Fault monitoring method, device and equipment based on odor, medium and program product

Abstract

The invention provides a fault monitoring method, device and equipment based on odor, a medium and a program product. The method comprises the following steps: acquiring at least one kind of smell sensing data inside to-be-detected equipment; aiming at each kind of smell sensing data, extracting data change characteristics of the kind of smell sensing data to obtain smell characteristic data corresponding to the kind of smell sensing data; the smell characteristic data is used for representing the smell change condition in the to-be-detected equipment; and monitoring the fault state of the to-be-detected equipment according to the smell characteristic data corresponding to each kind of smell sensing data. According to the invention, the fault monitoring precision and real-time performance of the to-be-monitored equipment can be improved, and the monitoring cost is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of odor monitoring, and in particular to an odor-based fault monitoring method, device, equipment, medium and program product. Background Art

[0002] The power supply and distribution system mainly includes high and low voltage switchgear, cable wells, cable interlayers, high voltage inverters and other equipment, which work together to ensure the smooth operation of the power supply and distribution system. In order to ensure the smooth operation of the power supply and distribution system, fault monitoring of the above-mentioned high and low voltage switchgear, cable wells, cable interlayers, high voltage inverters and other equipment is also particularly important.

[0003] Among them, the common faults of high and low voltage switchgear, cable wells, cable interlayers and high voltage inverters mainly include: partial discharge faults and cable overheating faults. Partial discharge faults will accelerate the aging and degradation of insulation materials, resulting in a decrease in its dielectric strength, thereby reducing insulation performance, and eventually causing insulation breakdown, resulting in equipment damage. Cable overheating faults will cause the cable insulation layer to age, or even insulation breakdown, resulting in phase-to-phase short circuit tripping. In addition, it may also cause fire accidents, threatening the safety of equipment and personnel. Therefore, it is necessary to monitor partial discharge faults and cable overheating faults in the power supply and distribution system in real time to ensure the safe operation of the power supply and distribution system.

[0004] In the related art, for partial discharge faults, pulse current method, ultra-high frequency method, transient earth voltage (TEV) method, ultrasonic detection method and other methods are usually used for temporary monitoring. However, the pulse current method is sensitive to sudden change signals, but its anti-electromagnetic interference ability is not strong and it is easily affected by electromagnetic interference. Ultrasonic detection method and ultra-high frequency method can effectively avoid the influence of electromagnetic interference, but the propagation characteristics of ultrasonic signals and ultra-high frequency electromagnetic wave signals inside the equipment are not easy to analyze, which makes it difficult to determine the detection results; TEV method detects partial discharge by capturing the instantaneous voltage signal to the ground generated on the surface of the equipment shell when the electromagnetic wave propagates in the air and the surface of the equipment shell, but it cannot quantitatively analyze the partial discharge amount and is easily interfered by external signals. In addition, most of the monitoring equipment used for partial discharge fault monitoring according to the above method is expensive, ranging from hundreds of thousands to more than one million. Moreover, the above monitoring method also requires long-term learning and experience accumulation of monitoring personnel to ensure a certain monitoring accuracy.

[0005] In case of cable overheating, distributed optical fiber temperature measurement technology, infrared imaging technology, HeatTag wireless sensors, etc. are usually used to monitor the cable temperature, thereby realizing the monitoring of cable overheating faults. However, most of the above temperature measurement methods require the use of professional temperature measurement equipment, which has high installation and maintenance costs, and the temperature measurement results are also affected by the installation location and usage, which makes it impossible to effectively guarantee the temperature measurement accuracy. In summary, in the related technologies, most of the monitoring methods for partial discharge faults and cable overheating faults have the problem of high monitoring costs and being easily affected by external interference, resulting in the inability to ensure high monitoring accuracy. Summary of the invention

[0006] The embodiments of the present invention provide a method, device, electronic device and storage medium for fault monitoring based on odor, so as to solve the problems of high fault monitoring cost and low monitoring accuracy in power supply and distribution system.

[0007] In a first aspect, an embodiment of the present invention provides a fault monitoring method based on smell, comprising: Acquiring at least one type of odor sensing data inside the device under test; For each type of odor sensing data, extract the data change characteristics of the odor sensing data to obtain odor characteristic data corresponding to the odor sensing data; the odor characteristic data is used to characterize the odor change in the device to be tested; The fault status of the device under test is monitored according to the odor characteristic data corresponding to each type of odor sensing data.

[0008] In a possible implementation, the odor characteristic data corresponding to each type of odor sensing data includes: odor characteristic data corresponding to each type of odor sensing data at different times; For each type of odor sensing data, extract the data change characteristics of the odor sensing data to obtain odor feature data corresponding to the odor sensing data, including: For each type of odor sensing data, obtain the time series data corresponding to the odor sensing data at the current moment; the time series data includes the odor sensing data corresponding to the current moment and different moments before the current moment; Determine first data and second data corresponding to the current moment; the first data is determined by odor sensor data within a first preset time interval in the time series data, and the second data is determined by odor sensor data within a second preset time interval in the time series data; the first preset time interval is earlier than the second preset time interval; the current moment is within the second preset time interval; According to the ratio of the first data to the second data, the odor characteristic data corresponding to the odor sensing data at the current moment is determined.

[0009] In a possible implementation manner, determining the first data corresponding to the current moment includes: Determine the first data corresponding to the previous moment as the first data corresponding to the current moment; Based on the ratio of the first data and the second data corresponding to the current moment, determine whether the odor of the device to be tested has reached a stable state. If so, update the first data corresponding to the current moment according to the time series data corresponding to the current moment; if not, keep the first data corresponding to the current moment.

[0010] In a possible implementation, judging whether the smell of the device under test has reached a stable state based on the ratio of the first data and the second data corresponding to the current moment, and if so, updating the first data corresponding to the current moment according to the time series data corresponding to the current moment; if not, maintaining the first data corresponding to the current moment, includes: Based on the ratio of the first data and the second data corresponding to the current moment, judging whether the first condition and the second condition are met, wherein the first condition is that the odor of the device to be tested reaches a stable state, and the second condition is that the accumulated duration exceeds the preset reset duration; If the first condition is met or the second condition is met, the first data corresponding to the current moment is updated according to the time series data corresponding to the current moment, and the accumulated duration is cleared; If the first condition is not met and the second condition is not met, the first data corresponding to the current moment is retained; For each type of odor sensing data, after determining the odor characteristic data corresponding to the type of odor sensing data at the current moment according to the ratio of the first data to the second data, the method further includes: For each type of odor sensing data, the cumulative duration corresponding to the odor sensing data is added to the unit duration to obtain an updated cumulative duration corresponding to the odor sensing data, wherein the unit duration is the time difference between two adjacent moments.

[0011] In a possible implementation, judging whether the first condition is satisfied based on the ratio of the first data to the second data corresponding to the current moment includes: Detecting whether the ratio of the first data and the second data corresponding to the current moment exceeds a first preset interval; If the ratio of the first data to the second data does not exceed a first preset interval, it is determined that the first condition is met; If the ratio of the first data to the second data exceeds a first preset interval, it is determined that the first condition is not met; the upper limit value of the first preset interval is greater than 1, and the lower limit value of the first preset interval is less than 1.

[0012] In a second aspect, an embodiment of the present invention provides a fault monitoring device based on smell, comprising: An acquisition module, used to acquire at least one kind of odor sensing data inside the device to be tested; An extraction module is used to extract data change characteristics of each type of odor sensing data, and obtain odor characteristic data corresponding to the odor sensing data; the odor characteristic data is used to characterize the odor change in the device to be tested; The monitoring module is used to monitor the fault status of the device under test according to the odor characteristic data corresponding to each type of odor sensing data.

[0013] In a third aspect, an embodiment of the present invention provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the processor implements the steps of the method described in the first aspect or any possible implementation manner of the first aspect.

[0014] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method described in the first aspect or any possible implementation method of the first aspect are implemented.

