的标志位包括空调机组未运行状态、空调机组开启状态和空调机组运行结束状态;
考虑室外环境温度变化不会太快,根据设定的时间间隔,选取对应所述时间间隔采集的多条历史数据,例如选取对应间隔1小时采集的多条历史数据;
根据设定的室外环境温度差值,选取相邻室外环境温差大于所述设定的室外环境温度差值的多条历史数据,例如选取相邻室外环境温差大于1摄氏度的多条历史数据。
在获取训练样本之后,将训练样本中的非性能指标数据输入到基准模型,得到基准模型输出的预测基准性能指标数据,根据训练样本中的基准性能指标数据与所述预测基准性能指标数据的偏差在0.2%以内,一致性较高,确定基准模型创建完成。
在得到预测基准性能指标数据之后,确定所述实际性能指标数据与对应预测基准性能指标数据的差值,并根据预先拟合的性能指标数据差值与运行时间的关系,确定所述空调机组的维护方案。
可以通过以下方法预先拟合的性能指标数据差值与运行时间的关系:
从中心服务器获取预设时间段内所述空调机组的多条历史数据,每条历史数据包括至少一个实际性能指标数据、实际非性能指标数据、预测基准性能指标数据及运行时间;
确定所述实际性能指标数据与对应预测基准性能指标数据的差值;
根据各条历史数据对应的差值及运行时间,拟合性能指标数据差值与运行时间的关系,其中,可以利用机器学习等方法,拟合性能指标数据差值与运行时间的关系曲线,例如使用XGBoost(eXtreme Gradient Boosting,极限梯度提升)算法建立性能指标数据差值与运行时间的模型。
当性能指标数据为HP时,将实际HP与所采集的对应的预测基准HP做差值,可以得到高压压力差值(ΔHP);根据各条历史数据对应的ΔHP及运行时间,拟合ΔHP与运行时间的关系。
当性能指标数据为P时,将实际P与所采集的对应的预测基准P做差值,可以得到功率差值(ΔP);根据各条历史数据对应的ΔP及运行时间,拟合ΔP与运行时间的关系。
在拟合性能指标数据差值与运行时间的关系之后,还可以根据后续实时数据的采集,不断地更新拟合的性能指标数据差值与运行时间的关系,做到了自学习,提高了预测的准确度,其具体步骤包括:
获取多条实时数据,每条实时数据包括所述实际性能指标数据、预测基准性能指标数据及实际运行时间;
确定所述实际性能指标数据与对应的预测基准性能指标数据的差值;
根据各条实时数据对应的差值及实际运行时间,更新拟合的性能指标数据差值与运行时间的关系。
根据预先拟合的性能指标数据差值与运行时间的关系,可以通过以下方法确定空调机组的维护方案:
根据预先拟合的ΔHP与运行时间的第一关系曲线和设定的ΔHP阈值,确定当前时间距离下次维护时间的第一间隔天数;
根据预先拟合的ΔP与运行时间的第二关系曲线和预先确定的目标维护间隔,确定当前时间距离下次维护时间的第二间隔天数;
根据第一间隔天数和第二间隔天数的较小值确定最终的维护间隔。
具体的,可以直接将第一间隔天数或第二间隔天数作为最终的维护间隔; 也可以当第一间隔天数大于第二间隔天数时,从可靠性的角度考虑,将第一间隔天数作为最终的维护间隔,或者从经济性的角度考虑,将第二间隔天数作为最终的维护间隔;还可以当第一间隔天数小于第二间隔天数时,从可靠性的角度考虑,将第一间隔天数作为最终的维护间隔。
从可靠性的角度出发,由于HP偏高会触发空调机组的自动保护,即空调机组自动关机,在一段时间内多次触发,会导致空调机组锁定,不能正常运行。因此,根据预先拟合的ΔHP与运行时间的关系,确定当前时间距离下次维护时间的第一间隔天数,以实现提前预警的作用,其具体步骤如下:
根据预先拟合的ΔHP与运行时间的第一关系曲线,确定当前ΔHP对应的第一运行时间;
其中,将采集的当前非性能指标数据带入基准模型,获得对应的当前预测基准HP,则所述当前HP与当前预测基准HP的差值为当前ΔHP。
根据所述第一关系曲线确定ΔHP阈值对应的第二运行时间;
根据所述第一运行时间和第二运行时间的差值,确定当前时间距离下次维护时间的第一间隔天数。
