WO2025237017A1 - 异常检测方法、异常检测装置、电器和可读存储介质 - Google Patents

异常检测方法、异常检测装置、电器和可读存储介质

Info

Publication number
WO2025237017A1
WO2025237017A1 PCT/CN2025/090513 CN2025090513W WO2025237017A1 WO 2025237017 A1 WO2025237017 A1 WO 2025237017A1 CN 2025090513 W CN2025090513 W CN 2025090513W WO 2025237017 A1 WO2025237017 A1 WO 2025237017A1
Authority
WO
WIPO (PCT)
Prior art keywords
deviation rate
preset
deviation
anomaly detection
current
Prior art date
Legal status (The legal status 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 status listed.)
Pending
Application number
PCT/CN2025/090513
Other languages
English (en)
French (fr)
Inventor
李冬冬
李斌
郑春元
姬学欢
赵浩伟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Midea Group Co Ltd
GD Midea Heating and Ventilating Equipment Co Ltd
Original Assignee
Midea Group Co Ltd
GD Midea Heating and Ventilating Equipment Co Ltd
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 Midea Group Co Ltd, GD Midea Heating and Ventilating Equipment Co Ltd filed Critical Midea Group Co Ltd
Publication of WO2025237017A1 publication Critical patent/WO2025237017A1/zh
Pending legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F11/00—Control or safety arrangements
    • F24F11/30—Control or safety arrangements for purposes related to the operation of the system, e.g. for safety or monitoring
    • F24F11/32—Responding to malfunctions or emergencies
    • F24F11/38—Failure diagnosis
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F04—POSITIVE - DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS FOR LIQUIDS OR ELASTIC FLUIDS
    • F04B—POSITIVE-DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS
    • F04B49/00—Control, e.g. of pump delivery, or pump pressure of, or safety measures for, machines, pumps, or pumping installations, not otherwise provided for, or of interest apart from, groups F04B1/00 - F04B47/00
    • F04B49/06—Control using electricity
    • F04B49/065—Control using electricity and making use of computers
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F04—POSITIVE - DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS FOR LIQUIDS OR ELASTIC FLUIDS
    • F04B—POSITIVE-DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS
    • F04B51/00—Testing machines, pumps, or pumping installations
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F11/00—Control or safety arrangements
    • F24F11/50—Control or safety arrangements characterised by user interfaces or communication
    • F24F11/61—Control or safety arrangements characterised by user interfaces or communication using timers
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F11/00—Control or safety arrangements
    • F24F11/62—Control or safety arrangements characterised by the type of control or by internal processing, e.g. using fuzzy logic, adaptive control or estimation of values
    • F24F11/63—Electronic processing
    • F24F11/64—Electronic processing using pre-stored data

