Method for controlling a device based on failure prediction, device

Abstract

The application provides a device control method and device based on fault prediction. The device control method based on fault prediction comprises the following steps: predicting the fault probability of the device; if the fault probability of the current device is within the interval range of [safe threshold, dangerous threshold], calculating the average time before failure of each component of the device under the current configuration; selecting the minimum value MTTF1 in the average time before failure, and recording the configuration of each component at this time; adjusting the configuration of each component respectively, recording the configuration of each component and the minimum value of the average time before failure corresponding to the configuration when the operation state of the device is improved or maintained, until the fault probability of the current device is less than the safe threshold; selecting the maximum value from the recorded minimum values of the average time before failure, and adjusting the configuration of each component of the device to the configuration of each component corresponding to the maximum value. The application can ensure the stability of the device operation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of device control, and in particular to a device control method based on fault prediction, which can control the device at a relatively low fault or for a longer stable time. BACKGROUND

[0002] The stability of device operation is an important indicator of good quality of the device.

[0003] In the prior art, various device fault prediction methods are given in order to avoid the sudden shutdown of subsequent devices due to faults, which affects the stability of device operation. However, these prior arts only have prediction, and no further improvement measures. In fact, the parts of the device will age during use. When the device has a high probability of failure, if measures are not taken in time, the device may not meet the user's needs, or a large amount of energy may be consumed.

[0004] Taking an air conditioner as an example, the air conditioner is a product with high stability requirements. Once the air conditioner cannot operate stably, it will have a great impact on the indoor temperature for people to work and study. In order to ensure the stability of the air conditioner operation, various technical solutions have been developed in the prior art.

[0005] For example, the prior art with publication number CN116163914A discloses an air conditioner compressor early warning scheme, which specifically includes the following steps.

[0006] Step S101: Obtain an air conditioner working condition signal. The air conditioner working condition signal can include: an acceleration signal A5(t) of the active end of the compressor foot pad and an acceleration signal A6(t) of the passive end.

[0007] Step S102: Fourier transform is performed on the acceleration signal of the active end of the compressor foot pad and the acceleration signal of the passive end, respectively.

[0008] Step S103: Based on the Fourier transformed acceleration signal of the active end of the compressor foot pad and the acceleration signal of the passive end, obtain the actual vibration attenuation amount of the compressor foot pad.

[0009] Step S104: Obtain a simulation vibration attenuation amount output by a simulation model based on the acceleration detection points of the active end and the passive end of the compressor foot pad.

[0010] Step S105: Compare the actual vibration attenuation amount with the simulation vibration attenuation amount, and calculate the error rate of the actual vibration attenuation amount and the simulation vibration attenuation amount.

[0011] Step S106: When the error rate is within the preset error range, and the actual vibration attenuation amount is lower than the preset attenuation threshold, calculating an external load spectrum received by the compressor foot pad based on the acceleration signal of the active end of the compressor foot pad and the acceleration signal of the passive end.

[0012] Step S107: Using a preset fatigue analysis model, performing risk analysis on the compressor foot pad based on the external load spectrum, and outputting an analysis result.

[0013] Although the prior art analyzes the risk of a fault of a certain component of an air conditioner based on certain vibration data of the air conditioner, no further solution is provided. However, the unit inevitably ages during operation, resulting in a decrease in maximum load and optimal operating state of the unit. When a fault occurs in a certain unit in the system, resulting in a decrease in output power, the user demand cannot be met in time before the fault unit is replaced or repaired. Even if the operating frequency of other units is increased, the service life of the unit may be reduced because the current load of the unit exceeds the actual maximum load, or the operating energy efficiency of the unit is not optimal, resulting in energy waste.

[0014] Therefore, how to provide a device that can be adjusted in time based on fault data to meet the current use demand and prolong the stable operation time is a technical problem to be solved. SUMMARY

[0015] To solve the technical problem of lack of further application of fault data to ensure the stability of the device in the prior art, a device control method and device based on fault prediction are provided.