[0015] In a fifth aspect, an embodiment of the present invention provides a computer program product, including a computer program, which, when executed by a processor, implements the method in the first aspect or any possible implementation manner of the first aspect.

[0016] The embodiment of the present invention provides a method, device, equipment, medium and program product for fault monitoring based on odor. By extracting the data change characteristics of the odor sensor data inside the device to be tested, the odor characteristic data can be determined, and then the fault state of the device to be tested can be monitored according to the odor characteristic data. Here, the odor characteristic data can be used to characterize the odor change inside the device to be tested. The embodiment of the present invention performs fault monitoring through the odor change inside the device to be tested, which will not be affected by external interference such as the electromagnetic environment inside the device to be tested, the installation location, and the experience of the monitoring personnel, and can effectively improve the fault monitoring accuracy. Moreover, the odor-based fault monitoring method provided by the present application has lower cost and is more economically feasible. In addition, the embodiment of the present invention can also realize the monitoring of partial discharge faults and cable overheating faults at the same time, without the need to monitor the two types of faults separately, which can effectively simplify the fault monitoring method in the power supply and distribution system. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0018] Figure 1 is a flow chart of an implementation of a fault monitoring method based on odor provided by an embodiment of the present invention; Figure 2 is a schematic diagram of a display interface provided by an embodiment of the present invention; Figure 3 is a flowchart for determining odor characteristic data provided by an embodiment of the present invention; Figure 4 is a flowchart for determining first data according to an embodiment of the present invention; Figure 5 is a flowchart for determining odor characteristic data provided by another embodiment of the present invention; Figure 6 is a schematic structural diagram of an odor-based fault monitoring device provided by an embodiment of the present invention; Figure 7 It is a schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0019] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present invention. However, it should be clear to those skilled in the art that the present invention may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present invention.

[0020] In the related technologies, most of them adopt specific monitoring methods to monitor partial discharge faults and cable overheating faults in the power supply and distribution system respectively. Among them, for partial discharge faults, pulse current method, ultra-high frequency method, transient earth voltage (TEV) method, ultrasonic detection method and other methods are usually used. However, the pulse current method has weak anti-electromagnetic interference ability, and its monitoring results are easily affected by electromagnetic interference. Ultrasonic detection method and ultra-high frequency method can effectively avoid the influence of electromagnetic interference, but the propagation characteristics of ultrasonic signals and ultra-high frequency electromagnetic wave signals inside the equipment are not easy to analyze, which makes it difficult to determine the fault monitoring results; TEV method detects partial discharge by capturing the instantaneous voltage signal to the ground generated on the surface of the equipment shell when the electromagnetic wave propagates in the air and the surface of the equipment shell, but it cannot quantitatively analyze the partial discharge amount and is also susceptible to external signal interference. In addition, the monitoring equipment used for partial discharge fault monitoring according to the above method is mostly expensive, ranging from hundreds of thousands to more than one million. In addition, the above monitoring method also requires long-term learning and experience accumulation of monitoring personnel to ensure a certain monitoring quality.

[0021] In case of cable overheating, distributed optical fiber temperature measurement technology, infrared imaging technology, HeatTag wireless sensors, etc. are usually used to monitor the cable temperature, thereby realizing the monitoring of cable overheating faults. However, most of the above temperature measurement methods require the use of professional temperature measurement equipment, which has high installation and maintenance costs, and the temperature measurement results are also affected by the installation location and usage, which makes it impossible to effectively guarantee the temperature measurement accuracy. In summary, in the related technologies, most of the monitoring methods for partial discharge faults and cable overheating faults have the problem of high monitoring costs and being easily affected by external interference, which results in the inability to ensure high monitoring accuracy.

[0022] In order to improve the fault monitoring accuracy of partial discharge faults and / or cable overheating faults and reduce the monitoring cost, in the implementation of the present application, the odor sensor data inside the device to be tested is obtained to determine the odor change inside the device to be tested, and the fault state inside the device to be tested is monitored according to the odor change. This is because the inventors found that each device in the power supply and distribution system usually produces different odors when a partial discharge fault or a cable overheating fault occurs. Therefore, the present application realizes the fault monitoring inside the device to be tested according to the odor change. On the one hand, the odor monitoring method will not be affected by external interference such as the electromagnetic environment, installation location, and experience of the monitoring personnel. On the other hand, each device in the power supply and distribution system, such as high and low voltage switch cabinets, cable wells, etc., is usually a semi-enclosed environment, which is convenient for accurately capturing the odor changes inside the equipment, thereby improving the fault monitoring accuracy. In addition, the odor-based fault monitoring method provided by the present application is lower in cost and more economically feasible. In addition, the fault monitoring method provided by the embodiment of the present invention can simultaneously realize the monitoring of partial discharge faults and / or cable overheating faults, without the need to monitor the two fault types separately, which can effectively simplify the fault monitoring method.

[0023] In order to make the purpose, technical solutions and advantages of the present invention more clear, specific embodiments will be described below in conjunction with the accompanying drawings.

[0024] Figure 1 The implementation flow chart of the odor-based fault monitoring method provided by the embodiment of the present invention is described in detail as follows: Step 101, obtaining at least one kind of odor sensing data inside the device to be tested.

[0025] Here, the equipment to be tested includes but is not limited to: high and low voltage switch cabinets, cable wells, cable interlayers, and high voltage inverters.

[0026] In the embodiment of the present invention, a semiconductor gas sensor, an olfactory sensor or an odor sensor may be used to collect odor sensing data. The odor sensing data may be used to characterize the odor condition inside the device to be tested.

[0027] Semiconductor gas sensors, olfactory sensors or odor sensors can respond to multiple gas types and volatile substances in a cross-interference manner to generate sensing data to achieve the purpose of identifying gases or odors of single or complex components, thereby reflecting the odor and gas conditions inside the device under test.

[0028] The above-mentioned sensor data can be called odor sensor data or gas sensor data. For the convenience of description, they are collectively referred to as odor sensor data below. It is understandable that when a certain sensor recognizes different types and concentrations of gases and / or volatile substances, the numerical value of the odor sensor data generated by it also changes accordingly. Depending on the type of sensor, the data representation form of the collected odor sensor data is different, and can usually be expressed in the form of voltage value, current value or resistance value.

[0029] In the embodiment of the present invention, a plurality of semiconductor gas sensors, olfactory sensors or odor sensors may be arranged inside the device under test to form a sensor array for collecting a variety of odor sensing data. Here, each sensor collects a type of odor sensing data. When selecting the sensor type, it may be determined based on the odor generated when the device under test is in a fault state.

[0030] For example, taking the high and low voltage switchgear as an example, when a partial discharge fault occurs inside the high and low voltage switchgear, a smell similar to the fishy smell will be generated, and when the local discharge intensity is high, a mixed charcoal smell will be generated. This fishy smell or mixed charcoal smell is mainly generated by ozone, carbon monoxide, hydrogen, nitrogen oxides and / or total volatile organic compounds emitted into the air. Accordingly, for partial discharge faults, semiconductor gas sensors, olfactory sensors or odor sensors that can respond to ozone, carbon monoxide, hydrogen, nitrogen oxides and / or total volatile organic compounds can be used.

[0031] When a cable overheats inside a high- and low-voltage switchgear, a rubber smell will be generated. When the cable continues to overheat, it will cause the insulation material to carbonize, and a burnt smell will be generated. This rubber smell is mainly caused by carbon monoxide, methane, carbon dioxide, nitrogen oxides, 2-ethylhexanol, dioctyl terephthalate, benzene, hydrogen chloride, carbon dioxide, alkyl and alkenyl aromatic products, unsaturated hydrogen, ozone, nitrogen oxides and / or aromatic hydrocarbon products emitted into the air. This burnt smell is mainly caused by ammonia, hydrogen, hydrogen sulfide and / or hydrocarbons emitted into the air.