Definitions

  • This application relates to the field of abnormality detection technology for electrical appliances, and in particular to an abnormality detection method, an abnormality detection device, an electrical appliance, and a non-volatile computer-readable storage medium.
  • This application provides an anomaly detection method, an anomaly detection device, an electrical component, and a non-volatile computer-readable storage medium.
  • the anomaly detection method provided in this application is applied to electrical appliances.
  • the method includes: obtaining the operating power of the electrical appliance whose operating time is within a preset time interval, and determining a basic deviation rate based on the operating power and a preset power threshold; determining a current deviation rate based on the current power of the electrical appliance and the preset power threshold; and performing anomaly detection based on the deviation between the current deviation rate and the basic deviation rate to determine the anomaly detection result.
  • the electrical appliance includes a compressor
  • the preset power threshold of the compressor is determined based on the compressor's sampling frequency, sampling low pressure, and sampling high pressure.
  • obtaining the operating power of an appliance whose operating time is within a preset time interval, and determining a basic deviation rate based on the operating power and a preset power threshold includes: obtaining the current power of the appliance at the current moment when the operating time is within the preset time interval; determining a first deviation rate based on the current power and a preset power threshold; and determining a basic deviation rate based on the first deviation rate and a first historical deviation rate, wherein the first historical deviation rate is determined based on the first deviation rate corresponding to moments prior to the current moment.
  • determining a base deviation rate based on a first deviation rate and a first historical deviation rate of the appliance includes: determining an average deviation rate of the first deviation rate and the first historical deviation rate; and determining the average deviation rate as the base deviation rate.
  • determining the current deviation rate based on the current power of the appliance and a preset power threshold includes: determining the current deviation rate based on the current power of the appliance and the preset power threshold when the running length is greater than the maximum value of a preset time interval.
  • determining the current deviation rate based on the current power of the appliance and a preset power threshold includes: when the running length is greater than the maximum value of a preset time interval, determining a second deviation rate based on the current power of the appliance at the current moment and the preset power threshold; and determining the current deviation rate based on the appliance's second historical deviation rate and the second deviation rate, wherein the second historical deviation rate is determined based on the second deviation rate corresponding to moments prior to the current moment.
  • anomaly detection is performed based on the deviation between the current deviation rate and the baseline deviation rate.
  • the method includes: determining the absolute value of the deviation between the current deviation rate and the baseline deviation rate; and determining that the appliance has malfunctioned if the absolute value is greater than a preset difference threshold.
  • anomaly detection based on the deviation between the current deviation rate and the baseline deviation rate is performed to determine the anomaly detection result, including: determining the remaining service life of the appliance based on the cumulative usage time of the appliance; determining a first target correction coefficient based on the remaining service life and a preset correlation, wherein the first preset correlation includes multiple preset remaining service lives and first preset deviation rate correction coefficients corresponding to each of the multiple preset remaining service lives; correcting the current deviation rate based on the first target correction coefficient to obtain a first corrected deviation rate; determining the absolute value of the deviation between the first corrected deviation rate and the baseline deviation rate; and determining that the appliance has an anomaly if the absolute value is greater than a preset difference threshold.
  • anomaly detection based on the deviation between the current deviation rate and the baseline deviation rate is performed to determine the anomaly detection result, including: determining the remaining service life of the appliance based on the cumulative usage time of the appliance; determining a second target correction coefficient based on the remaining service life and a preset correlation, wherein the preset correlation includes multiple preset remaining service lives and preset deviation rate correction coefficients corresponding to each of the multiple preset remaining service lives; correcting the baseline deviation rate based on the second target correction coefficient to obtain a second corrected deviation rate; determining the absolute value of the deviation between the current deviation rate and the second corrected deviation rate; and determining that the appliance has an anomaly if the absolute value is greater than a preset difference threshold.
  • the malfunction detection method further includes: turning off the appliance and adjusting the maximum value of a preset time interval to obtain a new time interval; turning on the appliance when the appliance's off time reaches a preset off time; using the new time interval as the preset time interval when the appliance's on time reaches a preset on time, and performing the steps of obtaining the operating power of appliances whose running time is within the preset time interval, and determining a basic deviation rate based on the operating power and a preset power threshold; and outputting a prompt message indicating that the appliance has malfunctioned when the absolute value is again determined to be greater than the preset difference threshold.
  • anomaly detection is performed based on the deviation between the current deviation rate and the baseline deviation rate to determine the anomaly detection result, including: calculating the deviation between the current deviation rate and the baseline deviation rate; determining that the working performance of the appliance has abnormally increased when the deviation is negative and the absolute value of the deviation is greater than a preset difference threshold; and determining that the working performance of the appliance has abnormally decreased when the deviation is positive and the deviation is greater than a preset difference threshold.
  • the preset duration range is determined based on the power deviation data and running time of operating appliances of the same type as the appliance in the cloud big data.
  • the anomaly detection device includes a first determining module, a second determining module, and an anomaly detection module.
  • the first determining module is used to acquire the operating power of an appliance whose operating time falls within a preset time interval, and to determine a base deviation rate based on the operating power and a preset power threshold.
  • the second determining module is used to determine the current deviation rate based on the current power of the appliance and the preset power threshold.
  • the anomaly detection module is used to perform anomaly detection based on the deviation between the current deviation rate and the base deviation rate to determine the anomaly detection result.
  • the electrical appliance provided in this application includes a processor, a memory, and a computer program.
  • the computer program is stored in the memory and executed by the processor.
  • the computer program includes instructions for performing the anomaly detection method of any of the above embodiments.
  • the non-volatile computer-readable storage medium provided in this application includes a computer program.
  • the computer program When the computer program is executed by a processor, it causes the processor to perform the anomaly detection method of any of the above embodiments.
  • the anomaly detection method, anomaly detection device, electrical appliance, and computer-readable storage medium provided in this application embodiment can set a preset time interval based on a preset time period corresponding to the electrical appliance's operating performance being within a reasonable range. Then, a base deviation rate is determined based on the electrical appliance's operating power and a preset power threshold within the preset time interval. This base deviation rate can be understood as the electrical appliance's allowable deviation rate. Next, a current deviation rate is determined based on the electrical appliance's current power and the preset power threshold, and anomaly detection is performed based on the deviation between the current deviation rate and the base deviation rate to obtain an anomaly detection result.
  • Figure 1 shows a flowchart of the anomaly detection method provided in an embodiment of this application
  • Figure 2 shows a flowchart of the anomaly detection method provided in an embodiment of this application
  • Figure 3 shows a flowchart of the anomaly detection method provided in an embodiment of this application
  • Figure 4 shows a flowchart of the anomaly detection method provided in an embodiment of this application
  • Figure 5 shows a flowchart of the anomaly detection method provided in an embodiment of this application
  • Figure 6 shows a flowchart of the anomaly detection method provided in an embodiment of this application.
  • Figure 7 shows a schematic diagram of the module of the anomaly detection device provided in an embodiment of this application.
  • Figure 8 shows a schematic diagram of the structure of the electrical appliance provided in an embodiment of this application.
  • Figure 9 shows a schematic diagram of the connection state between the non-volatile computer-readable storage medium and the processor provided in an embodiment of this application.
  • Electrical appliances refer to devices that utilize electrical energy to function normally. With increased usage time, the internal components of electrical appliances wear down, gradually reducing their performance. Once this decline reaches a certain level, the appliance is considered to be malfunctioning, resulting in poor performance.
  • the compressor as the core component of an air conditioner, inevitably experiences wear and tear during long-term operation. This leads to a loss in the performance of both the compressor and the air conditioner, such as abnormally high power consumption, and under extreme conditions, it may easily trigger frequency limiting or even protective shutdown. Simultaneously, the overall energy efficiency of the air conditioner will decrease, reducing the user experience.
  • electrical appliances may malfunction during operation, causing abnormal performance, such as abnormally high or low efficiency. Therefore, timely detection of abnormal electrical appliance performance is a crucial issue that needs to be addressed.
  • Another common method is to compare the collected specific performance parameters with preset specific performance parameters to determine whether an anomaly exists.
  • this method essentially emphasizes static comparison, and the preset specific performance parameters are often difficult to determine accurately for the following reasons:
  • a compressor is a complex electric compression device, and its manufacturing process has certain errors. These errors can only be obtained through on-line testing.
  • the laboratory will only select a few appliances from the same batch for testing to obtain specific performance parameters, but obviously, the specific performance parameters obtained in this way cannot accurately represent the specific performance parameters of each appliance in the batch.
  • specific performance parameters such as power models (physically driven, not big data driven), are generally obtained through laboratory calibration, which inevitably has certain deviations from the actual situation. These deviations include sensor acquisition errors and process calculation errors in the power model input parameters. These errors cannot be obtained statically and often need to be dynamically corrected.
  • this application provides an anomaly detection method.
  • An anomaly detection method provided in this application embodiment is applied to electrical appliances, such as air conditioners.
  • the anomaly detection method provided in this application embodiment includes the following steps:
  • Step 011 Obtain the operating power of appliances whose running time is within a preset time range, and determine the basic deviation rate based on the operating power and the preset power threshold;
  • the preset power threshold is the theoretical operating power of the appliance when its working performance is normal.
  • the preset power threshold can be determined after the appliance is installed in the workplace to ensure that the preset power threshold is more in line with the operating conditions of the appliance.
  • the electrical appliance could be a compressor, whose theoretical operating power can be determined using the AHRI (Advanced Human Relations Index) multi-coefficient model. After installation at the workplace, it can be confirmed that the compressor's operating performance is normal. At this point, the compressor's sampling frequency, low pressure, and high pressure can be collected. Then, based on the sampling frequency, low pressure, high pressure, and the AHRI multi-coefficient model, the compressor's preset power threshold can be determined. In this way, the preset power threshold related to external operating conditions can be obtained.
  • AHRI Advanced Human Relations Index
  • the appliance when the appliance is of other types, its theoretical operating power can be determined by obtaining the performance parameters that affect its theoretical operating power and the corresponding calculation formula.
  • the AHRI ten-coefficient model involves ten coefficients, which can be tested in a laboratory. Several compressors from the same batch can be sampled for testing to obtain the corresponding coefficients for that batch. However, the exact coefficients for each compressor may differ. The tested coefficients only ensure they fall within a reasonable range for each compressor in the batch; they do not represent the accurate coefficients for each individual compressor. Therefore, the theoretical power calculated using the AHRI ten-coefficient model may actually have some error compared to the appliance's true power.
  • the baseline value for judging whether the appliance's performance is abnormal i.e., the baseline deviation rate
  • This baseline deviation rate can be understood as the appliance's allowable deviation rate.
  • a preset time interval can be set based on the operating time during which the appliance's performance is within a reasonable range.
  • the reasonable range of performance can be understood as the range of performance under normal conditions or the range of performance under normal attenuation.
  • a reasonable range of operating performance is the range of operating performance under normal conditions. Manufacturing technology can ensure that the operating performance of electrical appliances remains within the normal range for a certain period after leaving the factory.
  • a preset time interval can be determined based on this time interval, such as 0-100 hours after leaving the factory. The minimum value of this preset time interval is typically 0, while the maximum value can be determined in various ways.
  • the preset duration interval is determined based on the power deviation data and runtime of operating appliances of the same type as the current appliance in the cloud big data.
  • the power deviation data and runtime of each operating appliance can be uploaded to the cloud big data.
  • Cluster analysis can then be performed on the power deviation data of operating appliances of the same type as the current appliance in the cloud big data to obtain a preset duration interval applicable to the current appliance. For example, assuming a power deviation threshold of 3%, a power deviation exceeding 3% indicates that the appliance's performance is within an unreasonable range.
  • the preset duration interval is determined based on the runtime of operating appliances with power deviations less than the power deviation threshold. For example, the maximum value of the preset duration interval can be determined based on the mode or mean of the runtimes of multiple operating appliances with power deviations less than the power deviation threshold, thereby determining the preset duration interval.
  • the preset duration range can be determined by the guide duration provided by the appliance manufacturer.
  • the appliance manufacturer can conduct preliminary experiments on the appliance to determine the longest duration under harsh operating conditions without significant wear, such as 100 hours, while meeting reliability requirements.
  • the appliance manufacturer can then determine the guide duration based on this longest duration, which can be used as the maximum value of the preset duration range.
  • the permissible operating range diagram of an electrical appliance is typically used to represent the operating limits of the appliance under different operating conditions. Therefore, the operating limits of the appliance in its actual operating environment can be determined based on the permissible operating range diagram. That is, typical state points are selected in the permissible operating range diagram, and the appliance is operated in a special mode according to the typical state points to obtain the longest duration during which the appliance can operate for a long time without significant wear. The maximum value of the preset duration interval is then determined based on this longest duration.
  • the horizontal and vertical axes of a compressor's permissible operating range diagram represent condensing and evaporating temperatures.
  • This diagram shows the condensing and evaporating temperatures that different compressor models can achieve when operating in different environments.
  • the condensing and evaporating temperatures that the same compressor model can achieve may differ in different regions; that is, the operating range of the same compressor model may vary in different regions. Therefore, after determining the current compressor model and its operating environment, multiple state points matching the ambient temperature of the compressor's operating environment can be obtained from the compressor's permissible operating range diagram to obtain multiple condensing and evaporating temperatures. Then, experiments are conducted based on these state points.
  • a compressor of the same type as the current compressor is used in the experiment and operates in the environment corresponding to the selected state points. This yields the longest duration that the current compressor can operate without significant wear under these conditions, thus determining the maximum value of the current compressor's preset time interval.
  • a reasonable range for operating performance is the range of operating performance under normal attenuation. Electrical appliances experience normal wear and tear after operation.
  • the preset time interval can be determined based on the operating time corresponding to this normal wear and tear. For example, if it can be determined that the appliance's operating performance is within the normal range for the first 100 hours after factory operation, the preset time interval can be set to 100 hours. Then, if it is found that the operating efficiency has decreased slightly between 150 and 200 hours of operation, the decrease can be attributed to normal wear and tear. Therefore, the preset time interval can be changed to 150-200 hours, allowing the influence of normal wear and tear to be excluded during subsequent anomaly detection, thus facilitating accurate identification of abnormal operating performance of the appliance.
  • the operating power of the appliance within a preset time range is obtained, and the basic deviation rate is determined based on the operating power and a preset power threshold.
  • the deviation rate can be calculated using the following formula:
  • pari is the currently acquired operating power
  • pcmp is the preset power threshold
  • ATTcmp is the deviation rate
  • the basic deviation rate can be determined based on the deviation rate corresponding to the current operating power, or it can be determined based on the deviation rate corresponding to all operating power within a preset time interval. In this way, the basic value required for anomaly detection can be obtained.
  • Step 012 Determine the current deviation rate based on the current power of the appliance and the preset power threshold
  • the current deviation rate can be determined based on the current power of the appliance and the preset power threshold.
  • the current deviation rate can be determined according to the above formula (1). It can be understood that the current deviation rate can be used to characterize the working performance of the appliance at the current moment. At this time, the current deviation rate can be determined only based on the current power, or the deviation rate can be determined based on the current power and the deviation rate can be determined based on the operating power within a preset time period before the current moment. Then, the average value of these deviation rates is obtained, and the average value is used as the current deviation rate.
  • Step 013 Perform anomaly detection based on the deviation between the current deviation rate and the baseline deviation rate to determine the anomaly detection result.
  • anomaly detection can be performed based on the deviation between the current deviation rate and the baseline deviation rate.
  • the deviation represents the difference between the current operating power and the operating power when the current operating power and performance are within a reasonable range.
  • the baseline deviation rate is determined based on a preset time interval. However, this preset time interval may not accurately represent the duration during which the operating performance is within a reasonable range, thus the baseline deviation rate may not accurately reflect the deviation rate when the operating performance is within a reasonable range.
  • the relationship between the current deviation rate and the baseline deviation rate is not fixed. Therefore, a preset difference threshold can be set. This threshold represents the maximum absolute value of the deviation when the appliance's current operating performance is within a reasonable range.
  • the preset difference threshold can be determined based on the preset time range and the type of appliance; for example, a longer preset time range results in a larger preset difference threshold.
  • anomaly detection is also required at this point, based on the difference and a preset difference threshold.
  • the absolute value of the deviation exceeds the preset difference threshold, it can be determined that the appliance's operating performance is abnormal, and the anomaly detection result can be "operating performance abnormal.”
  • the absolute value of the deviation does not exceed the preset difference threshold, it can be determined that the appliance's operating performance is normal, and the anomaly detection result can be "operating performance normal.”
  • a compressor performance warning indicator can be set, initially in the off state. If an abnormality in the appliance's performance is detected, the compressor performance warning indicator can be activated. If the appliance's performance is normal, the warning indicator can remain off. This ensures that technicians or users can promptly detect any abnormalities in the compressor's performance.
  • the anomaly detection method provided in this application can set a preset time interval based on a preset time period corresponding to the reasonable operating performance of an electrical appliance. Then, a base deviation rate is determined based on the operating power of the electrical appliance within the preset time interval and a preset power threshold. This base deviation rate can be understood as the allowable deviation rate of the electrical appliance. Next, the current deviation rate is determined based on the current power of the electrical appliance and the preset power threshold, and anomaly detection is performed based on the deviation between the current deviation rate and the base deviation rate to obtain the anomaly detection result. It can be understood that when the deviation shows a large difference between the current deviation rate and the base deviation rate, it can be confirmed that the operating performance of the electrical appliance is abnormal. In this way, during the operation of the electrical appliance, the operating performance of the electrical appliance can be monitored at all times based on the current deviation rate, thereby obtaining information on abnormal operating performance of the electrical appliance in a timely and accurate manner.
  • this application constructs a self-learning dynamic power threshold model, in which the power threshold used is the base deviation rate, which is determined based on the actual operating conditions of the appliance.
  • the power threshold used is the base deviation rate, which is determined based on the actual operating conditions of the appliance.
  • this application considers the factory differences of individual appliances and the error between the threshold model in the laboratory and real-world scenarios. It uses the operating power corresponding to a preset time interval after the appliance leaves the factory as the learning basis for the threshold model; that is, the base deviation rate is determined based on the operating power corresponding to the preset time interval after the appliance leaves the factory, rather than directly using data obtained in the laboratory as the learning basis for the threshold model.
  • the threshold model determines the threshold based on the actual situation after the appliance leaves the factory, i.e., to determine the preset power threshold based on the operating data after the appliance leaves the factory, thereby determining the base deviation rate, rather than simply using the operating time as the learning basis for the threshold model, i.e., using the operating time as the threshold for anomaly detection. Therefore, the threshold (i.e., the basic deviation rate) used in the threshold model of this application can more accurately detect the abnormality of electrical appliances, so as to solve the performance diagnosis problem during the operation of electrical appliances and enable more accurate acquisition of information on abnormal electrical appliance performance.
  • step 011 obtaining the operating power of appliances whose operating time is within a preset time interval, and determining the basic deviation rate based on the operating power and a preset power threshold, includes the following steps:
  • Step 0111 If the running time of the appliance is within a preset time range, obtain the current power of the appliance at the current moment;