例如,当前ΔHP对应的第一运行时间是3天,根据所述第一关系曲线确定ΔHP阈值对应的第二运行时间是7天,则当前时间距离下次维护时间的第一间隔天数为4天。
从经济性的角度出发,空调机组在运行过程中,受环境的影响,和基准P的差值ΔP总是存在,何时进行维护比较合适,可以考虑维护费用和电能消耗费用的关系。因此,确定当前时间距离下次维护时间的第二间隔天数的具体步骤如下:
获取预先确定的目标维护间隔;
根据当前ΔP在所述第二关系曲线对应的目标运行时间和所述目标维护间隔,即将所述目标维护间隔与当前ΔP在所述第二关系曲线对应的目标运行时间相减,确定当前时间距离下次维护时间的第二间隔天数;
其中,将采集的当前非性能指标数据带入基准模型,获得对应的当前预测 基准P,则所述当前P与当前预测基准P的差值为当前ΔP。
上述目标维护间隔可以根据以下方法确定:
预先根据预先拟合的ΔP与运行时间的第二关系曲线,及基准P与运行时间的第三关系曲线,确定按照不同时间间隔进行维护时预设时段的电能消耗费用;
确定按照不同时间间隔维护时,在预设时段的维护费用;
确定预设时段内维护费用与电能消耗费用总和最小对应的时间间隔为所述目标维护间隔。
其中,预设时段的电能消耗费用可以通过下述方法确定:
在预设时段内,假设每次维护后运行时长重置为零,确定不同维护间隔对应的各运行时长,空调机组在每次维护后,假设空调机组的性能恢复到运行时间为0的时候,这时的P也为0;
例如,当维护间隔为2天,则在4天中的第2天以及第4天进行了维护,由于每次维护后,运行时长重置为0,则第一个维护间隔,即第1至2天的实际P与第二个维护间隔,即第3至4天的实际P相同。
根据预先拟合的ΔP与运行时间的第二关系曲线,及基准P与运行时间的第三关系曲线,确定同一运行时长对应的实际P;
通过将空调机组在实际运行过程中采集的多条实际非性能指标数据输入到基准模型,可以获得对应的预测基准P。根据所述获得的预测基准P与其对应的运行时间,可以确定基准P与运行时间的第三关系曲线。
当上述的预设时段包括未来的一段时间时,需要根据基准P与运行时间的第三关系曲线,确定所述预设时段内的基准P,根据预先拟合的ΔP与运行时间的第二关系曲线,确定所述预设时段内的ΔP。将所述预设时段内的基准P和ΔP相加,得到同一运行时长对应的实际P。
根据各运行时长和对应的实际P及电价,确定电能消耗费用。
实施例2
基于相同的发明构思,本公开实施例还提供一种空调机组预测维护装置,由于该装置即是本公开实施例中的方法中的装置,并且该装置解决问题的原理 与该方法相似,因此该装置的实施可以参见方法的实施,重复之处不再赘述。
如图3所示,上述装置包括以下模块:
数据采集模块301,用于在空调机组运行过程中进行数据采集,所采集的数据包括至少一个实际性能指标数据及实际非性能指标数据;
预测基准性能指标数据获得模块302,用于将所述实际非性能指标数据输入到基准模型,得到所述基准模型输出的预测基准性能指标数据,所述基准模型为利用训练样本中的非性能指标数据作为输入,以输出所述训练样本中对应的基准性能指标数据而训练得到的模型,所述基准性能指标数据为仅根据空调机组的运行时长确定的数据;
维护方案确定模块303,用于确定所述实际性能指标数据与对应预测基准性能指标数据的差值,并根据预先拟合的性能指标数据差值与运行时间的关系,确定所述空调机组的维护方案。
作为一种可选的实施方式,预测基准性能指标数据获得模块通过如下方式获取所述训练样本:
从中心服务器获取预设时间段内所述空调机组的多条历史数据,每条历史数据包括至少一个实际性能指标数据、实际非性能指标数据及预测基准性能指标数据;
对所述多条历史数据按照数据规则进行数据清洗,清洗后得到训练样本。
作为一种可选的实施方式,所述预测基准性能指标数据获得模块,用于对所述多条历史数据按照数据规则进行数据清洗,包括如下至少一个步骤:
根据空调机组状态的标志位,选择标志位是空调运行状态的多条历史数据;