[0032] Accordingly, for cable overheating faults, semiconductor gas sensors, olfactory sensors or odor sensors that can respond to carbon monoxide, methane, carbon dioxide, nitrogen oxides, 2-ethylhexanol, dioctyl terephthalate, benzene, hydrogen chloride, carbon dioxide, alkyl and alkenyl aromatic products, unsaturated hydrogen, ozone, nitrogen oxides, aromatic hydrocarbon products, ammonia, hydrogen, hydrogen sulfide and other products and / or hydrocarbons can be used.

[0033] In summary, the semiconductor gas sensor, olfactory sensor or odor sensor in the embodiments of the present invention can respond to at least one of ozone, carbon monoxide, hydrogen, nitrogen oxides, total volatile organic compounds, methane, carbon dioxide, 2-ethylhexanol, dioctyl terephthalate, benzene, hydrogen chloride, alkyl and alkenyl aromatic products, unsaturated hydrogen, aromatic hydrocarbon products, ammonia and hydrogen sulfide to generate odor sensing data.

[0034] For example, for partial discharge faults, five semiconductor gas sensors may be used to collect odor sensing data, wherein the five sensors respond to ozone, carbon monoxide, hydrogen, nitrogen oxides, and total volatile organic compounds, respectively, and generate five types of odor sensing data.

[0035] In addition, an embodiment of the present invention also provides a display interface for displaying the odor characteristic data corresponding to each sensor to the user. The display interface may include a characteristic data change trend graph and a characteristic data spectrum of the device to be tested. Among them, the horizontal axis of the characteristic data change trend graph is time, and the vertical axis can be the odor characteristic data, which is used to characterize the change trend of different types of odor characteristic data inside the device to be tested over time. The characteristic data spectrum mainly includes a line graph and a radar graph. Among them, the horizontal axis of the line graph is different types of sensors, and the vertical axis is the odor characteristic data, which is used to characterize the odor characteristic data corresponding to each sensor in the device to be tested at the current moment. The radar graph is used to characterize the odor characteristic data corresponding to each sensor at the current moment.

[0036] For example, see Figure 2 , taking the above five sensors as an example, when the device to be tested is a high and low voltage switch cabinet, the AH characteristic data change trend chart is the characteristic data change trend chart of the high and low voltage switch cabinet, the horizontal axis of the AH characteristic data change trend chart is time, and the vertical axis is the odor characteristic data. Through the AH characteristic data change trend chart, users can conveniently observe the change trend of the odor characteristic data corresponding to each sensor at different times. GS1, GS2, GS3, GS4 and GS5 are used to represent sensors that can respond to ozone, carbon monoxide, hydrogen, nitrogen oxides and total volatile organic compounds, respectively. The horizontal axis of the line graph in the characteristic data spectrum is different sensors, and the vertical axis is the odor characteristic data. The radar chart is used to characterize the odor characteristic data corresponding to GS1, GS2, GS3, GS4 and GS5.

[0037] Step 102, for each type of odor sensing data, extract the data change characteristics of the odor sensing data to obtain odor characteristic data corresponding to the odor sensing data; the odor characteristic data is used to characterize the odor change in the device under test.

[0038] The odor condition inside the device under test can affect the size of the odor sensor data. However, the size of the odor sensor data is not only affected by the odor condition, but also by other external interference factors. For example, the odor sensor data corresponding to the same odor environment at different temperatures is different. Therefore, the embodiment of the present invention does not directly perform fault monitoring based on the size of the odor sensor data, but adopts a method of extracting data change characteristics of the odor sensor data to determine the odor change situation inside the device under test, and performs fault monitoring based on the odor change situation to avoid the adverse effects of external interference factors.

[0039] Step 103, monitoring the fault status of the device under test according to the odor characteristic data corresponding to each type of odor sensing data.

[0040] The odor characteristic data corresponding to each type of odor sensing data can characterize the change of an odor. The embodiment of the present invention monitors the fault state inside the device to be tested according to the change of each type of odor. In some embodiments, the odor feature data corresponding to each type of odor sensing data may be input into each pre-trained fault monitoring model to obtain the fault monitoring results output by each fault monitoring model. Each fault monitoring model corresponds to a preset fault state. The fault monitoring model determines whether the device under test is in the corresponding preset fault state according to the odor characteristic data inside the device under test.

[0041] For example, based on the odor characteristic data under partial discharge fault, a fault monitoring model for partial discharge fault can be trained, and the fault monitoring model is used to determine whether a partial discharge fault occurs in the equipment according to the input odor characteristic data. Similarly, based on the odor characteristic data under cable overheat fault, a fault monitoring model for cable overheat fault can be trained, and the fault monitoring model is used to determine whether a cable overheat fault occurs in the equipment according to the input odor characteristic data.

[0042] It can be understood that fault monitoring of the equipment under test is a real-time online monitoring process. Accordingly, the odor characteristic data corresponding to each type of odor sensing data here may include: odor characteristic data corresponding to each type of odor sensing data at different times.

[0043] For each moment, the odor feature data corresponding to each odor sensor at that moment can be input into each fault monitoring model to determine whether the device under test at that moment is in a fault state. Alternatively, the odor feature data corresponding to each odor sensor at that moment and at different moments before that moment can be input into each fault monitoring model to determine whether the device under test at that moment is in a fault state.

[0044] Compared with the prior art, the embodiment of the present invention can determine the odor characteristic data by extracting the data change characteristics of the odor sensor data inside the device to be tested, and then monitor the fault state of the device to be tested based on the odor characteristic data. Here, the odor characteristic data can be used to characterize the odor changes inside the device to be tested. The embodiment of the present invention performs fault monitoring through the odor changes inside the device to be tested, which will not be affected by external interference such as the electromagnetic environment inside the device to be tested, the installation location, and the experience of the monitoring personnel, and can effectively improve the fault monitoring accuracy. In addition, the odor-based fault monitoring method provided by the present application has lower cost and is more economically feasible. In addition, the fault monitoring method provided by the embodiment of the present invention can simultaneously realize the monitoring of partial discharge faults and / or cable overheating faults, without the need to monitor the two fault types separately, which can effectively simplify the fault monitoring method.

[0045] According to the above embodiments, the present application mainly determines the odor characteristic data by extracting the data change characteristics of the odor sensor data, thereby realizing the fault monitoring of the device under test. The specific method for determining the odor characteristic data is introduced in detail below. Figure 3 ,When determining the odor characteristic data, the following steps are mainly included: Step 301, for each type of odor sensing data, obtain the time series data corresponding to the odor sensing data at the current moment. The time series data includes the odor sensing data corresponding to the current moment and different moments before the current moment.

[0046] In the embodiment of the present invention, each sensor disposed in the device to be tested collects odor sensing data in real time, and each sensor corresponds to one type of odor sensing data. For each type of sensor, the odor sensing data collected by the sensor at the current moment and at different moments before the current moment can be combined in chronological order to form time series data.

[0047] Step 302, determining the first data and the second data corresponding to the current moment.

[0048] Here, the first data can be determined by the odor sensing data within the first preset time interval in the time series data, and is used to characterize the odor situation in the device under test within the first preset time interval. The second data can be determined by the odor sensing data within the second preset time interval in the time series data, and is used to characterize the odor situation in the device under test within the second preset time interval.

[0049] For example, the average value or median of the odor sensor data within the first preset time interval in the time series data can be used as the first data to characterize the odor situation in the device to be tested to avoid noise interference. Similarly, the average value or median of the odor sensor data within the second preset time interval in the time series data can be used as the second data.

[0050] The first preset time interval is earlier than the second preset time interval, and the current moment is within the second preset time interval. In essence, the second data can be used to characterize the odor condition of the device under test at the current moment. The first data can be used to characterize the odor condition of the device under test before the current moment.

[0051] In the embodiment of the present invention, the first preset time interval and the second preset time interval can be determined by setting parameters such as the first preset time interval length RaSQ, the second preset time interval length RgSQ, and the first time interval interval RaSI. Here, the first time interval interval is used to represent the time difference between the end moment of the first preset time interval and the beginning moment of the second preset time interval.