[0016] The device control method based on fault prediction provided by the present application comprises:

[0017] predicting a fault probability of the device;

[0018] if the fault probability of the current device is within the interval range of [safe threshold, dangerous threshold], calculating the average fault-free time of each component of the device under the current configuration;

[0019] selecting the minimum value MTTF1 in the average fault-free time, and recording the configuration of each component at this time;

[0020] adjusting the configuration of each component respectively, and when the operating state of the device improves or maintains, recording the configuration of each component and the minimum value of the average fault-free time at this time, until the fault probability of the current device is less than the safe threshold;

[0021] selecting the maximum value from the recorded minimum values of the average fault-free time, and adjusting the configuration of each component of the device to the configuration of each component corresponding to the maximum value.

[0022] Further, the average time-to-failure of the component is obtained from a probability density function of the reliability of the component.

[0023] Further, the probability density function of the reliability of the component is obtained from a reliability data curve of the component at the time of factory shipment.

[0024] Further, the probability density function of the reliability of the component is obtained from a formula The calculation is obtained, R(t) is the function of the reliability data curve, and t is the working time of the component.

[0025] Further, the average time-to-failure is obtained by a formula

[0026] Further, the failure probability of the device includes:

[0027] Obtain the vibration data and / or temperature rise change data of each component of the device;

[0028] Draw the vibration data and / or temperature rise data into a vibration curve and a temperature rise rate curve, respectively;

[0029] Based on the change trend of the vibration curve and / or the temperature rise rate curve, the failure probability of the device is predicted.

[0030] Further, the running state of the device is improved or maintained, including:

[0031] The adjusted output power meets the user demand, the failure probability is less than or equal to the danger threshold, and the minimum value of the average time-to-failure of all components is greater than or equal to MTTF1.

[0032] Further, the running state of the device is improved or maintained, including:

[0033] The adjusted output power meets the user demand, the failure probability is less than or equal to the safety threshold, and the minimum value of the average time-to-failure of all components is greater than or equal to MTTF1.

[0034] The device provided by the application comprises components and control modules for controlling the components, and the control modules control the components by using the device control method based on failure prediction.

[0035] Further, the device comprises an air conditioner.

[0036] The application is based on the prediction of the failure probability of the device and the calculation of the average time-to-failure of the components, and finds the configuration that can make the device run stably for the longest time by using an optimization algorithm, and dynamically updates to ensure the stability of the device. ​BRIEF DESCRIPTION OF DRAWINGS

[0037] The application will be described in further detail below with reference to the drawings, in which:

[0038] Figure 1 is the main flow chart of the application.

[0039] Figure 2 is the detailed flow chart of an embodiment of the application.

[0040] Figure 3 is the reliability table of the components of an embodiment of the application.

[0041] Figure 4 is Figure 3 the corresponding reliability curve.

[0042] Figure 5 is the module scheduling flow chart of an embodiment of the application. DETAILED DESCRIPTION

[0043] In order to make the technical problems to be solved by the application, technical solutions and beneficial effects clearer, the application will be further described in detail below in combination with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the application and not to limit the application.

[0044] Therefore, one feature indicated in the specification will be used to explain one feature of an embodiment of the application, and it is not implied that each embodiment of the application must have the explained feature. In addition, it should be noted that the specification describes many features. Although certain features can be combined together to show possible system designs, these features can also be used in other combinations that are not explicitly described. Therefore, unless otherwise stated, the described combinations are not intended to be limiting.

[0045] As Figure 1 shown, in order to solve the problem that the prior art only predicts faults and does not further process the case of the predicted faults, the application proposes a device control method based on fault prediction. In a basic embodiment, the device control method of the application mainly includes the following steps.

[0046] Step 1, predicting the failure probability of the device;

[0047] There are many different methods for predicting the failure probability of different devices in the prior art. The application will not be described in detail, and those skilled in the art can select the corresponding device failure probability prediction method according to the type of the specific device.

[0048] Step 2, if the failure probability of the current device is in the range of [safe threshold, dangerous threshold], calculate the average failure-free time of each component of the device under the current configuration.