  • Step 0112 Determine the first deviation rate based on the current power and the preset power threshold
  • Step 0113 Determine the basic deviation rate based on the first deviation rate and the first historical deviation rate of the electrical appliance.
  • the first historical deviation rate is determined based on the first deviation rate corresponding to the time before the current time.
  • the first deviation rate is calculated every time the current power is acquired, so that the base deviation rate can be determined based on multiple first deviation rates.
  • the base deviation rate can be determined using the following formula:
  • ATT cmp(base) is the base deviation rate
  • ATT cmp(i) corresponds to the first deviation rate and the first historical deviation rate.
  • all operating power of appliances within a preset time range can be uniformly obtained, and then the first deviation rate corresponding to each operating power can be determined based on the operating power and the preset power threshold. Finally, the base deviation rate can be determined based on the average of multiple first deviation rates.
  • the basic deviation rate can be determined based on the deviation rate corresponding to multiple running times within a preset time range of the appliance's running time. This allows the basic deviation rate to accurately represent the deviation rate when the working performance is within a reasonable range, thereby ensuring the accuracy of subsequent anomaly detection.
  • step 012 determining the current deviation rate based on the current power of the appliance and a preset power threshold includes the following steps:
  • Step 0121 If the running time exceeds the maximum value of the preset time interval, determine the current deviation rate based on the current power of the appliance and the preset power threshold.
  • runtime can be understood as the cumulative working time of the appliance after it leaves the factory.
  • the preset runtime range can be determined based on the working time within the normal performance range; for example, the preset runtime range could be 0-100 hours. Within the preset runtime range, the appliance's performance is considered to be within a reasonable range. Therefore, anomaly detection only needs to be performed when the runtime exceeds the maximum value of the preset runtime range. That is, when the runtime exceeds the maximum value of the preset runtime range, the current deviation rate is determined based on the appliance's current power and the preset power threshold. This eliminates the need for anomaly detection when the runtime is within the preset runtime range, thus reducing the computational load.
  • runtime can also be understood as the cumulative working time of the appliance each day, allowing the base deviation rate to be affected by the normal degradation of performance.
  • the preset runtime range corresponds to the range of daily working time, for example, 0-2 hours. It is understood that even when the runtime is within the preset runtime range, performance may still be abnormal, thus requiring anomaly detection. This can be achieved by acquiring the current power and performing anomaly detection based on the base deviation rate corresponding to the preset runtime range of the previous day. If the performance is normal when the runtime is within the preset runtime range, the base deviation rate can be re-determined based on the operating efficiency within the preset runtime range.
  • the current power is acquired again, and anomaly detection is performed based on the current power and the re-determined base deviation rate.
  • the base deviation rate can be affected by the normal degradation of performance, preventing normal degradation from being treated as an anomaly and triggering a warning during anomaly detection. This allows for accurate identification of abnormal performance changes in the appliance other than normal degradation.
  • step 0121 When the running time exceeds the maximum value of a preset time interval, the current deviation rate is determined based on the current power of the appliance and a preset power threshold, including the following steps:
  • Step 01211 If the running time exceeds the maximum value of the preset time interval, determine the second deviation rate based on the current power of the appliance at the current moment and the preset power threshold;
  • Step 01212 Determine the current deviation rate based on the second historical deviation rate and the second deviation rate of the electrical appliance.
  • the second historical deviation rate is determined based on the second deviation rate corresponding to the time before the current time.
  • the second deviation rate is calculated every time the current power is acquired, so that the current deviation rate can be determined based on multiple second deviation rates.
  • the current power of the appliance is obtained, and a second deviation rate is determined based on the current power and a preset power threshold. Then, the second deviation rates determined before the current time are obtained. For example, if the preset time interval is 0-90 hours and the current time is 120 hours, all second deviation rates determined within hours 91-119 need to be obtained, and these are defined as the second historical deviation rates. Finally, the average deviation rate is calculated by combining the current second deviation rate and the second historical deviation rate, and this average deviation rate is determined as the current deviation rate. For example, the current deviation rate can be determined using the following formula:
  • ATT cmp( stat ) is the current deviation rate
  • ATT cmp(i) corresponds to the second deviation rate and the second historical deviation rate.
  • the current deviation rate can be determined based on the second deviation rate of multiple corresponding runtimes that are greater than the maximum value of the preset runtime interval. This allows the current deviation rate to represent the working performance of the appliance during this period. Subsequent anomaly detection can accurately detect whether the working performance during this period is normal based on the current deviation rate, thereby ensuring the accuracy of anomaly detection.
  • the current deviation rate can also be determined directly based on the second deviation rate, so that the current deviation rate represents the working performance of the electrical appliance at the current moment, thereby ensuring the accuracy of anomaly detection while reducing the amount of calculation.
  • step 013: performing anomaly detection based on the deviation between the current deviation rate and the baseline deviation rate to determine the anomaly detection result includes the following steps:
  • Step 0131 Determine the absolute value of the deviation between the current deviation rate and the baseline deviation rate
  • Step 0132 If the absolute value is greater than the preset difference threshold, determine that the appliance is malfunctioning.
  • the relationship between the current deviation rate and the baseline deviation rate is not fixed, while the preset difference threshold is usually set to a positive number. Therefore, during anomaly detection, the absolute value of the deviation between the current deviation rate and the baseline deviation rate can be directly compared with the preset difference threshold. First, the absolute value of the deviation between the current deviation rate and the baseline deviation rate is calculated. If the absolute value is greater than the preset difference threshold, it can be confirmed that the difference between the two is large, and an abnormality in the appliance can be determined. If the absolute value is less than the preset difference threshold, it can be confirmed that the difference between the two is small, and the appliance can be determined to be working normally. In this way, anomalies can be quickly identified based on the absolute value of the deviation.
  • step 013: performing anomaly detection based on the deviation between the current deviation rate and the baseline deviation rate to determine the anomaly detection result includes the following steps:
  • Step 0133 Calculate the deviation between the current deviation rate and the baseline deviation rate
  • Step 0134 If the deviation is negative and the absolute value of the deviation is greater than the preset difference threshold, it is determined that the working performance of the electrical appliance has increased abnormally.
  • Step 0135 If the deviation is positive and greater than the preset difference threshold, determine that the working performance of the appliance is abnormally degraded.
  • the magnitude of the deviation can be used to determine whether the abnormal increase or decrease in performance is genuine.
  • the deviation rate calculation formula can be applied. As can be seen from Formula 1 above, the higher the current power, the smaller the deviation rate.
  • the first step is to calculate the deviation between the current deviation rate and the baseline deviation rate, which is obtained by subtracting the baseline deviation rate from the current deviation rate. Then, the sign of the deviation can be determined to indicate whether the performance is increasing or decreasing.
  • the deviation When the deviation is negative, it can be confirmed that the current deviation rate is less than the base deviation rate, and the current power is greater than the operating power corresponding to the base deviation rate. This indicates that the appliance's operating power has increased.
  • it is determined whether the absolute value of the deviation is greater than a preset difference threshold. If it is, it means the increase in the appliance's operating power is significant, confirming an abnormal increase in the appliance's performance. If it is less, it means the increase in the appliance's operating power is small and still within the allowable range, confirming that the appliance's performance is normal.
  • the deviation When the deviation is positive, it can be confirmed that the current deviation rate is greater than the base deviation rate, meaning the current power is less than the operating power corresponding to the base deviation rate. This indicates that the appliance's operating power has decreased.
  • the absolute value of the deviation is greater than a preset difference threshold. If it is, it indicates a significant decrease in the appliance's operating power, confirming an abnormal degradation in the appliance's performance. If it is less, it indicates a smaller decrease in the appliance's operating power, still within the acceptable range, confirming that the appliance's performance is normal.
  • the sign of the deviation and a preset difference threshold can be used to accurately determine whether the appliance's performance is abnormally increasing or decreasing.
  • the appliance can then emit corresponding abnormal signals based on the detection results.
  • the appliance may include an alarm light to indicate an abnormal detection result. If abnormal performance is detected as increasing, the alarm light may emit an orange light; if abnormal performance is detected as decreasing, the alarm light may emit a red light; if normal performance is detected, the alarm light may turn off or emit a green light. This allows users or technicians to directly determine the appliance's abnormal condition based on the abnormal signals, facilitating targeted repairs.
  • step 013: performing anomaly detection based on the deviation between the current deviation rate and the baseline deviation rate to determine the anomaly detection result includes the following steps:
  • Step 01311 Determine the remaining service life of the appliance based on its cumulative usage time
  • Step 01312 Determine the first target correction coefficient based on the remaining service life and the preset correlation.
  • the first preset correlation includes multiple preset remaining service lives and the first preset deviation rate correction coefficient corresponding to each of the multiple preset remaining service lives.
  • Step 01313 Correct the current deviation rate according to the first target correction coefficient to obtain the first corrected deviation rate
  • Step 01314 Determine the absolute value of the deviation between the first correction deviation rate and the basic deviation rate
  • Step 01315 If the absolute value is greater than the preset difference threshold, determine that the appliance is malfunctioning.
  • the remaining service life of an appliance is expressed by the remaining usage time. For example, if the total usage time of an appliance from the time it leaves the factory until it becomes unusable is T1, and the cumulative usage time after leaving the factory is T2, then the remaining usage time of the appliance is T1-T2, that is, the remaining service life is T1-T2.
  • the basic deviation rate is determined by "all first deviation rates for time periods of 0-79 hours". Based on this information, the appliance can be understood as a new device at this time. Therefore, there is almost no aging phenomenon in the hardware of the appliance, and the influence of aging phenomenon on the deviation rate can be ignored. That is, the basic deviation rate is obtained under the condition of no hardware aging. Therefore, after correcting the deviation rate calculated when there is hardware aging phenomenon using the remaining service life, the corrected deviation rate should be less than the original deviation rate. Therefore, the value range of the correction coefficient in the first preset correlation relationship is (0,1).
  • the remaining service life of the appliance is calculated. Then, by querying the first preset correlation relationship through the remaining service life of the appliance, the first preset deviation rate correction coefficient corresponding to the remaining service life of the appliance is obtained, that is, the first target correction coefficient is obtained. Then, the current deviation rate is corrected using the first target correction coefficient to obtain the first corrected deviation rate, which is less than the current deviation rate before correction.
  • the absolute value of the deviation between the first correction deviation rate and the basic deviation rate After obtaining the first correction deviation rate, calculate the absolute value of the deviation between the first correction deviation rate and the basic deviation rate. If the absolute value is greater than the preset difference threshold, determine that the appliance is abnormal; otherwise, determine that the appliance is normal.
  • the current deviation rate is corrected by taking into account the impact of hardware aging on the deviation rate.
  • the corrected current deviation rate is compared with the baseline deviation rate, the accuracy of judging whether the appliance is abnormal can be improved.
  • step 013: performing anomaly detection based on the deviation between the current deviation rate and the baseline deviation rate to determine the anomaly detection result includes the following steps:
  • Step 01321 Determine the remaining service life of the appliance based on its cumulative usage time
  • the second target correction coefficient is determined based on the remaining useful life and the preset correlation.
  • the preset correlation includes multiple preset remaining useful lives and the preset deviation rate correction coefficients corresponding to each of the multiple preset remaining useful lives.
  • Step 01322 Correct the basic deviation rate according to the second target correction coefficient to obtain the second corrected deviation rate
  • Step 01323 Determine the absolute value of the deviation between the current deviation rate and the second corrected deviation rate
  • Step 01324 If the absolute value is greater than the preset difference threshold, determine that the appliance is malfunctioning.
  • the basic deviation rate is determined by "all first deviation rates for time periods of 0-79 hours". Based on this information, the appliance can be understood as a new device at this time. Therefore, there is almost no aging phenomenon in the hardware of the appliance, and the influence of aging phenomenon on the deviation rate can be ignored. That is, the basic deviation rate is obtained under the condition of no hardware aging. Therefore, after correcting the deviation rate calculated when there is no aging phenomenon in the hardware using the remaining service life, the corrected deviation rate should be less than the original deviation rate. Therefore, the value of the correction coefficient in the second preset correlation relationship is greater than 1.
  • the remaining service life of the appliance is calculated. Then, by querying the above-mentioned second preset correlation through the remaining service life of the appliance, the second preset deviation rate correction coefficient corresponding to the remaining service life of the appliance is obtained, that is, the second target correction coefficient is obtained. Then, the basic deviation rate is corrected by the second target correction coefficient to obtain the second corrected deviation rate, which is greater than the basic deviation rate before correction.
  • the baseline deviation rate is corrected by taking into account the impact of hardware aging on the deviation rate. Comparing the current deviation rate with the corrected baseline deviation rate can improve the accuracy of determining whether an appliance is malfunctioning.
  • the malfunction detection method further includes the following steps:
  • step 011 is executed to obtain the operating power of the appliance whose operating time is within the preset time interval, and to determine the basic deviation rate based on the operating power and the preset power threshold.
  • an abnormality warning message is displayed for the output electrical appliance.
  • the abnormality of the appliance may be temporary and may disappear after being turned off and on again, in order to avoid misjudgment, if any of the absolute values of the deviation between the current deviation rate and the base deviation rate, the absolute values of the deviation between the first corrected deviation rate and the base deviation rate, and the absolute values of the deviation between the current deviation rate and the second corrected deviation rate are greater than a preset difference threshold, the appliance will be turned off first, and the shutdown duration will be started.
  • the maximum value of the preset duration interval will be adjusted to obtain a new time interval.
  • the maximum value of the new time interval is less than the maximum value of the time interval before the adjustment, or the maximum value of the new time interval is less than the maximum value of the time interval before the adjustment.
  • step 011 is executed. This allows the absolute values of the deviations between the current deviation rate and the base deviation rate, the first corrected deviation rate and the base deviation rate, and the current deviation rate and the second corrected deviation rate to be obtained again. If any of these three absolute values is greater than a preset difference threshold, it is further determined that the appliance is abnormal, and an error message indicating that the appliance is abnormal is output (e.g., a warning sign indicating abnormal appliance performance). This avoids misjudging that the appliance is abnormal and improves the accuracy of judging whether the appliance is abnormal.
  • the anomaly detection device 10 may include a first determining module 11, a second determining module 12, and an anomaly detection module 13.
  • the first determining module 11 is used to acquire the operating power of an appliance whose operating time is within a preset time interval, and determine a basic deviation rate based on the operating power and a preset power threshold;
  • the second determining module 12 is used to determine the current deviation rate based on the current power of the appliance and the preset power threshold;
  • the anomaly detection module 13 is used to perform anomaly detection based on the deviation between the current deviation rate and the basic deviation rate to determine the anomaly detection result.
  • the first determining module 11 is specifically used to obtain the current power of the appliance at the current moment when the running time of the appliance is within a preset time interval; determine a first deviation rate based on the current power and a preset power threshold; and determine a basic deviation rate based on the first deviation rate and the appliance's first historical deviation rate, wherein the first historical deviation rate is determined based on the first deviation rate corresponding to moments before the current moment.
  • the first determining module 11 is specifically used to determine the average deviation rate of the first deviation rate and the first historical deviation rate; and to determine the average deviation rate as the base deviation rate.
  • the second determining module 12 is specifically used to determine the current deviation rate based on the current power of the appliance and the preset power threshold when the running length exceeds the maximum value of the preset time interval.
  • the second determining module 12 is further configured to determine a second deviation rate based on the current power of the appliance at the current moment and a preset power threshold when the running length is greater than the maximum value of a preset time interval; and to determine the current deviation rate based on the appliance's second historical deviation rate and second deviation rate, wherein the second historical deviation rate is determined based on the second deviation rate corresponding to moments before the current moment.
  • the construction module 13 is specifically used to determine the absolute value of the deviation between the current deviation rate and the basic deviation rate; if the absolute value is greater than the preset difference threshold, it is determined that the appliance has malfunctioned.
  • the construction module 13 is further configured to: determine the remaining service life of the appliance based on its cumulative usage time; determine a first target correction coefficient based on the remaining service life and a preset correlation, wherein the first preset correlation includes multiple preset remaining service lives and first preset deviation rate correction coefficients corresponding to each of the multiple preset remaining service lives; correct the current deviation rate based on the first target correction coefficient to obtain a first corrected deviation rate; determine the absolute value of the deviation between the first corrected deviation rate and the base deviation rate; and determine that the appliance has malfunctioned if the absolute value is greater than a preset difference threshold.
  • the construction module 13 is further used to determine the remaining service life of the appliance based on the cumulative usage time of the appliance; determine a second target correction coefficient based on the remaining service life and a preset correlation relationship, wherein the preset correlation relationship includes multiple preset remaining service lives and preset deviation rate correction coefficients corresponding to each of the multiple preset remaining service lives; correct the base deviation rate based on the second target correction coefficient to obtain a second corrected deviation rate; determine the absolute value of the deviation between the current deviation rate and the second corrected deviation rate; and determine that the appliance has malfunctioned if the absolute value is greater than a preset difference threshold.
  • the anomaly detection device 10 further includes: a judgment unit, which is used to determine that an electrical appliance has malfunctioned when the absolute value is greater than a preset difference threshold, then turn off the electrical appliance and adjust the maximum value of a preset time interval to obtain a new time interval; turn on the electrical appliance when the appliance's off time reaches a preset off time; take the new time interval as the preset time interval when the appliance's on time reaches a preset on time, and perform the steps of obtaining the operating power of the electrical appliance whose running time is within the preset time interval, and determining the basic deviation rate based on the operating power and a preset power threshold; and output a prompt message indicating that an electrical appliance has malfunctioned when the absolute value is again greater than the preset difference threshold.
  • a judgment unit which is used to determine that an electrical appliance has malfunctioned when the absolute value is greater than a preset difference threshold, then turn off the electrical appliance and adjust the maximum value of a preset time interval to obtain a new time interval; turn on the
  • the construction module 13 is also used to calculate the difference between the current deviation rate and the basic deviation rate; when the deviation is negative and the absolute value of the deviation is greater than the preset difference threshold, it is determined that the working performance of the appliance has increased abnormally; when the deviation is positive and the deviation is greater than the preset difference threshold, it is determined that the working performance of the appliance has decreased abnormally.
  • the anomaly detection device 10 has been described above from the perspective of functional modules, with reference to the accompanying drawings.
  • These functional modules can be implemented in hardware, in software instructions, or in a combination of hardware and software modules.
  • the steps of the method embodiments in this application can be completed by integrated logic circuits in the processor's hardware and/or by software instructions.
  • the steps of the method disclosed in this application can be directly manifested as execution by a hardware encoding processor, or by a combination of hardware and software modules in the encoding processor.
  • the software module can reside in a mature storage medium in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. This storage medium is located in memory, and the processor reads information from the memory and, in conjunction with its hardware, completes the steps in the above method embodiments.
  • the electrical appliance 100 provided in this embodiment includes a processor 20, a memory 30, and a computer program.
  • the computer program is stored in the memory 30 and executed by the processor 20.
  • the computer program includes instructions for performing the anomaly detection method of any of the above embodiments.
  • This application embodiment also provides a computer-readable storage medium 300, on which a computer program 310 is stored.
  • the computer program 310 is executed by the processor 320, it implements the steps of the anomaly detection method of any of the above embodiments. For the sake of brevity, it will not be described again here.
  • references to terms such as “some embodiments,” “in one example,” “exemplarily,” etc. indicate that a specific feature, structure, material, or characteristic described in connection with an embodiment or example is included in at least one embodiment or example of this application.
  • the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example.
  • the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.
  • those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