根据设定的时间间隔,选取对应所述时间间隔采集的多条历史数据;
根据设定的室外环境温度差值,选取相邻室外环境温差大于所述设定的室外环境温度差值的多条历史数据。
作为一种可选的实施方式,所述维护方案确定模块,用于预先拟合的性能指标数据差值与运行时间的关系,包括:
从中心服务器获取预设时间段内所述空调机组的多条历史数据,每条历史 数据包括至少一个实际性能指标数据、实际非性能指标数据、预测基准性能指标数据及运行时间;
确定所述实际性能指标数据与对应预测基准性能指标数据的差值;
根据各条历史数据对应的差值及运行时间,拟合性能指标数据差值与运行时间的关系。
作为一种可选的实施方式,所采集的数据还包括实际运行时间,所述装置还包括:
实时数据获取模块,用于获取多条实时数据,每条实时数据包括所述实际性能指标数据、预测基准性能指标数据及实际运行时间;
差值确定模块,用于确定所述实际性能指标数据与对应预测基准性能指标数据的差值;
关系更新模块,用于根据各条实时数据对应的差值及实际运行时间,更新拟合的性能指标数据差值与运行时间的关系。
作为一种可选的实施方式,所述性能指标数据差值包括ΔHP和ΔP,维护方案确定模块,用于根据预先拟合的性能指标数据差值与运行时间的关系,确定所述空调机组的维护方案,包括:
根据预先拟合的ΔHP与运行时间的第一关系曲线和设定的ΔHP阈值,确定当前时间距离下次维护时间的第一间隔天数;
根据预先拟合的ΔP与运行时间的第二关系曲线和预先确定的目标维护间隔,确定当前时间距离下次维护时间的第二间隔天数;
根据第一间隔天数和第二间隔天数的较小值确定最终的维护间隔。
作为一种可选的实施方式,所述维护方案确定模块,用于根据预先拟合的ΔHP与运行时间的第一关系曲线和设定的ΔHP阈值,确定当前时间距离下次维护时间的第一间隔天数,包括:
根据预先拟合的ΔHP与运行时间的第一关系曲线,确定当前ΔHP对应的第一运行时间;
根据所述第一关系曲线确定ΔHP阈值对应的第二运行时间;
根据所述第一运行时间和第二运行时间的差值,确定当前时间距离下次维护时间的第一间隔天数。
作为一种可选的实施方式,所述维护方案确定模块,用于根据预先拟合的ΔP与运行时间的第二关系曲线和预先确定的目标维护间隔,确定当前时间距离下次维护时间的第二间隔天数,包括:
获取预先确定的目标维护间隔,所述目标维护间隔为预设时段内维护费用与电能消耗费用总和最小对应的时间间隔,其中,预先根据预先拟合的ΔP与运行时间的第二关系曲线,及基准P与运行时间的第三关系曲线,确定按照不同时间间隔进行维护时预设时段的电能消耗费用;确定按照不同时间间隔维护时,在预设时段的维护费用,根据所述预设时段的电能消耗费用和维护费用确定所述目标维护间隔;
根据当前ΔP在所述第二关系曲线对应的目标运行时间和所述目标维护间隔,确定当前时间距离下次维护时间的第二间隔天数。
作为一种可选的实施方式,所述维护方案确定模块,用于根据预先拟合的ΔP与运行时间的第二关系曲线,及基准P与运行时间的第三关系曲线,确定按照不同时间间隔进行维护时预设时段的电能消耗费用,包括:
在预设时段内,假设每次维护后运行时长重置为零,确定不同维护间隔对应的各运行时长;
根据预先拟合的ΔP与运行时间的第二关系曲线,及基准P与运行时间的第三关系曲线,确定同一运行时长对应的实际P;
根据各运行时长和对应的实际P及电价,确定电能消耗费用。
实施例3
基于相同的发明构思,本公开实施例中还提供了一种空调机组预测维护电子设备,由于该电子设备即是本公开实施例中的方法中的电子设备,并且该电子设备解决问题的原理与该方法相似,因此该电子设备的实施可以参见方法的实施,重复之处不再赘述。
下面参照图4来描述根据本公开的这种实施方式的电子设备40。图4显示 的电子设备40仅仅是一个示例,不应对本公开实施例的功能和使用范围带来任何限制。
如图4所示,电子设备40可以以通用计算设备的形式表现,例如其可以为终端设备。电子设备40的组件可以包括但不限于:上述至少一个处理器41、上述至少一个存储处理器可执行指令的存储器42、连接不同系统组件(包括存储器42和处理器41)的总线43。