[0052] Exemplarily, there is a 1 second difference between two adjacent moments, that is, the time difference between each data point in the time series data is 1 second. Taking the current moment as the 15th second as an example, the time series data corresponding to the current moment can be expressed as: R1, R2, R3, R4, R5, R6, R7, R8, R9, R10, R11, R12, R13, R14, R15. Among them, the odor sensor data corresponding to the current moment is R15. The first preset time interval length RaSQ and the second preset time interval length RgSQ can both be 6 seconds. The first time interval interval RaSI can be 1 second. Then the second preset time interval is the time interval formed from the 10th second to the 15th second, and the corresponding odor sensor data in the second preset time interval are R10, R11, R12, R13, R14, R15. The first preset time interval is the time interval formed from the 3rd second to the 8th second, and the corresponding odor sensing data within the first preset time interval are R3, R4, R5, R6, R7, and R8.

[0053] It can be understood that the first preset time interval length RaSQ and the second preset time interval length RgSQ are both integers greater than 1, and the first time interval interval RaSI is an integer greater than or equal to 0. The specific values ​​of the three can be determined according to actual conditions, and the embodiment of the present invention does not specifically limit this.

[0054] Step 303, determining the odor characteristic data corresponding to the odor sensing data at the current moment according to the ratio of the first data to the second data.

[0055] Here, the ratio of the first data to the second data substantially refers to the change in the odor situation at the current moment compared with the odor situation before the current moment, that is, the change in the odor at the current moment.

[0056] The embodiment of the present invention uses the ratio of the first data and the second data as odor characteristic data to monitor the fault state of the device under test at the current moment, thereby determining the fault state at the current moment based on the change in the odor at the current moment.

[0057] Similarly, during the operation of the device under test, the embodiment of the present invention can monitor the odor sensor data inside the device under test in real time online, and then determine the odor characteristic data at different times in real time, and based on the odor characteristic data at different times, determine the fault state of the device under test at the corresponding time.

[0058] According to the above embodiments, the embodiments of the present invention mainly determine the odor characteristic data based on the ratio of the first data to the second data, thereby realizing fault monitoring. The following focuses on how to determine the first data and the second data corresponding to the current moment.

[0059] In some embodiments, the second data corresponding to the current moment can be determined based on the average value of the odor sensor data within the second preset time interval in the time series data corresponding to the current moment.

[0060] Here, the average value of the odor sensor data within the second preset time interval in the time series data corresponding to the current moment can be directly determined as the second data corresponding to the current moment.

[0061] It should be noted that the second preset time interval here includes the current time, that is, the second preset time interval can be determined by taking the current time as the end time of the second preset time interval.

[0062] Still taking the current moment as the 15th second as an example, the time series data corresponding to the current moment can be expressed as: R1, R2, R3, R4, R5, R6, R7, R8, R9, R10, R11, R12, R13, R14, R15. Among them, the odor sensor data corresponding to the current moment is R15. The second preset time interval length RgSQ can be 6 seconds, that is, the second preset time interval is the time interval formed from the 10th second to the 15th second, and the corresponding odor sensor data in the second preset time interval is R10, R11, R12, R13, R14, R15. Correspondingly, the second data is the average value of R10, R11, R12, R13, R14, R15.

[0063] In some embodiments, when determining the first data corresponding to the current moment, the first data corresponding to the previous moment can be first determined as the first data corresponding to the current moment; then based on the ratio of the first data corresponding to the current moment and the second data, it is judged whether the odor of the equipment to be tested has reached a stable state. If so, the first data corresponding to the current moment is updated according to the time series data corresponding to the current moment; if not, the first data corresponding to the current moment is maintained.

[0064] Here, when judging whether the smell of the device under test has reached a stable state, two situations are mainly considered. One is that the smell inside the device under test has changed significantly, but after the significant change, it gradually dissipates and returns to the stable state before the significant change. The other is that the smell inside the device under test has never changed significantly, that is, it is always in a stable state. The stable state here can be understood as the state before the smell has changed significantly.

[0065] The following focuses on the first case. Considering that odor changes are usually a continuous process over a period of time, when the odor inside the device under test changes significantly, the degree of odor change can be determined based on the odor conditions when the odor begins to change and the odor conditions at the current moment, so as to better perform fault monitoring.

[0066] In order to accurately determine the degree of change of the odor, the first data is not updated in real time in the embodiment of the present invention. Instead, the first data corresponding to the previous moment is always determined as the first data corresponding to the current moment, and then the ratio of the first data corresponding to the current moment to the second data corresponding to the current moment is used to determine whether the odor in the device to be tested has reached a stable state.

[0067] In particular, for the starting moment, there is no previous moment. When executing the step of determining the first data corresponding to the previous moment as the first data corresponding to the current moment, the first data corresponding to the current moment can be directly determined based on the odor sensor data within the first preset time interval in the time series data corresponding to the starting moment, and then it can be judged whether the odor has reached a stable state.

[0068] Here, if the ratio of the first data corresponding to the current moment to the second data corresponding to the current moment does not exceed the first preset interval, it can be determined that the odor in the device to be tested has reached a stable state. If the ratio of the first data corresponding to the current moment to the second data corresponding to the current moment exceeds the first preset interval, it can be determined that the odor in the device to be tested has not reached a stable state. Here, the upper limit value of the first preset interval is greater than 1, and the lower limit value of the first preset interval is less than 1. Exemplarily, the first preset interval can be expressed as [1-D, 1+D]. Wherein D represents the deviation of the preset data. The value of the deviation of the preset data can be determined according to the actual situation, and the embodiment of the present invention is not specifically limited here. Exemplarily, the value of D can be 0.001, 0.002 or 0.003, etc.

[0069] If the odor in the device under test has not reached a stable state, it is necessary to keep the first data corresponding to the current moment, and then calculate the odor characteristic data to determine the degree of odor change. If the odor in the device under test reaches a stable state, the first data corresponding to the current moment can be updated according to the time series data corresponding to the current moment, so as to provide an odor reference basis for calculating the degree of odor change next time.

[0070] For the second case, if the odor in the device to be tested is always in a stable state, the first data can be updated in real time according to the time series data corresponding to the current moment, so as to provide an odor reference basis for the next odor change.

[0071] The above embodiments mainly consider two application scenarios, one is that the smell in the device under test does not change significantly, that is, it is always in a stable state. The other is that after the smell in the device under test changes significantly, it returns to the stable state before the smell changes.

[0072] Another application scenario is that after the smell in the device under test changes, it is impossible to restore to the stable state before the smell changes. If it takes a long time to restore to the stable state before the smell changes, it may be that the basic smell environment in the device under test has changed. To avoid misjudgment of this situation, the embodiment of the present invention introduces cumulative time to provide a new method for determining the first data corresponding to the current moment. Figure 4 , as detailed below: Step 401, based on the ratio of the first data and the second data corresponding to the current moment, determine whether the first condition and the second condition are met.

[0073] The first condition is that the smell of the device under test reaches a stable state, and the second condition is that the cumulative duration exceeds the preset reset duration. That is, when determining the first data, the embodiment of the present invention not only considers whether the smell reaches a stable state, but also considers whether the cumulative duration exceeds the preset reset duration.

[0074] According to the above embodiment, when the odor changes significantly, the embodiment of the present invention always determines the first data corresponding to the previous moment as the first data corresponding to the current moment, and keeps it continuously to determine the degree of change of the odor.

[0075] In the embodiment of the present invention, the cumulative duration is introduced to characterize the retention duration of the first data, that is, the duration from the moment when the smell begins to change to the current moment.

[0076] Step 402: If the first condition or the second condition is met, the first data corresponding to the current moment is updated according to the time series data corresponding to the current moment, and the accumulated duration is cleared.

[0077] According to the above embodiment, if the odor in the device to be tested reaches a stable state, the first data corresponding to the current moment is updated, thereby providing a reference basis for the next odor change.