Landscapes

  • Engineering & Computer Science (AREA)
  • Mechanical Engineering (AREA)
  • General Engineering & Computer Science (AREA)
  • Chemical & Material Sciences (AREA)
  • Combustion & Propulsion (AREA)
  • Signal Processing (AREA)
  • Computer Hardware Design (AREA)
  • Mathematical Physics (AREA)
  • Fuzzy Systems (AREA)
  • Physics & Mathematics (AREA)
  • Human Computer Interaction (AREA)
  • Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Testing And Monitoring For Control Systems (AREA)

Abstract

本申请公开了一种异常检测方法、异常检测装置、电器及非易失性计算机可读存储介质，该方法包括获取电器的运行时长位于预设时长区间内的电器的运行功率，并根据运行功率和预设功率阈值确定基础偏差率；根据电器的当前功率和预设功率阈值确定当前偏差率；基于当前偏差率和基础偏差率的偏差进行异常检测，以确定异常检测结果。可根据电器的工作性能位于合理范围内对应的预设时长设置预设时长区间，使得基础偏差率可以理解为电器的允许偏差率。可以理解，在偏差显示当前偏差率和基础偏差率之间的差距较大的情况下，可确认此时电器的工作性能发生异常。如此便可在电器的运行过程中，时刻监视电器的工作性能，从而可及时获取到电器工作性能异常的信息。