所述处理器通过运行所述可执行指令以实现如下步骤:
在空调机组运行过程中进行数据采集,所采集的数据包括至少一个实际性能指标数据及实际非性能指标数据;
将所述实际非性能指标数据输入到基准模型,得到所述基准模型输出的预测基准性能指标数据,所述基准模型为利用训练样本中的非性能指标数据作为输入,以输出所述训练样本中对应的基准性能指标数据而训练得到的模型,所述基准性能指标数据为仅根据空调机组的运行时长确定的数据;
确定所述预测基准性能指标数据与对应实际性能指标数据的差值,并根据预先拟合的性能指标数据差值与运行时间的关系,确定所述空调机组的维护方案。
所述处理器还通过运行所述可执行指令以实现实施例1中的空调机组预测维护方法,重复之处不再赘述。
存储器42可以包括易失性存储器形式的可读介质,例如随机存取存储器(RAM)421和/或高速缓存存储器422,还可以进一步包括只读存储器(ROM)423。
存储器42还可以包括具有一组(至少一个)程序模块424的程序/实用工具425,这样的程序模块424包括但不限于:操作系统、一个或者多个应用程序、其它程序模块以及程序数据,这些示例中的每一个或某种组合中可能包括网络环境的实现。
电子设备40也可以与一个或多个外部设备44(例如键盘、指向设备等)通信,还可与一个或者多个使得用户能与电子设备40交互的设备通信,和/或与使 得电子设备40能与一个或多个其它计算设备进行通信的任何设备(例如路由器、调制解调器等等)通信。这种通信可以通过输入/输出(I/O)接口45进行。并且,电子设备40还可以通过网络适配器46与一个或者多个网络(例如局域网(LAN),广域网(WAN)和/或公共网络,例如因特网)通信。如图所示,网络适配器46通过总线43与电子设备40的其它模块通信。应当明白,尽管图中未示出,可以结合电子设备40使用其它硬件和/或软件模块,包括但不限于:微代码、设备驱动器、冗余处理单元、外部磁盘驱动阵列、RAID系统、磁带驱动器以及数据备份存储系统等。
实施例4
在一些可能的实施方式中,本公开的各个方面还可以实现为一种程序产品的形式,其包括程序代码,当程序产品在终端设备上运行时,程序代码用于使终端设备执行本说明书上述“示例性方法”部分中描述的根据本公开各种示例性实施方式的智能化维护装置中各模块的步骤,例如,终端设备可以用于在空调机组运行过程中进行数据采集,所采集的数据包括至少一个实际性能指标数据及实际非性能指标数据;将所述实际非性能指标数据输入到基准模型,得到所述基准模型输出的预测基准性能指标数据,所述基准模型为利用训练样本中的非性能指标数据作为输入,以输出所述训练样本中对应的基准性能指标数据而训练得到的模型,所述基准性能指标数据为仅根据空调机组的运行时长确定的数据;确定所述实际性能指标数据与对应预测基准性能指标数据的差值,并根据预先拟合的性能指标数据差值与运行时间的关系,确定所述空调机组的维护方案等操作。
程序产品可以采用一个或多个可读介质的任意组合。可读介质可以是可读信号介质或者可读存储介质。可读存储介质例如可以是——但不限于——电、磁、光、电磁、红外线、或半导体的系统、装置或器件,或者任意以上的组合。可读存储介质的更具体的例子(非穷举的列表)包括:具有一个或多个导线的电连接、便携式盘、硬盘、随机存取存储器(RAM)、只读存储器(ROM)、可擦式可编程只读存储器(EPROM或闪存)、光纤、便携式紧凑盘只读存储器 (CD-ROM)、光存储器件、磁存储器件、或者上述的任意合适的组合。
如图5所示,描述了根据本公开的实施方式的用于空调机组预测维护的程序产品50,其可以采用便携式紧凑盘只读存储器(CD-ROM)并包括程序代码,并可以在终端设备,例如个人电脑上运行。然而,本公开的程序产品不限于此,在本文件中,可读存储介质可以是任何包含或存储程序的有形介质,该程序可以被指令执行系统、装置或者器件使用或者与其结合使用。