[0049] The average failure-free time is also known as MTTF (mean time to failure), which can be expressed by a simple formula: MTTF = total running time / failure times. Through this formula, the average failure-free time can be understood. This is only one way to calculate the average failure-free time, and other ways can also be used to calculate the average failure-free time. Each device can calculate the average failure-free time of all components, which will result in a large amount of calculation, and the finer the division of components, the larger the amount of calculation. Therefore, those skilled in the art can further screen according to the actual situation of the device, select the components that are more prone to failure, and calculate their average failure-free time, which can also achieve the purpose of the present application, while achieving the effect of reducing the amount of calculation, for example, according to the past data, only calculate the average failure-free time of the components with a failure rate greater than the failure threshold.

[0050] Step 3, after obtaining the average failure-free time of each component of a device, select the minimum value MTTF1 in the average failure-free time, and record it together with the configuration of each component at this time;

[0051] In this step, the components that have obtained the average failure-free time are selected from them to select the minimum value of the average failure-free time, because this minimum value is the stable running time of the device under the current configuration, once the minimum value is reached, a corresponding component of the device may fail or have a hidden danger, thereby breaking the stable running state of the device.

[0052] Step 4, adjust the configuration of each component respectively, when the running state of the device is improved or maintained, record the configuration of each component at this time and the corresponding minimum value of the average failure-free time, until the failure probability of the current device is less than the safe threshold;

[0053] Taking the compressor of an air conditioner as an example, the components of the compressor include an oil pump, a valve and the like, and adjusting the arrangement of the components is to arrange and combine the adjustable parameters of the components such as the oil pump and the valve, and each arrangement and combination is a configuration of the compressor. If a configuration can improve the operation state of the equipment or can maintain the current operation state, the configuration and the minimum value of the average time-to-failure of the corresponding components can be recorded. The improvement or maintenance of the operation state of the equipment is to be judged according to the specific equipment. For example, for a refrigerator, the maintenance of the current operation state of the refrigerator can be that the temperature is stabilized within a threshold range, or the output power is stabilized within a threshold range, on the basis of which the failure probability does not increase, and the minimum value of the average time-to-failure does not further decrease. If the operation state of the refrigerator is improved, the user's demand can be met while the energy efficiency is improved, on the basis of which the failure probability of the refrigerator is reduced, or the average time-to-failure is further increased.

[0054] Step 5: selecting the maximum value from the recorded minimum values of the average time-to-failure, and adjusting the arrangement of each component of the equipment to the arrangement of each component corresponding to the maximum value.

[0055] Through the above steps, when the failure probability of the equipment is greater than or equal to the safety threshold but less than or equal to the danger threshold, the arrangement of each component of the equipment is adjusted, and a configuration that can meet the user's current demand for using the equipment and can make the stable operation time of the equipment longer is screened out.

[0056] In a specific embodiment, the average time-to-failure of the component of the present application is obtained from the probability density function of the reliability of the component. The probability density refers to the probability of a random event, and thus the probability density function of the reliability of the component refers to a function that can represent the reliability of the component over time. The probability density function of the reliability of the component is much more accurate than the existing technology which simply divides the total operation time by the number of failures.

[0057] In a specific embodiment, the probability density function of the reliability of the component of the present application is obtained from the reliability data curve of the component when it is shipped. The probability density function of the reliability of the component requires test data of the component, which can be provided when the component is shipped. The reliability data curve of the component can be obtained from the test data, and some component manufacturers can directly provide the reliability data curve of the component. The probability density function of the reliability of the component can be obtained from the reliability data curve of the component.

[0058] In an embodiment, the probability density function of the reliability of the component of the present application is obtained from the formula The reliability data curve is a function of R(t), and t is the working time of the component. Through the formula, the change trend of the reliability can be obtained, so that the probability density function of the reliability can be obtained accurately at the corresponding working time of the component.

[0059] In one embodiment, the mean time to failure of the present application is calculated by the formula Through the formula, the mean time to failure of the component can be accurately quantified. It should be noted that when the corresponding configuration is selected, the reliability of each component also changes over time as the device works, so the failure probability of the device in the next cycle may be different even if the current configuration is used, and at the same time, the reliability of each component of the current configuration is also changing, i.e. the mean time to failure of each component is also changing.