Description

异常检测方法、异常检测装置、电器和可读存储介质
本申请要求于2024年05月14日提交国家知识产权局、申请号为2024105977505、申请名称为“异常检测方法、异常检测装置、电器和可读存储介质”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
本申请涉及电器的异常检测技术领域，尤其涉及一种异常检测方法、异常检测装置、电器和非易失性计算机可读存储介质。
背景技术
随着使用时长的增加，电器的工作性能会逐渐衰减，衰减到一定程度后，可认为此时工作性能异常，导致电器的工作效果较差。例如压缩机作为空调器的核心部件，其在长期运行过程中不可避免出现一定磨损，这会给压缩机及空调器的工作性能带来一定损失，比如功率异常偏大，极端工况下很容易进入限频，甚至保护停机，同时空调器整机运行能效也会出现一定的下降，降低用户体验。因此，如何及时获取到电器工作性能异常的信息成为亟需解决的问题。
发明内容
本申请提供一种异常检测方法、异常检测装置、电器和非易失性计算机可读存储介质。
本申请实施例提供的异常检测方法应用于电器，该异常检测方法方法包括：获取电器的运行时长位于预设时长区间内的电器的运行功率，并根据运行功率和预设功率阈值确定基础偏差率；根据电器的当前功率和预设功率阈值确定当前偏差率；基于当前偏差率和基础偏差率的偏差进行异常检测，以确定异常检测结果。
在一些实施例中，电器包括压缩机，压缩机的预设功率阈值根据压缩机的采集频率、采集低压和采集高压确定。
在一些实施例中，获取电器的运行时长位于预设时长区间内的电器的运行功率，并根据运行功率和预设功率阈值确定基础偏差率，包括：在电器的运行时长位于预设时长区间的情况下，获取电器在当前时刻的当前功率；根据当前功率和预设功率阈值确定第一偏差率；根据第一偏差率和第一历史偏差率确定基础偏差率，第一历史偏差率根据当前时刻以前的时刻对应的第一偏差率确定。
在一些实施例中，根据第一偏差率和电器的第一历史偏差率确定基础偏差率，包括：确定第一偏差率和第一历史偏差率的平均偏差率；将平均偏差率确定为基础偏差率。
在一些实施例中，根据电器的当前功率和预设功率阈值确定当前偏差率，包括：在运行时长大于预设时长区间的最大值的情况下，根据电器的当前功率和预设功率阈值确定当前偏差率。
在一些实施例中，根据电器的当前功率和预设功率阈值确定当前偏差率，包括：在运行时长大于预设时长区间的最大值的情况下，根据电器在当前时刻的当前功率和预设功率阈值确定第二偏差率；根据电器的第二历史偏差率和第二偏差率确定当前偏差率，第二历史偏差率根据当前时刻以前的时刻对应的第二偏差率确定。
在一些实施例中，基于当前偏差率和基础偏差率的偏差进行异常检测，以确定异常检测结果包括：确定当前偏差率和基础偏差率的偏差的绝对值；在绝对值大于预设差值阈值的情况下，确定电器发生异常。
在一些实施例中，基于当前偏差率和基础偏差率的偏差进行异常检测，以确定异常检测结果包括：根据电器的累计使用时长确定电器的剩余使用寿命；根据剩余使用寿命和预设关联关系，确定第一目标修正系数，第一预设关联关系包括多个预设剩余使用寿命和多个预设剩余使用寿命各自对应的第一预设偏差率修正系数；根据第一目标修正系数对当前偏差率进行修正，得到第一修正偏差率；确定第一修正偏差率和基础偏差率的偏差的绝对值；在绝对值大于预设差值阈值的情况下，确定电器发生异常。
在一些实施例中，基于当前偏差率和基础偏差率的偏差进行异常检测，以确定异常检测结果包括：根据电器的累计使用时长确定电器的剩余使用寿命；根据剩余使用寿命和预设关联关系，确定第二目标修正系数，预设关联关系包括多个预设剩余使用寿命和多个预设剩余使用寿命各自对应的预设偏差率修正系数；根据第二目标修正系数对基础偏差率进行修正，得到第二修正偏差率；确定当前偏差率和第二修正偏差率的偏差的绝对值；在绝对值大于预设差值阈值的情况下，确定电器发生异常。
在一些实施例中，在绝对值大于预设差值阈值的情况下，确定电器发生异常之后，该异常检测方法还包括：关闭电器，并调整预设时长区间的最大值，得到新的时间区间；在电器的关闭时长达到预设关闭时长的情况下，开启电器；在电器的开启时长达到预设开启时长的情况下，将新的时间区间作为预设时长区间，并执行获取电器的运行时长位于预设时长区间内的电器的运行功率，并根据运行功率和预设功率阈值确定基础偏差率的步骤；在再次确定绝对值大于预设差值阈值的情况下，输出电器发生异常的提示信息。
在一些实施例中，基于当前偏差率和基础偏差率的偏差进行异常检测，以确定异常检测结果包括：计算当前偏差率和基础偏差率的偏差；在偏差为负数，且偏差的绝对值大于预设差值阈值的情况下，确定电器的工作性能异常增加；在偏差为正数，且偏差大于预设差值阈值的情况下，确定电器的工作性能异常衰减。
在一些实施例中，预设时长区间根据云端大数据中与电器的类型相同的已运行电器的功率偏差数据及运行时长确定。
本申请实施例提供的异常检测装置包括第一确定模块、第二确定模块及异常检测模块。第一确定模块用于获取电器的运行时长位于预设时长区间内的电器的运行功率，并根据运行功率和预设功率阈值确定基础偏差率。第二确定模块用于根据电器的当前功率和预设功率阈值确定当前偏差率。异常检测模块用于基于当前偏差率和基础偏差率的偏差进行异常检测，以确定异常检测结果。
本申请实施例提供的电器包括处理器、存储器及计算机程序，其中，计算机程序被存储在存储器中，并且被处理器执行，计算机程序包括用于执行上述任一实施例的异常检测方法的指令。
本申请实施例提供的非易失性计算机可读存储介质包括计算机程序，计算机程序被处理器执行时，使得处理器执行上述任一实施例的异常检测方法。
本申请实施例提供的异常检测方法、异常检测装置、电器和计算机可读存储介质，可根据电器的工作性能位于合理范围内对应的预设时长设置预设时长区间，然后根据预设时长区间内的电器的运行功率和预设功率阈值确定基础偏差率，该基础偏差率可以理解为电器的允许偏差率。接着，可根据电器的当前功率和预设功率阈值确定当前偏差率，并基于当前偏差率和基础偏差率的偏差进行异常检测，得到异常检测结果。可以理解，在偏差显示当前偏差率和基础偏差率之间的差距较大的情况下，可确认此时电器的工作性能发生异常。如此便可在电器的运行过程中，可根据当前偏差率来时刻监视电器的工作性能，从而可及时且准确地获取到电器工作性能异常的信息。
本申请的实施例的附加方面和优点将在下面的描述中部分给出，部分将从下面的描述中变得明显，或通过本申请的实施例的实践了解到。
附图说明
本申请的上述和/或附加的方面和优点从结合下面附图对实施例的描述中将变得明显和容易理解，其中：
图1示出了本申请实施例提供的异常检测方法的流程示意图；
图2示出了本申请实施例提供的异常检测方法的流程示意图；
图3示出了本申请实施例提供的异常检测方法的流程示意图；
图4示出了本申请实施例提供的异常检测方法的流程示意图；
图5示出了本申请实施例提供的异常检测方法的流程示意图；
图6示出了本申请实施例提供的异常检测方法的流程示意图；
图7示出了本申请实施例提供的异常检测装置的模块示意图；
图8示出了本申请实施例提供的电器的结构示意图；
图9示出了本申请实施例提供的非易失性计算机可读存储介质和处理器的连接状态示意图。
具体实施方式
下面详细描述本申请的实施例，实施例的示例在附图中示出，其中，相同或类似的标号自始至终表示相同或类似的元件或具有相同或类似功能的元件。下面通过参考附图描述的实施例是示例性的，仅用于解释本申请的实施例，而不能理解为对本申请的实施例的限制。
电器指的是利用电能来进行正常工作的器件。随着使用时长的增加，电器内部的元件会发生损耗，随着使用时长的增加，电器的工作性能会逐渐衰减，衰减到一定程度后，可认为此时工作性能异常，导致电器的工作效果较差。例如压缩机作为空调器的核心部件，其在长期运行过程中不可避免出现一定磨损，这会给压缩机及空调器的工作性能带来一定损失，比如功率异常偏大，极端工况下很容易进入限频，甚至保护停机，同时空调器整机运行能效也会出现一定的下降，降低用户体验。除此之外，电器工作过程中还可能会发生故障，导致电器工作性能异常，例如工作效率异常增高或异常衰减。因此，如何及时获取到电器工作性能异常的信息成为亟需解决的问题。
针对上述问题，一种常用的方法是将电器运行时长作为评价电器性能磨损的指标之一，但该方法忽略了外部实际运行工况对磨损的影响，实际上运行工况越恶劣，在相同运行时长内磨损的程度越大。
另一种常用的方法是比较采集到的特定性能与预设的特定性能参数，从而判断出异常与否。但该方法本质上还是强调静态对比法，预设的特定性能参数往往很难准确确定，原因如下：首先，电器的生产工艺通常存在一定的误差，例如压缩机是个复杂的电动压缩式器件，其生产工艺存在一定误差，这部分误差是需要通过下线实测验证才可获取的。实验室只会从同一批电器中抽取若干个电器来进行测试，得到特定性能参数，但是显然这样得到的特定性能参数无法准确地代表这批电器中每一个电器的特定性能参数。其次，特定性能参数比如功率模型(物理驱动型，非大数据驱动型)一般是通过实验室标定得到，这与实际情况必然存在一定偏差，这部分偏差包括功率模型入参的传感器采集误差和过程计算误差，这部分误差无法静态获得，往往是需要动态修正的。
为解决上述技术问题，本申请实施例提供一种异常检测方法。
下面将对本申请的异常检测方法进行详细阐述：
请参阅图1，本申请实施例提供的一种异常检测方法，应用于电器，例如空调。本申请实施例提供的异常检测方法包括以下步骤：
步骤011：获取电器的运行时长位于预设时长区间内的电器的运行功率，并根据运行功率和预设功率阈值确定基础偏差率；
具体地，预设功率阈值为电器的工作性能正常时的理论运行功率，预设功率阈值可在电器安装在工作场所后确定，以确保预设功率阈值较符合电器的运行情况。
例如电器可为压缩机，压缩机的理论运行功率可根据AHRI十系数模型确定。在安装在工作场所后，可确认此时压缩机的工作性能正常，此时可采集压缩机的采集频率、采集低压和采集高压，然后根据采集频率、采集低压、采集高压和AHRI十系数模型来确定压缩机的预设功率阈值。如此便可获取与外部的运行工况相关的预设功率阈值。
类似的，在电器为其他类型的电器的情况下，可通过获取影响该电器的理论运行功率的性能参数及对应的计算公式来确定该电器的理论运行功率。
但是，电器通常是大批量生产，而在确定预设功率阈值时，往往只在其中选取少部分电器作为样本来确定预设功率阈值。同时，即使是同一生产批次的电器，不同电器之间也可能会存在一定的差异，导致电器在工作性能正常时的运行功率不一定为理论运行功率。例如，AHRI十系数模型涉及十个系数，该十个系数可在实验室中测试得到，在同一批生产出来的压缩机中可抽取其中几台压缩机进行测试，以得到该批压缩机对应的系数。然而，每个压缩机对应的准确的系数可能不相同，测试得到的系数只能确保位于该批中各个压缩机对应的系数的合理范围内，测试得到的系数并不能够代表各个压缩机的准确的系数，因此利用AHRI十系数模型计算得到的理论功率实际上与电器的真实功率之间可能存在一定的误差。
因此，需要根据电器真实运行时，且工作性能正常的情况下对应的实际运行功率和预设功率阈值来确定判断电器的工作性能是否异常的基础值，即基础偏差率，该基础偏差率可以理解为电器的允许偏差率。此时可根据电器的工作性能处于合理范围内的工作时长来设置预设时长区间。其中工作性能的合理范围可理解为处于正常状态的工作性能的范围，或是处于正常衰减状态的工作性能的范围。
在一些实施例中，工作性能的合理范围为处于正常状态的工作性能的范围。生产技术可确保电器在出厂后运行的一段时长内工作性能一定处于正常范围内，可根据该时长确定预设时长区间，如预设时长区间为出厂后0-100小时。此时预设时长区间的最小值通常为0，而预设预设时长区间的最大值的确定方式有多种。
在一个实施例中，预设时长区间根据云端大数据中与电器的类型相同的已运行电器的功率偏差数据及运行时长确定。每个已运行电器的功率偏差数据和运行时长都可上传到云端大数据中。此时可对云端大数据中，与当前电器的类型相同的已运行电器的功率偏差数据进行聚类分析，以得到适用于当前电器的预设时长区间。如，假设功率偏差阈值为3％，一旦功率偏差超过3％即代表该电器的工作性能处于不合理范围。那么，在确定当前电器的类型后，可从云端大数据中获取与当前电器的类型相同的已运行电器的功率偏差数据，以确定功率偏差小于或等于功率偏差阈值的已运行电器，并获取他们的运行时长。最后根据功率偏差小于功率偏差阈值的已运行电器的运行时长确定预设时长区间，例如可根据多个功率偏差小于功率偏差阈值的已运行电器的运行时长的众数或平均数确定预设时长区间的最大值，从而确定预设时长区间。
在另一个实施例中，预设时长区间可由电器厂家给出的指导时长确定。电器厂家可对电器进行预先的实验，确定在电器按照满足可靠性要求下开展恶劣工况长期运行仍无明显磨损的最长时长，比如100h，然后电器厂家便可根据该最长时长确定指导时长，而指导时长可作为预设时长区间的最大值。