[0078] If the second condition is met, that is, the duration from the moment when the smell begins to change to the current moment is too long, at this time, even if the smell in the device to be tested has not reached the stable state before the smell begins to change, the first data is still updated.

[0079] The reason for this design is that when the odor changes significantly and does not return to the stable state before the significant change for a long time, the basic odor environment inside the device to be tested may have changed. In order to avoid misjudgment due to changes in the basic odor environment, the embodiment of the present invention updates the first data when it is determined that the cumulative time exceeds the preset reset time, which is equivalent to determining the current odor situation as a new stable state. By calculating the first data in the new stable state, a reference odor basis is provided for the next odor change.

[0080] It can be understood that the accumulated time can be used to represent the retention time of the first data. Accordingly, after the first data is updated, the accumulated time needs to be cleared.

[0081] Step 403: If the first condition and the second condition are not satisfied, the first data corresponding to the current moment is retained.

[0082] On the contrary, if the odor in the device to be tested has not returned to a stable state and the accumulated time exceeds the preset reset time, the first data will not be updated, but the first data corresponding to the previous moment will continue to be used to determine the degree of odor change.

[0083] In addition to introducing the cumulative duration, the embodiment of the present invention also needs to update the cumulative duration in real time so that the entire process can be cyclically operated. To this end, for each type of odor sensing data, after determining the odor feature data corresponding to the odor sensing data at the current moment according to the ratio of the first data to the second data, the following is also included: For each type of odor sensing data, the cumulative duration corresponding to the odor sensing data is added to the unit duration to obtain an updated cumulative duration corresponding to the odor sensing data, wherein the unit duration is the time difference between two adjacent moments.

[0084] That is to say, for each type of odor sensing data at each moment, after calculating the odor characteristic data corresponding to that moment, it is necessary to update the cumulative duration according to the time difference between that moment and the next moment.

[0085] In addition, the accumulated time may be initialized at the start time, and the initialized accumulated time may be 0, so that the accumulated time may start to accumulate from 0.

[0086] The above embodiment mainly introduces the process of updating or maintaining the first data according to the first condition and the second condition to determine the first data corresponding to the current moment. The following is a detailed introduction to the method of updating the first data.

[0087] When updating the first data corresponding to the current moment according to the time series data corresponding to the current moment, it is possible to first detect whether the accumulated duration is greater than the preset threshold duration. The preset threshold duration is less than the preset reset duration. Here, the embodiment of the present invention does not specifically limit the values ​​of the preset threshold duration and the preset reset duration. Exemplarily, the preset threshold duration may be 1 second, 2 seconds, or 3 seconds, etc. The preset reset duration may be 300 seconds, 400 seconds, or 500 seconds, etc.

[0088] If the accumulated duration corresponding to the current moment is greater than the preset threshold duration, the second data corresponding to the current moment is determined as the updated first data corresponding to the current moment.

[0089] If the accumulated duration corresponding to the current moment is less than or equal to the preset threshold duration, the updated first data corresponding to the current moment is determined based on the average value of the odor sensor data within the first preset time interval in the time series data corresponding to the current moment.

[0090] Regarding the situation where the odor in the device to be tested has reached stability at the current moment, by detecting whether the cumulative time is greater than the preset threshold time, it can be determined whether the time interval between the moment when the odor begins to change and the current moment is too large. If the time interval is too large, it means that the odor has changed significantly and the change time has lasted for a long time. At this time, if the first data is determined directly based on the odor sensor data within the first preset time interval, the first preset time interval may happen to be the odor change interval, and the first data obtained by the updated calculation may not be able to accurately reflect the odor situation that has reached a stable state, resulting in misjudgment.

[0091] The embodiment of the present invention takes into account that after the odor reaches a stable state at the current moment, the first data and the second data are close. Therefore, the second data corresponding to the current moment is directly updated to the first data corresponding to the current moment, so that the updated first data can accurately reflect the odor situation after the odor reaches a stable state.

[0092] If the accumulated duration is less than or equal to the preset threshold duration, it means that the odor has changed for a short time and has returned to a stable state. It can be simply understood that the odor has not changed significantly. At this time, the first data corresponding to the current moment can be determined based on the odor sensing data within the first preset time interval corresponding to the current moment. The specific determination method is detailed in the above embodiment and will not be repeated here.

[0093] To improve the accuracy of fault monitoring, the preset threshold time length may be set as small as possible. For example, the preset threshold time length may be 1 second, 2 seconds or 3 seconds.

[0094] Taking 1 second as an example, when the accumulated duration is greater than 1, it means that the first data has been used for at least one unit of time, that is, the smell in the device to be tested has changed and has returned to a stable state before the change. At this time, in order to avoid misjudgment, the second data corresponding to the current moment is directly used to update the first data corresponding to the current moment.

[0095] When the accumulated duration is less than or equal to 1, it indicates that the odor in the device under test has not changed significantly, and the first data is in a real-time updating state. At this time, the first data corresponding to the current moment is still updated according to the time series data corresponding to the current moment.

[0096] In the case where the accumulated duration exceeds the preset reset duration, it is possible to directly determine that the accumulated duration exceeds the preset threshold duration, and update the first data corresponding to the current moment according to the second data corresponding to the current moment.

[0097] The above embodiment mainly introduces the case where the first data is updated when the first condition is met or the second condition is met. The following further introduces the case where the first condition is not met or the second condition is not met.

[0098] If the first condition and the second condition are not met, it proves that the odor in the device under test has changed significantly and has not yet returned to the stable state before the change began, and the accumulated duration has not exceeded the preset reset duration. At this time, in order to avoid the situation where the accumulated duration has reached the preset reset duration when the odor in the device under test is still in a continuous change, the embodiment of the present invention determines whether to reset the accumulated duration by detecting whether the odor situation at the current moment has reached a new stable state.

[0099] It is understandable that if the odor condition of the device to be tested is still in a state of continuous fluctuation at the current moment, in order to avoid the situation in which the cumulative duration exceeds the preset reset duration in this case, the cumulative duration can be forcibly cleared to avoid the situation in which the cumulative duration exceeds the preset reset duration and the first data is updated.

[0100] On the contrary, if the odor situation of the device under test is relatively stable at the current moment and no longer fluctuates, it proves that the odor situation in the device under test is already in a new stable state. At this time, there is no need to reset the cumulative time. When the cumulative time reaches the preset reset time, the first data can be directly updated to determine the odor situation in the new stable state, providing an odor reference basis for the next odor change.

[0101] Specifically, in some embodiments, if the first condition is not met and the second condition is not met, after maintaining the first data corresponding to the current moment, the third data and the fourth data corresponding to the current moment can also be determined based on the time series data corresponding to the current moment.

[0102] Here, the third data is determined by the odor sensing data within a third preset time interval in the time series data. The fourth data is determined by the odor sensing data within a fourth preset time interval in the time series data.

[0103] Here, the third data may be an average value of the odor sensing data within a third preset time interval in the time series data. The fourth data may be an average value of the odor sensing data within a fourth preset time interval in the time series data.

[0104] Among them, the current moment is within the third preset time interval, and the fourth preset time interval is earlier than the third preset time interval. In essence, the third data can be used to characterize the odor situation at the current moment. The fourth data can characterize the odor situation before the current moment. The embodiment of the present invention determines whether the odor change situation at the current moment tends to be stable based on the odor situation at the current moment and the odor situation between the current moments.

[0105] If the ratio of the third data to the fourth data exceeds the second preset interval, it proves that the odor change at the current moment has not stabilized and is still in continuous fluctuation. At this time, the cumulative duration will be cleared to zero to avoid the situation where the cumulative duration exceeds the preset reset duration during the continuous fluctuation of the odor, thereby updating the first data.

[0106] If the ratio of the third data to the fourth data does not exceed the second preset interval, it proves that the odor change at the current moment tends to be stable. At this time, there is no need to reset the cumulative time. When the cumulative time reaches the preset reset time, the first data can be directly updated to determine the odor situation under the new stable state, providing an odor reference basis for the next odor change.