可读信号介质可以包括在基带中或者作为载波一部分传播的数据信号,其中承载了可读程序代码。这种传播的数据信号可以采用多种形式,包括——但不限于——电磁信号、光信号或上述的任意合适的组合。可读信号介质还可以是可读存储介质以外的任何可读介质,该可读介质可以发送、传播或者传输用于由指令执行系统、装置或者器件使用或者与其结合使用的程序。
可读介质上包含的程序代码可以用任何适当的介质传输,包括——但不限于——无线、有线、光缆、RF等等,或者上述的任意合适的组合。
可以以一种或多种程序设计语言的任意组合来编写用于执行本公开操作的程序代码,程序设计语言包括面向对象的程序设计语言—诸如Java、C++等,还包括常规的过程式程序设计语言—诸如“C”语言或类似的程序设计语言。程序代码可以完全地在用户计算设备上执行、部分地在用户设备上执行、作为一个独立的软件包执行、部分在用户计算设备上部分在远程计算设备上执行、或者完全在远程计算设备或服务器上执行。在涉及远程计算设备的情形中,远程计算设备可以通过任意种类的网络——包括局域网(LAN)或广域网(WAN)—连接到用户计算设备,或者,可以连接到外部计算设备(例如利用因特网服务提供商来通过因特网连接)。
应当注意,尽管在上文详细描述中提及了系统的若干模块或子模块,但是这种划分仅仅是示例性的并非强制性的。实际上,根据本公开的实施方式,上文描述的两个或更多模块的特征和功能可以在一个模块中具体化。反之,上文描述的一个模块的特征和功能可以进一步划分为由多个模块来具体化。
此外,尽管在附图中以特定顺序描述了本公开系统各模块的操作,但是, 这并非要求或者暗示必须按照该特定顺序来执行这些操作,或是必须执行全部所示的操作才能实现期望的结果。附加地或备选地,可以省略某些操作,将多个操作合并为一个操作执行,和/或将一个操作分解为多个操作执行。
本领域内的技术人员应明白,本公开的实施例可提供为方法、系统、或计算机程序产品。因此,本公开可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。而且,本公开可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器和光学存储器等)上实施的计算机程序产品的形式。
本公开是参照根据本公开实施例的方法、设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器,使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的设备。
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理设备以特定方式工作的计算机可读存储器中,使得存储在该计算机可读存储器中的指令产生包括指令设备的制造品,该指令设备实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些计算机程序指令也可装载到计算机或其他可编程数据处理设备上,使得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机实现的处理,从而在计算机或其他可编程设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
本领域技术人员在考虑说明书及实践这里公开的发明后,将容易想到本公开的其它实施方案。本申请旨在涵盖本公开的任何变型、用途或者适应性变化,这些变型、用途或者适应性变化遵循本公开的一般性原理并包括本公开未公开 的本技术领域中的公知常识或惯用技术手段。说明书和实施例仅被视为示例性的,本公开的真正范围和精神由下面的权利要求指出。
应当理解的是,本公开并不局限于上面已经描述并在附图中示出的精确结构,并且可以在不脱离其范围进行各种修改和改变。本公开的范围仅由所附的权利要求来限制。