在又一个实施例中，电器允许运行范围图通常用于表示该电器在在不同工况下的运行界限。因此，可以根据电器允许运行范围图确定电器在其真实运行环境下的运行界限，即在电器允许运行范围图选取典型状态点，并使得电器根据典型状态点进行特殊模式运行，得到电器长期运行仍无明显磨损的最长时长，从而根据该最长时长确定预设时长区间的最大值。
例如，压缩机的允许运行范围图的横纵坐标为冷凝温度和蒸发温度，压缩机允许运行范围图中整理了不同型号的压缩机在不同环境中运行时能够达到的冷凝温度和蒸发温度。不同地区的同种型号的压缩机运行时能够达到的冷凝温度和蒸发温度可能不同，即同种型号的压缩机在不同地区运行时的运行范围可能不同。因此，在确定当前压缩机的型号以及当前压缩机的运行环境后，可在压缩机允许运行范围图中获取与压缩机的运行环境的环境温度匹配的多个状态点，以获取多个冷凝温度和蒸发温度。然后根据多个状态点进行实验，令用于实验的与当前压缩机的类型相同的压缩机在选取的状态点对应的环境下运行，从而得到当前压缩机在该环境下长期运行仍无明显磨损的最长时长，从而可确定当前压缩机的预设时长区间的最大值。
在一些实施例中，工作性能的合理范围为处于正常衰减状态的工作性能的范围。电器在运行后会正常损耗，此时可根据正常损耗时对应的运行时长来确定预设时长区间。如，可确定出厂工作100小时以内，电器的工作性能一定处于正常范围内，此时可设定预设时长区间为100小时。然后在工作150-200小时的时候，发现这段时间内虽然工作效率有所下降，但是下降的幅度并不大，此时可认为这段时间内电器的工作效率下降的原因是正常损耗。那么，就可以将预设时长区间改为150-200小时，使得后续在进行异常检测时可排除正常损耗的影响，以便于准确地识别出电器的工作性能异常的情况。
然后获取电器的运行时长位于预设时长区间内的运行功率，并根据运行功率和预设功率阈值确定基础偏差率。例如可根据下述公式计算偏差率：
其中pari为当前获取的运行功率，pcmp为预设功率阈值，ATTcmp为偏差率。
此时可根据当前的运行功率对应的偏差率来确定基础偏差率，或者可根据预设时长区间内的所有运行功率对应的偏差率来确定基础偏差率，如此便可得到异常检测时需要的基础值。
步骤012：根据电器的当前功率和预设功率阈值确定当前偏差率；
具体地，可根据电器的当前功率和预设功率阈值确定当前偏差率，如可根据上述公式(1)确定当前偏差率，可以理解，当前偏差率可用于表征电器在当前时刻的工作性能。此时可只根据当前功率确定当前偏差率，或者可根据当前功率确定偏差率，并根据当前时刻前一段预设时长内的运行功率确定偏差率，然后求得这些偏差率的平均值，并将平均值作为当前偏差率。
步骤013：基于当前偏差率和基础偏差率的偏差进行异常检测，以确定异常检测结果。
具体地，可基于当前偏差率和基础偏差率的偏差进行异常检测，偏差可表征当前运行功率和工作性能位于合理范围内时对应的运行功率的差值。基础偏差率是根据预设时长区间确定的，预设时长区间可能无法非常准确地表达工作性能位于合理范围内的时长，使得基础偏差率可能无法非常准确地表达工作性能位于合理范围内时的偏差率。同时，当前偏差率和基础偏差率的大小关系并不固定。因此此时还可设置预设差值阈值，预设差值阈值为在电器的当前工作性能位于合理范围内的情况下，偏差的绝对值的最大值，一旦两者之间的偏差的绝对值超过预设差值阈值，则可确定电器的工作性能异常。其中，预设差值阈值可根据预设时长范围和电器的类型确定，例如预设时长范围越长，预设差值阈值越大。
因此，此时还需要基于差值和预设差值阈值进行异常检测。当偏差的绝对值超过预设差值阈值的时候，可确定电器的工作性能异常，此时异常检测结果可为工作性能异常。当偏差的绝对值未超过预设差值阈值的时候，可确定电器的工作性能正常，此时异常检测结果可为工作性能正常。
例如，可设置压缩机性能预警标志，压缩机性能预警标志初始状态为关闭状态。在确定电器的工作性能异常的情况下，可启动压缩机性能预警标志。在确定电器的工作性能正常的情况下，可持续关闭启动压缩机性能预警标志。如此便可确保技术人员或用户可及时发现压缩机的工作性能异常这一情况。
本申请实施例提供的异常检测方法可根据电器的工作性能位于合理范围内对应的预设时长设置预设时长区间，然后根据预设时长区间内的电器的运行功率和预设功率阈值确定基础偏差率，该基础偏差率可以理解为电器的允许偏差率。接着，可根据电器的当前功率和预设功率阈值确定当前偏差率，并基于当前偏差率和基础偏差率的偏差进行异常检测，得到异常检测结果。可以理解，在偏差显示当前偏差率和基础偏差率之间的差距较大的情况下，可确认此时电器的工作性能发生异常。如此便可在电器的运行过程中，可根据当前偏差率来时刻监视电器的工作性能，从而可及时且准确地获取到电器工作性能异常的信息。
如此，本申请构建了一种自学习的动态功率阈值模型，其中使用的功率阈值为基础偏差率，而基础偏差率根据电器的实际运行情况确定。具体地，本申请考虑了电器单品的出厂差异，和阈值模型在实验室与真实场景的误差，把将电器出厂后的预设时长区间对应的运行功率作为阈值模型的学习基础，即根据电器出厂后的预设时长区间对应的运行功率确定基础偏差率，而不是直接根据实验室中得到的数据作为阈值模型的学习基础。同时还考虑了外部运行工况的影响，根据电器在具体环境中以正常工作性能运行时的状态参数如高压，低压，频率确定的预设功率阈值作为阈值模型的学习基础，使得阈值模型能够根据电器出厂后的实际情况确定阈值，即根据电器出厂后的运行数据确定预设功率阈值，从而确定基础偏差率，而不是简单地以运行时长作为阈值模型的学习基础，即以运行时长作为阈值来进行异常检测。因此，本申请的阈值模型中所使用的阈值(即基础偏差率)能够更准确地进行电器的异常检测，以解决电器运行过程中性能诊断问题，使得能够更准确地获取到电器工作性能异常的信息。
请参阅图2，在一些实施例中，步骤011：获取电器的运行时长位于预设时长区间内的电器的运行功率，并根据运行功率和预设功率阈值确定基础偏差率，包括以下步骤：
步骤0111：在电器的运行时长位于预设时长区间的情况下，获取电器在当前时刻的当前功率；
步骤0112：根据当前功率和预设功率阈值确定第一偏差率；
步骤0113：根据第一偏差率和电器的第一历史偏差率确定基础偏差率，第一历史偏差率根据当前时刻以前的时刻对应的第一偏差率确定。
具体地，在电器的运行时长位于预设时长区间的情况下，每获取一次当前功率就计算一次第一偏差率，使得后续可根据多个第一偏差率来确定基础偏差率。
首先，获取电器在当前时刻的当前功率，并根据当前功率和预设功率阈值确定第一偏差率。然后再获取达到当前时刻以前的时刻确定的第一偏差率，例如当前时刻为80小时的情况下，则获取时刻为0-79小时确定的所有第一偏差率，并将此时得到的所有第一偏差率确定为第一历史偏差率。然后结合此时求得的第一偏差率和第一历史偏差率求得预设时长区间内的平均偏差率，并将该平均偏差率确定为基础偏差率。例如，可根据下述公式确定基础偏差率：
其中，i＝1,…,n，ATTcmp(base)为基础偏差率，ATTcmp(i)对应第一偏差率和第一历史偏差率。
或者，也可统一获取电器的运行时长位于预设时长区间内的所有运行功率，然后再根据运行功率和预设功率阈值确定每个运行功率对应的第一偏差率，最后再根据多个第一偏差率的均值确定基础偏差率。
如此，可根据电器的运行时长位于预设时长区间内的多个运行时长对应的偏差率来确定基础偏差率，使得基础偏差率可较准确的代表工作性能位于合理范围内时的偏差率，从而可确保后续异常检测的准确性。
请参阅图3，在一些实施例中，步骤012：根据电器的当前功率和预设功率阈值确定当前偏差率包括以下步骤：
步骤0121：在运行时长大于预设时长区间的最大值的情况下，根据电器的当前功率和预设功率阈值确定当前偏差率。
具体地，运行时长可理解为是电器在出厂后的累计工作时长，此时预设时长区间可根据工作性能处于正常范围内的工作时长确定，例如预设时长区间可为0-100小时。在预设时长区间内可认为电器的工作性能一定处于合理范围内，因此只需要在运行时长大于预设时长区间的最大值的情况下进行异常检测，即在运行时长大于预设时长区间的最大值的情况下，根据电器的当前功率和预设功率阈值确定当前偏差率，使得在运行时长位于预设时长区间的时候不需要进行异常检测，从而减少计算量。
在另一些实施例中，运行时长也可理解为电器在每天工作时的累计工作时长，使得基础偏差率可受到工作性能正常衰减所带来的影响，此时预设时长区间对应每天的工作时长的区间，例如为0-2小时。可以理解，此时在运行时长位于预设时长区间的情况下，工作性能也可能会发生异常，因此需要做异常检测，此时可获取当前功率，并根据前一天的预设时长区间对应的基础偏差率进行异常检测。在确定运行时长位于预设时长区间时，工作性能正常的情况下，可根据运行时长位于预设时长区间的运行效率重新确定基础偏差率。然后再获取当前功率，并根据当前功率和重新确定的基础偏差率来进行异常检测。如此，基础偏差率可受到工作性能正常衰减所带来的影响，使得在进行异常检测时不会将正常衰减作为异常而进行警告，从而能够准确地排查出电器除正常衰减以外的工作性能异常变化的情况。
请参阅图4，在一些实施例中，步骤0121：在运行时长大于预设时长区间的最大值的情况下，根据电器的当前功率和预设功率阈值确定当前偏差率，包括以下步骤：
步骤01211：在运行时长大于预设时长区间的最大值的情况下，根据电器在当前时刻的当前功率和预设功率阈值确定第二偏差率；
步骤01212：根据电器的第二历史偏差率和第二偏差率确定当前偏差率，第二历史偏差率根据当前时刻以前的时刻对应的第二偏差率确定。
具体地，与第一偏差率类似的，在运行时长大于预设时长区间的最大值的情况下，每获取一次当前功率就计算一次第二偏差率，使得后续可根据多个第二偏差率来确定当前偏差率。
在运行时长大于预设时长区间的最大值后需要进行异常检测，此时获取电器在当前时刻的当前功率，并根据当前功率和预设功率阈值确定第二偏差率。然后再获取达到当前时刻以前的时刻确定的第二偏差率，例如预设时长区间为0-90小时，当前时刻为120小时，因此需要获取第91-119小时内确定的所有第二偏差率，并将此时得到的所有第二偏差率确定为第二历史偏差率。然后结合此时求得的第二偏差率和第二历史偏差率求得平均偏差率，并将该平均偏差率确定为当前偏差率。例如，可根据下述公式确定当前偏差率：
其中，i＝1,…,n，ATTcmp(stat)为当前偏差率，ATTcmp(i)对应第二偏差率和第二历史偏差率。
如此，可根据多个对应的运行时长大于预设时长区间的最大值的第二偏差率来确定当前偏差率，使得当前偏差率可代表电器在这段时间内的工作性能，后续异常检测可根据当前偏差率准确地检测这段时间内的工作性能是否正常，从而确保异常检测的准确性。
当然，也可直接根据第二偏差率确定当前偏差率，使得当前偏差率代表电器在当前时刻的工作性能，从而在确保异常检测的准确性的同时，还可减小计算量。
请参阅图5，在一些实施例中，步骤013：基于当前偏差率和基础偏差率的偏差进行异常检测，以确定异常检测结果，包括以下步骤：
步骤0131：确定当前偏差率和基础偏差率的偏差的绝对值；
步骤0132：在绝对值大于预设差值阈值的情况下，确定电器发生异常。
具体地，当前偏差率和基础偏差率之间的大小关系不固定，而预设差值阈值通常设置为正数。因此在进行异常检测的时候，可直接根据当前偏差率和基础偏差率的偏差的绝对值与预设差值阈值进行比较。此时首先计算当前偏差率和基础偏差率的偏差的绝对值，在绝对值大于预设差值阈值的情况下，可确认两者之间相差较大，此时可确定电器发生异常情况。在绝对值小于预设差值阈值的情况下，可确认两者相差较小，此时可确定电器正常工作。如此便可根据偏差的绝对值快速地确定出异常情况。
请参阅图6，在一些实施例中，步骤013：基于当前偏差率和基础偏差率的偏差进行异常检测，以确定异常检测结果，包括以下步骤：
步骤0133：计算当前偏差率和基础偏差率的偏差；
步骤0134：在偏差为负数，且偏差的绝对值大于预设差值阈值的情况下，确定电器的工作性能异常增加；
步骤0135：在偏差为正数，且偏差大于预设差值阈值的情况下，确定电器的工作性能异常衰减。
具体地，电器的工作性能发生异常的时候，工作性能可能会异常增加或异常衰减。此时可根据偏差的大小确定工作性能是异常增加还是异常衰减。同时，根据偏差率计算公式(上述公式1)可知，当前功率越大，偏差率越小。
在进行异常检测的时候，首先需要计算当前偏差率和基础偏差率的偏差，即将当前偏差率减去基础偏差率，以得到两者之间的差值。然后，可判断偏差的正负号，以判断工作性能是增加还是衰减。
在偏差为负数的情况下，可确认当前偏差率小于基础偏差率，当前功率大于基础偏差率对应的运行功率，可确定此时电器的运行功率增加。接着判断偏差的绝对值是否大于预设差值阈值，若大于，则代表电器的运行功率增加的幅度较大，可确定电器的工作性能异常增加。若小于，则代表电器的运行功率增加的幅度较小，还在允许范围内，可确定电器的工作性能正常。
在偏差为正数的情况下，可确认当前偏差率大于基础偏差率，即当前功率小于基础偏差率对应的运行功率，可确定此时电器的运行功率减小。接着判断偏差的绝对值是否大于预设差值阈值，若大于，则代表电器的运行功率减小的幅度较大，可确定电器的工作性能异常衰减。若小于，则代表电器的运行功率减小的幅度较小，还在允许范围内，可确定电器的工作性能正常。