[0107] Here, the upper limit value of the second preset interval is greater than 1, and the lower limit value is less than 1.

[0108] Similar to the first preset time interval and the second preset time interval, the embodiment of the present invention can determine the third preset time interval and the fourth preset time interval by setting parameters such as the third preset time interval length RgCVSQ, the fourth preset time interval length RgBVSQ and the second time interval interval RgSI.

[0109] It is understandable that the third preset time interval length RgCVSQ and the fourth preset time interval length RgBVSQ are both integers greater than 1, and the second time interval interval RgSI is an integer greater than or equal to 0. The specific values ​​of the three can be determined according to actual conditions, and the embodiment of the present invention does not specifically limit this.

[0110] Still taking the time series data from 1 to 15 seconds as an example, the time series data corresponding to the 15th second can be expressed as: R1, R2, R3, R4, R5, R6, R7, R8, R9, R10, R11, R12, R13, R14, R15. Among them, the odor sensor data corresponding to the current moment is R15. The third preset time interval length RgCVSQ and the fourth preset time interval length RgBVSQ can both be 3 seconds. The second time interval interval RgSI can be 1 second. Then the third preset time interval is the time interval formed from the 13th second to the 15th second, and the corresponding odor sensor data in the fourth preset time interval are R13, R14, and R15. The fourth preset time interval is the time interval formed from the 9th second to the 11th second. The corresponding odor sensor data in the third preset time interval are R9, R10, and R11.

[0111] Combine the following Figure 5 , a general introduction is given to the process of determining the first data and the second data corresponding to the current moment.

[0112] First, the system is initialized, mainly initializing the first data (Ra), the second data (Rg) and the accumulated time (RaLockCount), and setting the data collection frequency of the sensor accordingly. Here, the data collection frequency can determine the time difference between two adjacent moments.

[0113] Secondly, the first data (RaLast) corresponding to the previous moment is determined as the first data (Ra) corresponding to the current moment. Determine whether the first data (Ra) corresponding to the current moment is 0. If Ra= =0, it means that the current moment is the starting moment. At this time, the average value of the odor sensor data in the first preset time interval in the time series data corresponding to the current moment is determined as the first data corresponding to the current moment. The formula is expressed as: Ra=Avg(R[0-RaSQ-RgSQ-RaSI],RaSQ). Here, R represents the time series data corresponding to the current moment. If the first data (Ra) corresponding to the current moment is not 0, it proves that the current moment is not the starting moment. At this time, directly jump to the step of calculating the second data corresponding to the current moment.

[0114] Next, the second data corresponding to the current moment is determined according to the average value of the odor sensor data in the second preset time interval in the time series data corresponding to the current moment. The public expression is: Rg=Avg(R[0-RgSQ], RgSQ).

[0115] Then, it is determined whether the accumulated time (RaLockCount) exceeds the preset reset time (RaResetCount) and whether the ratio of the first data to the second data corresponding to the current moment exceeds the first preset interval. In the embodiment of the present invention, the first preset interval can be expressed as [1-D, 1+D].

[0116] One situation is: if the cumulative duration exceeds the preset reset duration, or the ratio of the first data and the second data corresponding to the current moment does not exceed the first preset interval, that is, RaLockCount>RaResetCount, or 1-D<= Ra÷Rg<= 1+ D, then it is further determined whether the cumulative duration is greater than the preset threshold duration. In this embodiment of the present invention, the value of the preset threshold duration is 1.

[0117] If the accumulated duration is greater than the preset threshold duration, the second data corresponding to the current moment is updated to the first data corresponding to the current moment.

[0118] If the accumulated duration is less than or equal to the preset threshold duration, the average value of the odor sensor data in the first preset time interval corresponding to the current moment is updated to the first data corresponding to the current moment.

[0119] After the first data (Ra) is updated, the accumulated time (RaLockCount) is cleared, and the odor characteristic data (RaF) corresponding to the current moment is determined based on the ratio of the updated first data and the second data corresponding to the current moment.

[0120] Another situation is: the accumulated duration does not exceed the preset reset duration, and the ratio of the first data and the second data corresponding to the current moment exceeds the first preset interval. At this time, the first data corresponding to the current moment still uses the first data corresponding to the previous moment.

[0121] Then, the third data (RgCV) corresponding to the current moment is determined according to the average value of the odor sensing data in the third preset time interval in the time series data corresponding to the current moment. The public expression is: RgCV=Avg(R[0-RgCVSQ],RgCVSQ).

[0122] According to the average value of the odor sensor data in the fourth preset time interval in the time series data corresponding to the current moment, the fourth data (RgBV) corresponding to the current moment is determined. The public expression is: RgBV=Avg(R[0-RgCVSQ-RgBVSQ-RgSI],RgBVSQ).

[0123] Next, the ratio of the third data to the fourth data is calculated to determine whether the odor situation in the device to be tested is stable at the current moment.

[0124] If the ratio of the third data to the fourth data exceeds the second preset interval, the cumulative time clearing step is performed, and the odor characteristic data is calculated based on the first data locked and used and the second data corresponding to the current moment. If the ratio of the third data to the fourth data does not exceed the second preset interval, the cumulative time clearing step is not performed, and the odor characteristic data is directly calculated based on the first data locked and used and the second data corresponding to the current moment. Here, the second preset interval can be expressed as [1-D, 1+D].

[0125] Finally, after determining the odor characteristic data (RaF) at the current moment, the accumulated time (RaLockCount) is updated, and the step of re-jumping to determine the first data corresponding to the current moment as the first data corresponding to the current moment is continued to calculate the odor characteristic data corresponding to the next moment.

[0126] Furthermore, in combination with the specific values ​​of each parameter, a specific example is provided to illustrate the method for determining odor characteristic data. In this example, the data collection frequency is 1 second. Assuming that the current moment is the 15th second, the time series data corresponding to the current moment can be expressed as: R1, R2, R3, R4, R5, R6, R7, R8, R9, R10, R11, R12, R13, R14, R15. The cumulative duration of the current moment is 1. The length of the first preset time interval and the length of the second preset time interval are both 6 seconds, the length of the third preset time interval and the length of the fourth preset time interval are both 3 seconds, and the interval of the first time interval and the interval of the second time interval are both 1 second. The first preset interval and the second preset interval are both [0.998, 1.002]. The value of the preset threshold duration is 1 second. The value of the preset reset duration is 300 seconds.

[0127] First, the first data corresponding to the 14th second is determined as the first data (Ra) corresponding to the 15th second.

[0128] Next, the average value of the odor sensor data in the second preset time interval (RgSQ) in the time series data corresponding to the 15th second is determined as the second data (Rg) corresponding to the current moment. That is, the average value of R10, R11, R12, R13, R14, and R15 is determined as the second data (Rg) corresponding to the 15th second.

[0129] Then, it is determined whether the accumulated time (RaLockCount) exceeds the preset reset time (RaResetCount) and whether the ratio of the first data (Ra) to the second data (Rg) corresponding to the current moment exceeds the first preset interval [1-D, 1+D].

[0130] At this point, the accumulated duration has not exceeded the preset reset duration.

[0131] One case is: if the ratio of the first data (Ra) to the second data (Rg) corresponding to the current moment is within [0.998, 1.002], it is further determined whether the accumulated duration (RaLockCount) is greater than 1.

[0132] At this time, the accumulated duration (RaLockCount) is equal to 1, and the first data (Ra) is updated according to the average value of the odor sensor data within the first preset time interval (RaSQ), that is, the average value of R3, R4, R5, R6, R7, and R8 is updated to the first data corresponding to the current moment.

[0133] After the first data (Ra) is updated, the accumulated time (RaLockCount) is cleared, and the odor characteristic data (RaF) corresponding to the current moment is determined based on the ratio of the updated first data and the second data corresponding to the current moment.

[0134] Another situation is: the ratio of the first data and the second data corresponding to the current moment exceeds [0.998, 1.002]. At this time, the first data corresponding to the 15th second still uses the first data corresponding to the 14th second.