[0060] Although the present application does not specifically describe the prediction of the failure probability of the device in detail, the present application will introduce three general ways to predict the failure probability of the device.

[0061] The first way is to obtain the vibration data of each component of the device. Generally, the components of the device will vibrate when the device is working. By studying the vibration data of the main components or all components of the device, the failure probability of the device can be predicted.

[0062] Specifically, the vibration data can be plotted into a vibration curve, and based on the change trend of the vibration curve, the failure probability of the device can be predicted.

[0063] The second way is to obtain the temperature rise change data of each component of the device. The temperature of some components of the device will change when the device is working. By studying the temperature rise change data of the main components or all components of the device, the failure probability of the device can be predicted.

[0064] Specifically, the temperature rise change data can be plotted into a temperature rise rate curve, and based on the change trend of the temperature rise rate curve, the failure probability of the device can be predicted.

[0065] The third way is to consider both the vibration data and the temperature rise change data of the device, obtain the vibration data and the temperature rise change data of the main components or all components of the device, plot the vibration data and the temperature rise data into a vibration curve and a temperature rise rate curve respectively, and comprehensively judge based on the change trend of the vibration curve and the temperature rise rate curve to predict the failure probability of the device.

[0066] The specific device can select one of the three ways according to the actual situation of the device. Through the above three ways, the failure prediction of most devices can be basically covered, so that the technicians in the art can make a simple prediction of the failure of most devices, and there are many further detailed failure prediction steps in the prior art. Based on one of the above three ways, the detailed failure prediction steps in the prior art can also be implemented to obtain a more accurate device failure probability.

[0067] In one embodiment, the improved or maintained running state of the device includes that the adjusted output power meets the user demand, the failure probability is less than or equal to the danger threshold, and the minimum value of the average time-to-failure of all components is greater than or equal to MTTF1. This embodiment mainly focuses on that the device can have a longer stable running time. For example, for an air conditioner, a longer stable running time is an important indicator, so when the device is an air conditioner, this way can be used to judge whether the running state of the device is improved or maintained.

[0068] In another embodiment, the improved or maintained running state of the device includes that the adjusted output power meets the user demand, the failure probability is less than the safety threshold, and the minimum value of the average time-to-failure of all components is greater than or equal to MTTF1. This embodiment mainly focuses on the failure probability of the device, and further considers the longer stable running time of the device on the basis of the lower failure probability of the device. This way can be applied in occasions where the failure probability of the device is required to be higher to meet the user's demand.

[0069] The present application also protects the corresponding device, which includes components and a control module for controlling the components, and the control module of the device uses the above-mentioned device control method based on failure prediction to control the components.

[0070] The device referred to in the present application can include various devices, such as air conditioners, refrigerators, televisions, etc.

[0071] When the device is an air conditioner, the components referred to in the present application include at least one of the sensors, oil pumps, and electronic valves. These are some main components of the air conditioner, and by calculating and screening the average time-to-failure of these components, the minimum value of the maximum average time-to-failure is selected from multiple configurations, so that the air conditioner can have a longer stable running time.

[0072] The following takes an air conditioner as a specific device for example to facilitate further understanding of the present application.

[0073] As Figure 2As shown, the unit of the air conditioner can obtain the vibration conditions of each component of the air conditioner under the current operating condition and other monitorable parameters through the sensor during the operation of the unit, and then judge and predict the possible faults of the unit. There are relatively mature prediction methods at present, such as average moving line, machine learning and the like.

[0074] For example, the vibration conditions of the compressor, valve and the like, the temperature rise rate and the like can be monitored in a cycle during the operation of the unit, and the corresponding data is saved, and then the saved data is analyzed to predict the change trend and the probability of failure, and once the vibration data of the corresponding component is monitored to exceed the preset value, or the temperature rise rate exceeds the preset value, or the predicted failure probability is greater than the danger threshold P1, the warning information can be given immediately.

[0075] When the failure probability of the air conditioner is obtained, for the unit that does not appear failure and the failure is between the safety threshold P2 and the danger threshold P1, the above control method of the application can be used to adjust and control the air conditioner.