Claims (20)

  1. 一种空调机组预测维护方法,其特征在于,该方法包括:
    在空调机组运行过程中进行数据采集,所采集的数据包括至少一个实际性能指标数据及实际非性能指标数据;
    将所述实际非性能指标数据输入到基准模型,得到所述基准模型输出的预测基准性能指标数据,所述基准模型为利用训练样本中的非性能指标数据作为输入,以输出所述训练样本中对应的基准性能指标数据而训练得到的模型,所述基准性能指标数据为仅根据空调机组的运行时长确定的数据;
    确定所述实际性能指标数据与对应预测基准性能指标数据的差值,并根据预先拟合的性能指标数据差值与运行时间的关系,确定所述空调机组的维护方案。
  2. 根据权利要求1所述的方法,其特征在于,通过如下方式获取所述训练样本:
    从中心服务器获取预设时间段内所述空调机组的多条历史数据,每条历史数据包括至少一个实际性能指标数据、实际非性能指标数据及预测基准性能指标数据;
    对所述多条历史数据按照数据规则进行数据清洗,清洗后得到训练样本。
  3. 根据权利要求2所述的方法,其特征在于,所述对所述多条历史数据按照数据规则进行数据清洗,包括如下至少一个步骤:
    根据空调机组状态的标志位,选择标志位是空调运行状态的多条历史数据;
    根据设定的时间间隔,选取对应所述时间间隔采集的多条历史数据;
    根据设定的室外环境温度差值,选取相邻室外环境温差大于所述设定的室外环境温度差值的多条历史数据。
  4. 根据权利要求1所述的方法,其特征在于,所述预先拟合的性能指标数据差值与运行时间的关系,包括:
    从中心服务器获取预设时间段内所述空调机组的多条历史数据,每条历史数据包括至少一个实际性能指标数据、实际非性能指标数据、预测基准性能指标数据及运行时间;
    确定所述实际性能指标数据与对应预测基准性能指标数据的差值;
    根据各条历史数据对应的差值及运行时间,拟合性能指标数据差值与运行时间的关系。
  5. 根据权利要求4所述的方法,其特征在于,所采集的数据还包括实际运行时间,拟合性能指标数据差值与运行时间的关系之后,还包括:
    获取多条实时数据,每条实时数据包括所述实际性能指标数据、预测基准性能指标数据及实际运行时间;
    确定所述实际性能指标数据与对应预测基准性能指标数据的差值;
    根据各条实时数据对应的差值及实际运行时间,更新拟合的性能指标数据差值与运行时间的关系。
  6. 根据权利要求1所述的方法,其特征在于,所述性能指标数据差值包括高压压力差值ΔHP和功率差值ΔP,根据预先拟合的性能指标数据差值与运行时间的关系,确定所述空调机组的维护方案,包括:
    根据预先拟合的ΔHP与运行时间的第一关系曲线和设定的ΔHP阈值,确定当前时间距离下次维护时间的第一间隔天数;
    根据预先拟合的ΔP与运行时间的第二关系曲线和预先确定的目标维护间隔,确定当前时间距离下次维护时间的第二间隔天数;
    根据第一间隔天数和第二间隔天数的较小值确定最终的维护间隔。
  7. 根据权利要求6所述的方法,其特征在于,所述根据预先拟合的ΔHP与运行时间的第一关系曲线和设定的ΔHP阈值,确定当前时间距离下次维护时间的第一间隔天数,包括:
    根据预先拟合的ΔHP与运行时间的第一关系曲线,确定当前ΔHP对应的第一运行时间;
    根据所述第一关系曲线确定ΔHP阈值对应的第二运行时间;
    根据所述第一运行时间和第二运行时间的差值,确定当前时间距离下次维护时间的第一间隔天数。
  8. 根据权利要求6所述的方法,其特征在于,所述根据预先拟合的ΔP与运行时间的第二关系曲线和预先确定的目标维护间隔,确定当前时间距离下次维护时间的第二间隔天数,包括:
    获取预先确定的目标维护间隔,所述目标维护间隔为预设时段内维护费用与电能消耗费用总和最小对应的时间间隔,其中,预先根据预先拟合的ΔP与运行时间的第二关系曲线,及基准P与运行时间的第三关系曲线,确定按照不同时间间隔进行维护时预设时段的电能消耗费用;确定按照不同时间间隔维护时,在预设时段的维护费用,根据所述预设时段的电能消耗费用和维护费用确定所述目标维护间隔;
    根据当前ΔP在所述第二关系曲线对应的目标运行时间和所述目标维护间隔,确定当前时间距离下次维护时间的第二间隔天数。
  9. 根据权利要求8所述的方法,其特征在于,所述根据预先拟合的ΔP与运行时间的第二关系曲线,及基准P与运行时间的第三关系曲线,确定按照不同时间间隔进行维护时预设时段的电能消耗费用,包括:
    在预设时段内,假设每次维护后运行时长重置为零,确定不同维护间隔对应的各运行时长;
    根据预先拟合的ΔP与运行时间的第二关系曲线,及基准P与运行时间的第三关系曲线,确定同一运行时长对应的实际P;
    根据各运行时长和对应的实际P及电价,确定电能消耗费用。
  10. 一种空调机组预测维护装置,其特征在于,所述装置包括:
    数据采集模块,用于在空调机组运行过程中进行数据采集,所采集的数据包括至少一个实际性能指标数据及实际非性能指标数据;
    预测基准性能指标数据获得模块,用于将所述实际非性能指标数据输入到基准模型,得到所述基准模型输出的预测基准性能指标数据,所述基准模型为利用训练样本中的非性能指标数据作为输入,以输出所述训练样本中对应的基 准性能指标数据而训练得到的模型,所述基准性能指标数据为仅根据空调机组的运行时长确定的数据;