[0135] Subsequently, according to the average value of the odor sensor data in the third preset time interval (RgCVSQ) in the time series data corresponding to the 15th second, the third data (RgCV) corresponding to the current moment is determined, that is, the average value of R13, R14, and R15 is determined as the third data (RgCV).

[0136] According to the average value of the odor sensing data in the fourth preset time interval (RgBVSQ) in the time series data corresponding to the current moment, that is, the average value of R9, R10, and R11 is determined as the fourth data (RgBV).

[0137] Next, determine whether the ratio of the third data (RgCV) to the fourth data (RgBV) exceeds [0.998, 1.002]. If so, execute the cumulative duration (RaLockCount) clearing step, and calculate the odor characteristic data (RaF) based on the first data at the 14th second and the second data corresponding to the 15th second.

[0138] If the ratio of the third data to the fourth data does not exceed [0.998, 1.002], the cumulative duration clearing step will not be performed, and the odor characteristic data (RaF) will be calculated directly based on the first data at the 14th second and the second data corresponding to the 15th second.

[0139] Finally, after determining the odor characteristic data at the 15th second, the cumulative duration (RaLockCount) is updated, and the step of re-jumping to determine the first data corresponding to the previous moment as the first data corresponding to the current moment is continued to calculate the odor characteristic data corresponding to the 16th second.

[0140] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0141] The following is an embodiment of the device of the present invention. For details not described in detail therein, reference may be made to the corresponding method embodiment described above.

[0142] Figure 6 The structure diagram of the odor-based fault monitoring device provided by an embodiment of the present invention is shown. For the convenience of description, only the part related to the embodiment of the present invention is shown, which is described in detail as follows: like Figure 6As shown, the odor-based fault monitoring device 6 includes: an acquisition module 61 , an extraction module 62 and a monitoring module 63 .

[0143] An acquisition module 61 is used to acquire at least one kind of odor sensing data inside the device to be tested; The extraction module 62 is used to extract the data change characteristics of each type of odor sensing data, and obtain the odor characteristic data corresponding to the odor sensing data; the odor characteristic data is used to characterize the odor change in the device under test; The monitoring module 63 is used to monitor the fault status of the device under test according to the odor characteristic data corresponding to each type of odor sensing data.

[0144] Optionally, the odor characteristic data corresponding to each type of odor sensing data includes: odor characteristic data corresponding to each type of odor sensing data at different moments; The extraction module 62 is specifically used for: For each type of odor sensing data, obtain the time series data corresponding to the odor sensing data at the current moment; the time series data includes the odor sensing data corresponding to the current moment and different moments before the current moment; Determine the first data and the second data corresponding to the current moment; the first data is determined by the odor sensor data within a first preset time interval in the time series data, and the second data is determined by the odor sensor data within a second preset time interval in the time series data; the first preset time interval is earlier than the second preset time interval; the current moment is within the second preset time interval; According to the ratio of the first data to the second data, the odor characteristic data corresponding to the odor sensing data at the current moment is determined.

[0145] Optionally, the extraction module 62 is specifically used for: Determine the second data corresponding to the current moment according to the average value of the odor sensor data within the second preset time interval in the time series data corresponding to the current moment; The extraction module 62 is specifically used for: Determine the first data corresponding to the previous moment as the first data corresponding to the current moment; Based on the ratio of the first data and the second data corresponding to the current moment, determine whether the odor of the device to be tested has reached a stable state. If so, update the first data corresponding to the current moment according to the time series data corresponding to the current moment; if not, keep the first data corresponding to the current moment.

[0146] Optionally, the extraction module 62 is specifically used for: Based on the ratio of the first data and the second data corresponding to the current moment, determine whether the first condition and the second condition are met, wherein the first condition is that the odor of the device to be tested reaches a stable state, and the second condition is that the accumulated time exceeds the preset reset time; If the first condition is met or the second condition is met, the first data corresponding to the current moment is updated according to the time series data corresponding to the current moment, and the accumulated duration is cleared; If the first condition is not met and the second condition is not met, the first data corresponding to the current moment is retained; The extraction module 62 is further used for: For each type of odor sensing data, the cumulative duration corresponding to the odor sensing data is added to the unit duration to obtain an updated cumulative duration corresponding to the odor sensing data, wherein the unit duration is the time difference between two adjacent moments.

[0147] Optionally, the extraction module 62 is specifically used for: Check whether the accumulated duration is greater than the preset threshold duration; the preset threshold duration is less than the preset reset duration; If the accumulated duration corresponding to the current moment is greater than the preset threshold duration, the second data corresponding to the current moment is determined as the updated first data corresponding to the current moment; If the accumulated duration corresponding to the current moment is less than or equal to the preset threshold duration, the updated first data corresponding to the current moment is determined based on the average value of the odor sensor data within the first preset time interval in the time series data corresponding to the current moment.

[0148] Optionally, the extraction module 62 is specifically used for: Detecting whether the ratio of the first data and the second data corresponding to the current moment exceeds a first preset interval; If the ratio of the first data to the second data does not exceed the first preset interval, it is determined that the first condition is met; If the ratio of the first data to the second data exceeds the first preset interval, it is determined that the first condition is not met; the upper limit value of the first preset interval is greater than 1, and the lower limit value of the first preset interval is less than 1.

[0149] Optionally, the extraction module 62 is further used for: According to the time series data corresponding to the current moment, the third data and the fourth data corresponding to the current moment are determined; the third data is determined by the odor sensor data within the third preset time interval in the time series data, and the fourth data is determined by the odor sensor data within the fourth preset time interval in the time series data; the fourth preset time interval is earlier than the third preset time interval; the current moment is within the third preset time interval; If the ratio of the third data to the fourth data exceeds the second preset interval, the accumulated duration is reset to zero; the upper limit value of the second preset interval is greater than 1, and the lower limit value is less than 1.

[0150] Optionally, the monitoring module 63 is specifically used for: The odor feature data corresponding to each type of odor sensing data is input into each pre-trained fault monitoring model to obtain the fault monitoring results output by each fault monitoring model; each fault monitoring model corresponds to a preset fault state; the fault monitoring model determines whether the device is in the corresponding preset fault state based on the odor feature data inside the device.

[0151] Optionally, the odor sensing data is collected by a semiconductor gas sensor, an olfactory sensor or an odor sensor; the semiconductor gas sensor, the olfactory sensor or the odor sensor responds to at least one of ozone, carbon monoxide, hydrogen, nitrogen oxides, total volatile organic compounds, methane, carbon dioxide, 2-ethylhexanol, dioctyl terephthalate, benzene, hydrogen chloride, alkyl and alkenyl aromatic products, unsaturated hydrogen, aromatic hydrocarbon products, ammonia and hydrogen sulfide.

[0152] The real-time example of the device can be used to implement the above method embodiment, and the implementation effect of its technical principle is the same as that of the above method embodiment, which will not be repeated here.

[0153] Figure 7 Schematic diagram of an electronic device provided by an embodiment of the present invention. Figure 7 As shown, the electronic device 7 of this embodiment includes: a processor 70 and a memory 71. The memory 71 stores a computer program 72. When the processor 70 executes the computer program 72, the steps in the above-mentioned method embodiments are implemented. Alternatively, when the processor 70 executes the computer program 72, the functions of each module / unit in the above-mentioned device embodiments are implemented.

[0154] Exemplarily, the computer program 72 may be divided into one or more modules / units, which are stored in the memory 71 and executed by the processor 70 to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, which are used to describe the execution process of the computer program 72 in the electronic device 7.

[0155] The electronic device 7 may include, but is not limited to, a processor 70 and a memory 71. Those skilled in the art will appreciate that Figure 7 It is only an example of the electronic device 7 and does not constitute a limitation of the electronic device 7. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the electronic device 7 may also include input and output devices, network access devices, buses, etc.

[0156] The processor 70 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc.