如此，可根据偏差的正负号和预设差值阈值来准确地判断电器的工作性能是异常增加还是异常衰减，后续电器可根据检测结果的不同发出对应的异常信号，例如电器可包括警报灯，警报灯用于发出异常检测结果。在判断工作性能异常增加的情况下，警报灯可发出橙色光线；在判断工作性能异常衰减的情况下，警报灯可发出红色光线；在判断工作性能正常的情况下，警报灯可关闭或发出绿色光线。如此，用户或技术人员可根据异常信号直接确定电器的异常状况，以便于后续进行针对性的维修。
在一些实施例中，步骤013：基于当前偏差率和基础偏差率的偏差进行异常检测，以确定异常检测结果，包括以下步骤：
步骤01311：根据电器的累计使用时长确定电器的剩余使用寿命；
步骤01312：根据剩余使用寿命和预设关联关系，确定第一目标修正系数，第一预设关联关系包括多个预设剩余使用寿命和多个预设剩余使用寿命各自对应的第一预设偏差率修正系数；
步骤01313：根据第一目标修正系数对当前偏差率进行修正，得到第一修正偏差率；
步骤01314：确定第一修正偏差率和基础偏差率的偏差的绝对值；
步骤01315：在绝对值大于预设差值阈值的情况下，确定电器发生异常。
电器的剩余使用寿命通过电器的剩余使用时长表示。例如在电器运行的情况下，电器从刚出厂到不能使用期间，电器的总使用时长为T1，如果出厂后电器的累计使用时长为T2，那么电器的剩余使用时长为T1-T2，即剩余使用寿命为T1-T2。
考虑到电器使用时间的增加会导致硬件逐渐老化，这种老化现象会直接影响检测单元对运行功率检测的准确性，进而干扰偏差率的精确计算。为了解决这一问题，事先通过一系列实验事先标定了电器的剩余使用寿命对偏差率的影响，并据此确定了不同剩余使用寿命对应的偏差率修正系数，称为第一预设偏差率修正系数，这一措施能够有效校修正由于硬件老化所引起的偏差率计算误差，从而得到准确的偏差率。
通过上述基础偏差率的计算过程，基础偏差率通过“时刻为0-79小时的所有第一偏差率”确定，即基于此信息可以得到电器此时可以理解为一台新设备，故电器中硬件的老化现象几乎没有，即可以忽略老化现象对于偏差率的影响，即基础偏差率是在没有硬件老化的情况下得到，因此，采用剩余使用寿命对硬件存在老化现象时计算得到的偏差率进行修正之后，修正后的偏差率应该是小于修正之前的偏差率的，故第一预设关联关系中的修正系数的取值范围为(0,1]。
在得到当前偏差率和基础偏差率之后，计算出电器的剩余使用寿命，然后通过电器的剩余使用寿命查询上述第一预设关联关系，得到电器的剩余使用寿命对应的第一预设偏差率修正系数，即得到第一目标修正系数，然后采用第一目标修正系数对当前偏差率进行修正，得到第一修正偏差率，第一修正偏差率小于修正之前的当前偏差率。
得到第一修正偏差率之后，计算第一修正偏差率和基础偏差率的偏差的绝对值，如果该绝对值大于预设差值阈值，确定电器发生异常，否则，确定电器是正常的。
通过电器的剩余使用寿命对当前偏差率进行修正，考虑了电器中硬件的老化现象对应偏差率的影响，将修正后的当前偏差率与基础偏差率进行比较时，可以提高对于判断电器是否异常的准确性。
在一些实施例中，步骤013：基于当前偏差率和基础偏差率的偏差进行异常检测，以确定异常检测结果，包括以下步骤：
步骤01321：根据电器的累计使用时长确定电器的剩余使用寿命；
根据剩余使用寿命和预设关联关系，确定第二目标修正系数，预设关联关系包括多个预设剩余使用寿命和多个预设剩余使用寿命各自对应的预设偏差率修正系数；
步骤01322：根据第二目标修正系数对基础偏差率进行修正，得到第二修正偏差率；
步骤01323：确定当前偏差率和第二修正偏差率的偏差的绝对值；
步骤01324：在绝对值大于预设差值阈值的情况下，确定电器发生异常。
考虑到电器使用时间的增加会导致硬件逐渐老化，这种老化现象会直接影响检测单元对运行功率检测的准确性，进而干扰偏差率的精确计算。为了解决这一问题，事先通过一系列实验事先标定了电器的剩余使用寿命对偏差率的影响，并据此确定了不同剩余使用寿命对应的偏差率修正系数，称为第二预设偏差率修正系数，这一措施能够有效校修正由于硬件老化所引起的偏差率计算误差，从而得到准确的偏差率。
通过上述基础偏差率的计算过程，基础偏差率通过“时刻为0-79小时的所有第一偏差率”确定，即基于此信息可以得到电器此时可以理解为一台新设备，故电器中硬件的老化现象几乎没有，即可以忽略老化现象对于偏差率的影响，即基础偏差率是在没有硬件老化的情况下得到，因此，采用剩余使用寿命对硬件不存在老化现象时计算得到的偏差率进行修正之后，修正后的偏差率应该是小于修正之前的偏差率的，故第二预设关联关系中的修正系数的取值是大于1的。
在得到当前偏差率和基础偏差率之后，计算出电器的剩余使用寿命，然后通过电器的剩余使用寿命查询上述第二预设关联关系，得到电器的剩余使用寿命对应的第二预设偏差率修正系数，即得到第二目标修正系数，然后采用第二目标修正系数对基础偏差率进行修正，得到第二修正偏差率，第二修正偏差率大于修正之前的基础偏差率。
得到第二修正偏差率之后，计算第二修正偏差率和当前偏差率的偏差的绝对值，如果该绝对值大于预设差值阈值，确定电器发生异常，否则，确定电器是正常的。
通过电器的剩余使用寿命对基础偏差率进行修正，考虑了电器中硬件的老化现象对应偏差率的影响，将当前偏差率与修正后的基础偏差率进行比较时，可以提高对于判断电器是否异常的准确性。
在一些实施例中，步骤013：在当前偏差率和基础偏差率的偏差的绝对值、第一修正偏差率和基础偏差率的偏差的绝对值，以及当前偏差率和第二修正偏差率的偏差的绝对值中的一者大于预设差值阈值的情况下，确定电器发生异常之后，异常检测方法还包括以下步骤：
关闭电器，并调整预设时长区间的最大值，得到新的时间区间；
在电器的关闭时长达到预设关闭时长的情况下，开启电器；
在电器的开启时长达到预设开启时长的情况下，将新的时间区间作为预设时长区间，并执行步骤011，获取电器的运行时长位于预设时长区间内的电器的运行功率，并根据运行功率和预设功率阈值确定基础偏差率；
在再次确定当前偏差率和基础偏差率的偏差的绝对值、第一修正偏差率和基础偏差率的偏差的绝对值，以及当前偏差率和第二修正偏差率的偏差的绝对值中的一者大于预设差值阈值的情况下，输出电器发生异常的提示信息。
考虑到电器出现的异常可能短暂的，可能在经历关机和开机之后就会消除，因此为了避免误判，在当前偏差率和基础偏差率的偏差的绝对值、第一修正偏差率和基础偏差率的偏差的绝对值，以及当前偏差率和第二修正偏差率的偏差的绝对值中的一者大于预设差值阈值的情况下，先关闭电器，并开始关闭时长计时，同时调整预设时长区间的最大值，得到新的时间区间，新的时间区间的最大值小于调整之前的时间区间的最大值，或者，新的时间区间的最大值小于调整之前的时间区间的最大值。
在电器的关闭时长达到预设关闭时长时，然后开启电器，并开始开启时长计时，在电器的开启时长达到预设开启时长，将新的时间区间作为调整之前的预设时长区间，并执行步骤011，从而可以再次得到当前偏差率和基础偏差率的偏差的绝对值、第一修正偏差率和基础偏差率的偏差的绝对值，以及当前偏差率和第二修正偏差率的偏差的绝对值，如果这三个绝对值中的一者大于预设差值阈值，则进一步确定电器发生异常，并输出电器发生异常的提示信息(例如输出表示电器性能异常的预警标志)，从而避免电器发生异常的误判，提高了对于判断电器是否异常的准确性。
请参阅图7，为便于更好地实施本申请实施例提供的异常检测方法，本申请实施例还提供一种异常检测装置10。该异常检测装置10可以包括第一确定模块11、第二确定模块12和异常检测模块13。第一确定模块11用于获取电器的运行时长位于预设时长区间内的电器的运行功率，并根据运行功率和预设功率阈值确定基础偏差率；第二确定模块12用于根据电器的当前功率和预设功率阈值确定当前偏差率；异常检测模块13用于基于当前偏差率和基础偏差率的偏差进行异常检测，以确定异常检测结果。
第一确定模块11具体用于在电器的运行时长位于预设时长区间的情况下，获取电器在当前时刻的当前功率；根据当前功率和预设功率阈值确定第一偏差率；根据第一偏差率和电器的第一历史偏差率确定基础偏差率，第一历史偏差率根据当前时刻以前的时刻对应的第一偏差率确定。
第一确定模块11具体用于确定第一偏差率和第一历史偏差率的平均偏差率；将平均偏差率确定为基础偏差率。
第二确定模块12具体用于在运行时长大于预设时长区间的最大值的情况下，根据电器的当前功率和预设功率阈值确定当前偏差率。
第二确定模块12具体还用于在运行时长大于预设时长区间的最大值的情况下，根据电器在当前时刻的当前功率和预设功率阈值确定第二偏差率；根据电器的第二历史偏差率和第二偏差率确定当前偏差率，第二历史偏差率根据当前时刻以前的时刻对应的第二偏差率确定。
构建模块13具体用于确定当前偏差率和基础偏差率的偏差的绝对值；在绝对值大于预设差值阈值的情况下，确定电器发生异常。
构建模块13具体还用于根据电器的累计使用时长确定电器的剩余使用寿命；根据剩余使用寿命和预设关联关系，确定第一目标修正系数，第一预设关联关系包括多个预设剩余使用寿命和多个预设剩余使用寿命各自对应的第一预设偏差率修正系数；根据第一目标修正系数对当前偏差率进行修正，得到第一修正偏差率；确定第一修正偏差率和基础偏差率的偏差的绝对值；在绝对值大于预设差值阈值的情况下，确定电器发生异常。
构建模块13具体还用于根据电器的累计使用时长确定电器的剩余使用寿命；根据剩余使用寿命和预设关联关系，确定第二目标修正系数，预设关联关系包括多个预设剩余使用寿命和多个预设剩余使用寿命各自对应的预设偏差率修正系数；根据第二目标修正系数对基础偏差率进行修正，得到第二修正偏差率；确定当前偏差率和第二修正偏差率的偏差的绝对值；在绝对值大于预设差值阈值的情况下，确定电器发生异常。
该异常检测装置10还包括：判断单元，判断单元用于在绝对值大于预设差值阈值的情况下，确定电器发生异常之后，关闭电器，并调整预设时长区间的最大值，得到新的时间区间；在电器的关闭时长达到预设关闭时长的情况下，开启电器；在电器的开启时长达到预设开启时长的情况下，将新的时间区间作为预设时长区间，并执行获取电器的运行时长位于预设时长区间内的电器的运行功率，并根据运行功率和预设功率阈值确定基础偏差率的步骤；在再次确定绝对值大于预设差值阈值的情况下，输出电器发生异常的提示信息。
构建模块13具体还用于计算当前偏差率和基础偏差率的差值；在偏差为负数，且偏差的绝对值大于预设差值阈值的情况下，确定电器的工作性能异常增加；在偏差为正数，且偏差大于预设差值阈值的情况下，确定电器的工作性能异常衰减。
上文中结合附图从功能模块的角度描述了异常检测装置10，该功能模块可以通过硬件形式实现，也可以通过软件形式的指令实现，还可以通过硬件和软件模块组合实现。具体地，本申请实施例中的方法实施例的各步骤可以通过处理器中的硬件的集成逻辑电路和/或软件形式的指令完成，结合本申请实施例公开的方法的步骤可以直接体现为硬件编码处理器执行完成，或者用编码处理器中的硬件及软件模块组合执行完成。可选地，软件模块可以位于随机存储器，闪存、只读存储器、可编程只读存储器、电可擦写可编程存储器、寄存器等本领域的成熟的存储介质中。该存储介质位于存储器，处理器读取存储器中的信息，结合其硬件完成上述方法实施例中的步骤。
请参阅图8，本申请实施例提供的电器100包括处理器20、存储器30及计算机程序，其中，计算机程序被存储在存储器30中，并且被处理器20执行，计算机程序包括用于执行上述任一实施例的异常检测方法的指令。
请参阅图9，本申请实施例还提供了一种计算机可读存储介质300，其上存储有计算机程序310，计算机程序310被处理器320执行的情况下，实现上述任意一种实施例的异常检测方法的步骤，为了简洁，在此不再赘述。
在本说明书的描述中，参考术语“某些实施例”、“一个例子中”、“示例地”等的描述意指结合实施例或示例描述的具体特征、结构、材料或者特点包含于本申请的至少一个实施例或示例中。在本说明书中，对上述术语的示意性表述不一定指的是相同的实施例或示例。而且，描述的具体特征、结构、材料或者特点可以在任何的一个或多个实施例或示例中以合适的方式结合。此外，在不相互矛盾的情况下，本领域的技术人员可以将本说明书中描述的不同实施例或示例以及不同实施例或示例的特征进行结合和组合。
流程图中或在此以其他方式描述的任何过程或方法描述可以被理解为，表示包括一个或更多个用于实现特定逻辑功能或过程的步骤的可执行指令的代码的模块、片段或部分，并且本申请的优选实施例的范围包括另外的实现，其中可以不按所示出或讨论的顺序，包括根据所涉及的功能按基本同时的方式或按相反的顺序，来执行功能，这应被本申请的实施例所属技术领域的技术人员所理解。
尽管上面已经示出和描述了本申请的实施例，可以理解的是，上述实施例是示例性的，不能理解为对本申请的限制，本领域的普通技术人员在本申请的范围内可以对上述实施例进行变化、修改、替换和变型。