    维护方案确定模块,用于确定所述实际性能指标数据与对应预测基准性能指标数据的差值,并根据预先拟合的性能指标数据差值与运行时间的关系,确定所述空调机组的维护方案。
  11. 根据权利要求10所述的装置,其特征在于,预测基准性能指标数据获得模块通过如下方式获取所述训练样本:
    从中心服务器获取预设时间段内所述空调机组的多条历史数据,每条历史数据包括至少一个实际性能指标数据、实际非性能指标数据及预测基准性能指标数据;
    对所述多条历史数据按照数据规则进行数据清洗,清洗后得到训练样本。
  12. 根据权利要求11所述的装置,其特征在于,所述预测基准性能指标数据获得模块,用于对所述多条历史数据按照数据规则进行数据清洗,包括如下至少一个步骤:
    根据空调机组状态的标志位,选择标志位是空调运行状态的多条历史数据;
    根据设定的时间间隔,选取对应所述时间间隔采集的多条历史数据;
    根据设定的室外环境温度差值,选取相邻室外环境温差大于所述设定的室外环境温度差值的多条历史数据。
  13. 根据权利要求10所述的装置,其特征在于,所述维护方案确定模块,用于预先拟合的性能指标数据差值与运行时间的关系,包括:
    从中心服务器获取预设时间段内所述空调机组的多条历史数据,每条历史数据包括至少一个实际性能指标数据、实际非性能指标数据、预测基准性能指标数据及运行时间;
    确定所述实际性能指标数据与对应预测基准性能指标数据的差值;
    根据各条历史数据对应的差值及运行时间,拟合性能指标数据差值与运行时间的关系。
  14. 根据权利要求13所述的装置,其特征在于,所采集的数据还包括实际 运行时间,所述装置还包括:
    实时数据获取模块,用于获取多条实时数据,每条实时数据包括所述实际性能指标数据、预测基准性能指标数据及实际运行时间;
    差值确定模块,用于确定所述实际性能指标数据与对应预测基准性能指标数据的差值;
    关系更新模块,用于根据各条实时数据对应的差值及实际运行时间,更新拟合的性能指标数据差值与运行时间的关系。
  15. 根据权利要求10所述的装置,其特征在于,所述性能指标数据差值包括高压压力差值ΔHP和功率差值ΔP,维护方案确定模块,用于根据预先拟合的性能指标数据差值与运行时间的关系,确定所述空调机组的维护方案,包括:
    根据预先拟合的ΔHP与运行时间的第一关系曲线和设定的ΔHP阈值,确定当前时间距离下次维护时间的第一间隔天数;
    根据预先拟合的ΔP与运行时间的第二关系曲线和预先确定的目标维护间隔,确定当前时间距离下次维护时间的第二间隔天数;
    根据第一间隔天数和第二间隔天数的较小值确定最终的维护间隔。
  16. 根据权利要求15所述的装置,其特征在于,所述维护方案确定模块,用于根据预先拟合的ΔHP与运行时间的第一关系曲线和设定的ΔHP阈值,确定当前时间距离下次维护时间的第一间隔天数,包括:
    根据预先拟合的ΔHP与运行时间的第一关系曲线,确定当前ΔHP对应的第一运行时间;
    根据所述第一关系曲线确定ΔHP阈值对应的第二运行时间;
    根据所述第一运行时间和第二运行时间的差值,确定当前时间距离下次维护时间的第一间隔天数。
  17. 根据权利要求15所述的装置,其特征在于,所述维护方案确定模块,用于根据预先拟合的ΔP与运行时间的第二关系曲线和预先确定的目标维护间隔,确定当前时间距离下次维护时间的第二间隔天数,包括:
    获取预先确定的目标维护间隔,所述目标维护间隔为预设时段内维护费用 与电能消耗费用总和最小对应的时间间隔,其中,预先根据预先拟合的ΔP与运行时间的第二关系曲线,及基准P与运行时间的第三关系曲线,确定按照不同时间间隔进行维护时预设时段的电能消耗费用;确定按照不同时间间隔维护时,在预设时段的维护费用,根据所述预设时段的电能消耗费用和维护费用确定所述目标维护间隔;
    根据当前ΔP在所述第二关系曲线对应的目标运行时间和所述目标维护间隔,确定当前时间距离下次维护时间的第二间隔天数。
  18. 根据权利要求17所述的装置,其特征在于,所述维护方案确定模块,用于根据预先拟合的ΔP与运行时间的第二关系曲线,及基准P与运行时间的第三关系曲线,确定按照不同时间间隔进行维护时预设时段的电能消耗费用,包括:
    在预设时段内,假设每次维护后运行时长重置为零,确定不同维护间隔对应的各运行时长;
    根据预先拟合的ΔP与运行时间的第二关系曲线,及基准P与运行时间的第三关系曲线,确定同一运行时长对应的实际P;
    根据各运行时长和对应的实际P及电价,确定电能消耗费用。
  19. 一种电子设备,其特征在于,包括:处理器;用于存储处理器可执行指令的存储器;其中,所述处理器通过运行所述可执行指令以实现权利要求1至9任一项所述方法的步骤。
  20. 一种计算机可读写存储介质,其上存储有计算机指令,其特征在于,该指令被处理器执行时实现权利要求1至9任一项所述方法的步骤。
PCT/CN2022/084760 2021-12-16 2022-04-01 一种空调机组预测维护方法、装置、设备及介质 WO2023108953A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202111545585.1 2021-12-16
CN202111545585.1A CN116266319A (zh) 2021-12-16 2021-12-16 一种空调机组预测维护方法、装置、设备及介质