[0157] The memory 71 may be an internal storage unit of the electronic device 7, such as a hard disk or memory of the electronic device 7. The memory 71 may also be an external storage device of the electronic device 7, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the electronic device 7. Further, the memory 71 may also include both an internal storage unit and an external storage device of the electronic device 7. The memory 71 is used to store the computer program 72 and other programs and data required by the electronic device 7. The memory 71 may also be used to temporarily store data that has been output or is to be output.

[0158] For the convenience and simplicity of description, only the division of the above functional modules / units is used as an example for illustration. In actual applications, the above functions can be assigned to different functional modules / units as needed. The above modules / units can be implemented in the form of hardware, software, or a combination of hardware and software.

[0159] The embodiment of the present invention further provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the methods in the above method embodiments are implemented.

[0160] The embodiment of the present invention further provides a computer program product, including a computer program. When the computer program is executed by a processor, the methods in the above method embodiments are implemented.

[0161] The computer program includes computer program code, which may be in source code form, object code form, executable file or some intermediate form, etc. Computer readable media may include: any entity or device capable of carrying computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc.

[0162] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments. If there is no special explanation and logical conflict, the terms and / or descriptions between different embodiments are consistent and can be referenced to each other. The technical features in different embodiments can be combined to form a new embodiment according to their internal logical relationship.

[0163] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A fault monitoring method based on smell, characterized in that: include: Acquiring at least one type of odor sensing data inside the device under test; For each type of odor sensing data, extract the data change characteristics of the odor sensing data to obtain odor characteristic data corresponding to the odor sensing data; the odor characteristic data is used to characterize the odor change in the device to be tested; The fault status of the device under test is monitored according to the odor characteristic data corresponding to each type of odor sensing data.

2. The odor-based fault monitoring method according to claim 1, characterized in that: The odor characteristic data corresponding to each type of odor sensing data includes: odor characteristic data corresponding to each type of odor sensing data at different times; For each type of odor sensing data, extract the data change characteristics of the odor sensing data to obtain odor feature data corresponding to the odor sensing data, including: For each type of odor sensing data, obtain the time series data corresponding to the odor sensing data at the current moment; the time series data includes the odor sensing data corresponding to the current moment and different moments before the current moment; Determine first data and second data corresponding to the current moment; the first data is determined by odor sensor data within a first preset time interval in the time series data, and the second data is determined by odor sensor data within a second preset time interval in the time series data; the first preset time interval is earlier than the second preset time interval; the current moment is within the second preset time interval; According to the ratio of the first data to the second data, the odor characteristic data corresponding to the odor sensing data at the current moment is determined.

3. The odor-based fault monitoring method according to claim 2, characterized in that: The determining of the second data corresponding to the current moment includes: Determine the second data corresponding to the current moment according to the average value of the odor sensor data within the second preset time interval in the time series data corresponding to the current moment; The determining of the first data corresponding to the current moment includes: Determine the first data corresponding to the previous moment as the first data corresponding to the current moment; Based on the ratio of the first data and the second data corresponding to the current moment, determine whether the odor of the device to be tested has reached a stable state. If so, update the first data corresponding to the current moment according to the time series data corresponding to the current moment; if not, keep the first data corresponding to the current moment.

4. The odor-based fault monitoring method according to claim 3, characterized in that: Based on the ratio of the first data and the second data corresponding to the current moment, judging whether the odor of the device to be tested has reached a stable state, and if so, updating the first data corresponding to the current moment according to the time series data corresponding to the current moment; If not, keep the first data corresponding to the current time, including: Based on the ratio of the first data and the second data corresponding to the current moment, judging whether the first condition and the second condition are met, wherein the first condition is that the odor of the device to be tested reaches a stable state, and the second condition is that the accumulated duration exceeds the preset reset duration; If the first condition is met or the second condition is met, the first data corresponding to the current moment is updated according to the time series data corresponding to the current moment, and the accumulated duration is cleared; If the first condition is not met and the second condition is not met, the first data corresponding to the current moment is retained; For each type of odor sensing data, after determining the odor characteristic data corresponding to the type of odor sensing data at the current moment according to the ratio of the first data to the second data, the method further includes: For each type of odor sensing data, the cumulative duration corresponding to the odor sensing data is added to the unit duration to obtain an updated cumulative duration corresponding to the odor sensing data, wherein the unit duration is the time difference between two adjacent moments.

5. The odor-based fault monitoring method according to claim 4, characterized in that: Updating first data corresponding to the current moment according to the time series data corresponding to the current moment includes: Detecting whether the accumulated duration is greater than a preset threshold duration; the preset threshold duration is less than the preset reset duration; If the accumulated duration corresponding to the current moment is greater than the preset threshold duration, the second data corresponding to the current moment is determined as the updated first data corresponding to the current moment; If the accumulated duration corresponding to the current moment is less than or equal to the preset threshold duration, the updated first data corresponding to the current moment is determined based on the average value of the odor sensor data within the first preset time interval in the time series data corresponding to the current moment.

6. The odor-based fault monitoring method according to claim 4 or 5, characterized in that: Judging whether the first condition is met based on the ratio of the first data and the second data corresponding to the current moment includes: Detecting whether the ratio of the first data and the second data corresponding to the current moment exceeds a first preset interval; If the ratio of the first data to the second data does not exceed a first preset interval, it is determined that the first condition is met; If the ratio of the first data to the second data exceeds a first preset interval, it is determined that the first condition is not met; the upper limit value of the first preset interval is greater than 1, and the lower limit value of the first preset interval is less than 1.

7. The odor-based fault monitoring method according to claim 4 or 5, characterized in that: If the first condition is not satisfied and the second condition is not satisfied, then after the first data corresponding to the current moment is kept, the method further includes: According to the time series data corresponding to the current moment, third data and fourth data corresponding to the current moment are determined; the third data is determined by the odor sensor data within a third preset time interval in the time series data, and the fourth data is determined by the odor sensor data within a fourth preset time interval in the time series data; the fourth preset time interval is earlier than the third preset time interval; the current moment is within the third preset time interval; If the ratio of the third data to the fourth data exceeds a second preset interval, the accumulated duration is cleared; the upper limit value of the second preset interval is greater than 1, and the lower limit value is less than 1.

8. The odor-based fault monitoring method according to any one of claims 1 to 5, characterized in that: The step of monitoring the fault state of the device under test according to the odor characteristic data corresponding to each type of odor sensing data includes: The odor feature data corresponding to each type of odor sensing data is input into each pre-trained fault monitoring model to obtain the fault monitoring result output by each fault monitoring model; each fault monitoring model corresponds to a preset fault state; the fault monitoring model determines whether the device is in the corresponding preset fault state based on the odor feature data inside the device.

9. The odor-based fault monitoring method according to any one of claims 1 to 5, characterized in that: The odor sensing data is collected by a semiconductor gas sensor, an olfactory sensor or an odor sensor; the semiconductor gas sensor, the olfactory sensor or the odor sensor responds to at least one of ozone, carbon monoxide, hydrogen, nitrogen oxides, total volatile organic compounds, methane, carbon dioxide, 2-ethylhexanol, dioctyl terephthalate, benzene, hydrogen chloride, alkyl and alkenyl aromatic products, unsaturated hydrogen, aromatic hydrocarbon products, ammonia and hydrogen sulfide.

10. A fault monitoring device based on smell, characterized in that: include: An acquisition module, used to acquire at least one kind of odor sensing data inside the device to be tested; An extraction module is used to extract data change characteristics of each type of odor sensing data, and obtain odor characteristic data corresponding to the odor sensing data; the odor characteristic data is used to characterize the odor change in the device to be tested; The monitoring module is used to monitor the fault status of the device under test according to the odor characteristic data corresponding to each type of odor sensing data.

11. An electronic device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, and the processor implements the method according to any one of claims 1 to 9 when executing the computer program.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 9 is implemented.

13. A computer program product, characterized in that The method comprises a computer program, which implements the method according to any one of claims 1 to 9 when being executed by a processor.

Images