Claims (15)

  1. 一种异常检测方法，其中，应用于电器，所述方法包括：
    获取电器的运行时长位于预设时长区间内的电器的运行功率，并根据所述运行功率和预设功率阈值确定基础偏差率；
    根据所述电器的当前功率和所述预设功率阈值确定当前偏差率；
    基于所述当前偏差率和所述基础偏差率的偏差进行异常检测，以确定异常检测结果。
  2. 根据权利要求1所述的异常检测方法，其中，所述电器包括压缩机，所述压缩机的预设功率阈值根据所述压缩机的采集频率、采集低压和采集高压确定。
  3. 根据权利要求1所述的异常检测方法，其中，所述获取电器的运行时长位于预设时长区间内的电器的运行功率，并根据所述运行功率和预设功率阈值确定基础偏差率，包括：
    在所述电器的运行时长位于预设时长区间的情况下，获取电器在当前时刻的当前功率；
    根据所述当前功率和所述预设功率阈值确定第一偏差率；
    根据所述第一偏差率和所述电器的第一历史偏差率确定所述基础偏差率，所述第一历史偏差率根据当前时刻以前的时刻对应的第一偏差率确定。
  4. 根据权利要求3所述的异常检测方法，其中，所述根据所述第一偏差率和所述电器的第一历史偏差率确定所述基础偏差率，包括：
    确定所述第一偏差率和所述第一历史偏差率的平均偏差率；
    将所述平均偏差率确定为所述基础偏差率。
  5. 根据权利要求1所述的异常检测方法，其中，所述根据所述电器的当前功率和所述预设功率阈值确定当前偏差率，包括：
    在所述运行时长大于所述预设时长区间的最大值的情况下，根据所述电器的当前功率和预设功率阈值确定当前偏差率。
  6. 根据权利要求5所述的异常检测方法，其中，所述在所述运行时长大于所述预设时长区间的最大值的情况下，根据所述电器的当前功率和预设功率阈值确定当前偏差率，包括：
    在所述运行时长大于所述预设时长区间的最大值的情况下，根据所述电器在当前时刻的当前功率和所述预设功率阈值确定第二偏差率；
    根据所述电器的第二历史偏差率和所述第二偏差率确定所述当前偏差率，所述第二历史偏差率根据当前时刻以前的时刻对应的第二偏差率确定。
  7. 根据权利要求1所述的异常检测方法，其中，所述基于所述当前偏差率和所述基础偏差率的偏差进行异常检测，以确定异常检测结果包括：
    确定所述当前偏差率和所述基础偏差率的偏差的绝对值；
    在所述绝对值大于预设差值阈值的情况下，确定所述电器发生异常。
  8. 根据权利要求1所述的异常检测方法，其中，所述基于所述当前偏差率和所述基础偏差率的偏差进行异常检测，以确定异常检测结果包括：
    根据所述电器的累计使用时长确定所述电器的剩余使用寿命；
    根据所述剩余使用寿命和预设关联关系，确定第一目标修正系数，所述第一预设关联关系包括多个预设剩余使用寿命和所述多个预设剩余使用寿命各自对应的第一预设偏差率修正系数；
    根据所述第一目标修正系数对所述当前偏差率进行修正，得到第一修正偏差率；
    确定所述第一修正偏差率和所述基础偏差率的偏差的绝对值；
    在所述绝对值大于预设差值阈值的情况下，确定所述电器发生异常。
  9. 根据权利要求1所述的异常检测方法，其中，所述基于所述当前偏差率和所述基础偏差率的偏差进行异常检测，以确定异常检测结果包括：
    根据所述电器的累计使用时长确定所述电器的剩余使用寿命；
    根据所述剩余使用寿命和预设关联关系，确定第二目标修正系数，所述预设关联关系包括多个预设剩余使用寿命和所述多个预设剩余使用寿命各自对应的预设偏差率修正系数；
    根据所述第二目标修正系数对所述基础偏差率进行修正，得到第二修正偏差率；
    确定所述当前偏差率和所述第二修正偏差率的偏差的绝对值；
    在所述绝对值大于预设差值阈值的情况下，确定所述电器发生异常。
  10. 根据权利要求7-8中任意一项所述的异常检测方法，其中，所述在所述绝对值大于预设差值阈值的情况下，确定所述电器发生异常之后，所述异常检测方法还包括：
    关闭所述电器，并调整所述预设时长区间的最大值，得到新的时间区间；
    在所述电器的关闭时长达到预设关闭时长的情况下，开启所述电器；
    在所述电器的开启时长达到预设开启时长的情况下，将所述新的时间区间作为所述预设时长区间，并执行所述获取电器的运行时长位于预设时长区间内的电器的运行功率，并根据所述运行功率和预设功率阈值确定基础偏差率的步骤；
    在再次确定所述绝对值大于所述预设差值阈值的情况下，输出所述电器发生异常的提示信息。
  11. 根据权利要求1所述的异常检测方法，其中，所述基于所述当前偏差率和所述基础偏差率的偏差进行异常检测，以确定异常检测结果包括：
    计算所述当前偏差率和所述基础偏差率的偏差；
    在所述偏差为负数，且所述偏差的绝对值大于所述预设差值阈值的情况下，确定所述电器的工作性能异常增加；
    在所述偏差为正数，且所述偏差大于所述预设差值阈值的情况下，确定所述电器的工作性能异常衰减。
  12. 根据权利要求1所述的异常检测方法，其中，所述预设时长区间根据云端大数据中与所述电器的类型相同的已运行电器的功率偏差数据及运行时长确定。
  13. 一种异常检测装置，其中，应用于电器，所述装置包括：
    第一确定模块，用于获取电器的运行时长位于预设时长区间内的电器的运行功率，并根据所述运行功率和预设功率阈值确定基础偏差率；
    第二确定模块，用于根据所述电器的当前功率和所述预设功率阈值确定当前偏差率；
    异常检测模块，用于基于所述当前偏差率和所述基础偏差率的偏差进行异常检测，以确定异常检测结果。
  14. 一种电器，其中，包括：
    处理器、存储器；及
    计算机程序，其中，所述计算机程序被存储在所述存储器中，并且被所述处理器执行，所述计算机程序包括用于执行权利要求1至12任意一项所述的异常检测方法的指令。
  15. 一种包含计算机程序的非易失性计算机可读存储介质，所述计算机程序被处理器执行时，使得所述处理器执行权利要求1-12任意一项所述的异常检测方法。
PCT/CN2025/090513 2024-05-14 2025-04-22 异常检测方法、异常检测装置、电器和可读存储介质 Pending WO2025237017A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202410597750.5 2024-05-14
CN202410597750.5A CN118532313A (zh) 2024-05-14 2024-05-14 异常检测方法、异常检测装置、电器和可读存储介质

Publications (1)

Publication Number Publication Date
WO2025237017A1 true WO2025237017A1 (zh) 2025-11-20

Family

ID=92392656

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2025/090513 Pending WO2025237017A1 (zh) 2024-05-14 2025-04-22 异常检测方法、异常检测装置、电器和可读存储介质

Country Status (2)

Country Link
CN (1) CN118532313A (zh)
WO (1) WO2025237017A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN121387901A (zh) * 2025-12-24 2026-01-23 杭州华罡智能科技有限公司 一种电能表智能数据处理方法、系统及终端

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN118532313A (zh) * 2024-05-14 2024-08-23 广东美的暖通设备有限公司 异常检测方法、异常检测装置、电器和可读存储介质
CN119780586B (zh) * 2025-03-10 2025-06-24 宁波戴维医疗器械股份有限公司 一种高频发生器的老化测试方法、设备及存储介质
CN120542749B (zh) * 2025-07-29 2025-09-19 中石油深圳新能源研究院有限公司 风电场设备管理方法、装置、设备及存储介质

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2009020721A (ja) * 2007-07-12 2009-01-29 Daikin Ind Ltd 空調機の劣化判定装置、空調システム、劣化判定方法および劣化判定プログラム
CN112629900A (zh) * 2020-12-10 2021-04-09 浙江国视科技有限公司 一种家用电器监测方法及系统
CN113219272A (zh) * 2021-04-06 2021-08-06 国网江苏省电力有限公司营销服务中心 一种基于非侵入式监测预测家用电器老化的方法及系统
CN113647835A (zh) * 2021-08-19 2021-11-16 珠海格力电器股份有限公司 家电设备功率异常降低的判断方法、装置和家电设备
CN116186479A (zh) * 2022-11-28 2023-05-30 中联重科股份有限公司 用于泵车工况数据的处理方法、处理器、装置及存储介质
CN118532313A (zh) * 2024-05-14 2024-08-23 广东美的暖通设备有限公司 异常检测方法、异常检测装置、电器和可读存储介质

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2009020721A (ja) * 2007-07-12 2009-01-29 Daikin Ind Ltd 空調機の劣化判定装置、空調システム、劣化判定方法および劣化判定プログラム
CN112629900A (zh) * 2020-12-10 2021-04-09 浙江国视科技有限公司 一种家用电器监测方法及系统
CN113219272A (zh) * 2021-04-06 2021-08-06 国网江苏省电力有限公司营销服务中心 一种基于非侵入式监测预测家用电器老化的方法及系统
CN113647835A (zh) * 2021-08-19 2021-11-16 珠海格力电器股份有限公司 家电设备功率异常降低的判断方法、装置和家电设备
CN116186479A (zh) * 2022-11-28 2023-05-30 中联重科股份有限公司 用于泵车工况数据的处理方法、处理器、装置及存储介质
CN118532313A (zh) * 2024-05-14 2024-08-23 广东美的暖通设备有限公司 异常检测方法、异常检测装置、电器和可读存储介质

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN121387901A (zh) * 2025-12-24 2026-01-23 杭州华罡智能科技有限公司 一种电能表智能数据处理方法、系统及终端

Also Published As

Publication number Publication date
CN118532313A (zh) 2024-08-23

Similar Documents

Publication Publication Date Title
WO2025237017A1 (zh) 异常检测方法、异常检测装置、电器和可读存储介质
US7908505B2 (en) Apparatus, system, and method for event, time, and failure state recording mechanism in a power supply
KR20190003580A (ko) 감시 장치, 대상 장치의 감시 방법 및 프로그램
CN109556646B (zh) 一种气体传感器校准方法、装置、存储介质及终端
CN114578280B (zh) 智能电表故障确定方法、装置、计算机设备和存储介质
CN112648710A (zh) 排气温度传感器故障检测方法、装置及空调设备
CN112629709A (zh) 温度传感器故障检测方法、检测装置及电动车控制器
JP6832485B2 (ja) ガスメータ
CN119493039A (zh) 电池系统健康检测方法、电子设备、存储介质及程序产品
CN111880028A (zh) 一种电器设备检测方法、装置、检测设备及存储介质
CN118483573A (zh) 继电器状态评估方法、装置、存储介质及车辆
CN114490213A (zh) 电脑主板故障的监测诊断方法、系统、装置及存储介质
CN121112492B (zh) 一种采暖热水炉用风压开关控制系统及其方法
CN111059696B (zh) 一种功率模块温度检测控制方法、计算机可读存储介质及空调
CN113375287B (zh) 低压传感器故障识别控制方法、装置及多联机空调系统
CN115388518A (zh) 用于空调的故障预警方法及装置、空调、存储介质
CN111674224B (zh) 空调器的控制方法、装置、空调器和存储介质
CN112804741B (zh) 功率控制的方法和装置
CN106991028B (zh) 一种计算机的温度异常处理方法及系统
JP6767078B2 (ja) 異常検出装置及び異常検出方法
US11181444B2 (en) Method for the detection of a glow plug replacement
JP7437163B2 (ja) 診断装置、診断方法およびプログラム
CN116066957B (zh) 一种空调缺氟的判断方法、装置、电子设备和存储介质
CN119718043B (zh) 适用于微型电脑主机的电源状态监测方法
US12066481B2 (en) System and method for monitoring activity over time-domain of electrical devices

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: 25802466

Country of ref document: EP

Kind code of ref document: A1