Publications (1)

Publication Number Publication Date
WO2023108953A1 true WO2023108953A1 (zh) 2023-06-22

Family

ID=86742814

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/084760 WO2023108953A1 (zh) 2021-12-16 2022-04-01 一种空调机组预测维护方法、装置、设备及介质

Country Status (2)

Country Link
CN (1) CN116266319A (zh)
WO (1) WO2023108953A1 (zh)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110736509A (zh) * 2019-11-13 2020-01-31 格力电器(武汉)有限公司 一种空调外机测试方法、系统、存储介质及上位机
CN112036581A (zh) * 2019-05-15 2020-12-04 上海杰之能软件科技有限公司 交通工具空调系统的性能检测方法及装置、存储介质、终端
CN112200479A (zh) * 2020-10-20 2021-01-08 厦门航空有限公司 民机空调系统散热器性能量化评估方法
WO2021063033A1 (zh) * 2019-09-30 2021-04-08 北京国双科技有限公司 空调能耗模型训练方法与空调系统控制方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112036581A (zh) * 2019-05-15 2020-12-04 上海杰之能软件科技有限公司 交通工具空调系统的性能检测方法及装置、存储介质、终端
WO2021063033A1 (zh) * 2019-09-30 2021-04-08 北京国双科技有限公司 空调能耗模型训练方法与空调系统控制方法
CN110736509A (zh) * 2019-11-13 2020-01-31 格力电器(武汉)有限公司 一种空调外机测试方法、系统、存储介质及上位机
CN112200479A (zh) * 2020-10-20 2021-01-08 厦门航空有限公司 民机空调系统散热器性能量化评估方法

Also Published As

Publication number Publication date
CN116266319A (zh) 2023-06-20

Similar Documents

Publication Publication Date Title
CN113835344B (zh) 设备的控制优化方法、展示平台、云服务器及存储介质
Fan et al. Temporal knowledge discovery in big BAS data for building energy management
Yu et al. A novel methodology for knowledge discovery through mining associations between building operational data
Ke et al. Cloud computing platform for real-time measurement and verification of energy performance
US20190302713A1 (en) Method and device for automatically diagnosing and controlling apparatus in intelligent building
US8321188B2 (en) Weather-related energy-usage analysis
US8086544B2 (en) Analysis of energy-related factors for selecting computational job locations
CN107860102B (zh) 一种控制中央空调的方法及装置
KR101448453B1 (ko) 건물 에너지에 대한 예측제어, 실시간제어, 운영 및 관리에 따른 통합 솔루션 시스템
US10498585B2 (en) Sensor data analytics and alarm management
US20150323948A1 (en) System for analyzing building energy consumption information
WO2023193458A1 (zh) 基于数字孪生的产线优化方法、装置、电子设备及介质
KR101762061B1 (ko) 클라우드 기반 에너지 사용량 예측 시스템 및 방법
CN103853106A (zh) 一种建筑能源供应设备的能耗预测参数优化方法
CN108320042B (zh) 循环水系统的优化方法与装置
JP2012190442A (ja) データ・センタの効率分析及び最適化のための方法、システム、コンピュータ・プログラム
Pawlish et al. Free cooling: A paradigm shift in data centers
CN111486555A (zh) 一种人工智能ai专家系统进行中央空调节能调控的方法
CN110160210B (zh) 风机盘管的故障检测方法、装置、存储介质和电子设备
Stanica et al. A methodology to support the decision-making process for energy retrofitting at district scale
CN115796393A (zh) 基于多能互动的能源网管理优化方法、系统及存储介质
Lei et al. Prediction method of energy consumption for high building based on LMBP neural network
CN113790516B (zh) 中央空调制冷站全局优化节能控制方法、系统和电子设备
WO2023108953A1 (zh) 一种空调机组预测维护方法、装置、设备及介质
Crowe et al. From theory to practice: Lessons learned from an advanced M&V Commercial Pilot

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22905723

Country of ref document: EP

Kind code of ref document: A1