[0076] The input and output power is recorded in real time, so as to obtain the unit configuration in the best operating state of itself and the energy efficiency in each operating state on the basis of ensuring that the output power is sufficient.

[0077] The probability density function f1(t) of each component of the unit under the current configuration and the minimum value MTTF1 of the mean time to failure MTTF (i.e. the mathematical expectation of the failure-free working time T) are calculated, and the configuration status and f1(t) and MTTF1 at this time are recorded and saved. The components of the unit can be subdivided to each sensor, oil pump, electronic valve and other components that can have a greater impact on the operating condition of the unit.

[0078] The probability density function f(t) can be obtained by obtaining the reliability data curve of each product component at the factory, i.e. the following conversion is performed on the component reliability R(t):

[0079] In actual production, the reliability estimate value is the ratio of the number of products that can complete the target function in a specified time interval to the total number of products put into work in the time.

[0080] The MTTF can be obtained by the following formula:

[0081] After the unit can be stably operated for a period of time, the failure probability is less than or equal to P1, and the output power is sufficient, the configuration parameters of the components corresponding to MTTF1 are adjusted again.

[0082] If the failure probability is greater than P1 after adjustment, or the output power of the unit is insufficient, or the minimum value of the average failure-free time of the parts is less than MTTF1, then restore to the configuration before adjustment; otherwise, record the configuration at this time and f2(t), MTTF2.

[0083] Repeat several times, or repeat a predetermined number of times N, until the failure probability is less than or equal to P2, compare the recorded values MTTF1, MTTF2,..., MTTFN, find the maximum value MTTFmax, and adjust the configuration of each part of the unit to the configuration corresponding to MTTFmax.

[0084] If the failure probability is less than P2 after adjustment, keep the current configuration of the unit, get the input and output power, calculate the energy efficiency, and get the optimal working state.

[0085] Figure 3 The reliability estimate value table of a certain part of the unit is given, and the R(t) = e -0.002t According to the curve, when the time approaches 0, R(t) approaches 1, i.e. when the working time of the part approaches 0, basically all products put into work can complete the target function without failure; when the time approaches ∞, R(t) approaches 0, at this time almost all products put into work cannot complete the work. The specific curve can be referred to Figure 4 , where the Y axis represents f(t) and the X axis represents working time.

[0086] Substitute R(t) into the corresponding density probability function to get f(t) = 0.002e -0.002t , so the MTTF at this time is 500h, i.e. the part will fail after an average of 500h of work.

[0087] When the air conditioner is a multi-module system, assume that the failure probability of one of the units is greater than the danger threshold P1, and the modular scheduling can be performed to reduce the operating power of the unit until the vibration and temperature rise deviation no longer exceeds the preset range, and when the failure probability is less than P1, the unit configuration is adjusted through MTTF. If it cannot simultaneously satisfy the failure probability less than P1 and the output power is sufficient, it is preferred to satisfy the failure probability less than P1. For the reduced output power, the operating power of other units is increased to make up for it, and the failure probability and energy efficiency of other units are calculated to configure their operating parameters, so that the failure probability is less than P1 while the output power is improved, in order to ensure the stability of the unit operation.

[0088] As Figure 5As shown, when a unit appears to have a shutdown failure, the unit is closed, and other units are dispatched to make up the output through modularization. Once it is calculated that other units cannot make up the missing output while satisfying the failure probability less than P1, the other units are kept running in the optimal configuration, the failure is reported, and maintenance is notified.

[0089] The present application can make the unit run in the maximum average time before failure with a lower failure rate, thereby ensuring the continuous and stable operation of the unit.

[0090] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. Any modification, equivalent replacement, and improvement within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A device control method based on fault prediction, characterized in that, include: Predict the probability of equipment failure; If the current failure probability of the equipment is within the range of [safe threshold, dangerous threshold], calculate the average time before failure for each component of the equipment under the current configuration; Select the minimum value MTTF1 in the mean time to failure and record it together with the configuration of each component at this time; Adjust the configuration of each component individually. When the operating status of the equipment improves or is maintained, record the configuration of each component and the minimum value of the corresponding mean time before failure, until the failure probability of the current equipment is less than the safety threshold. Select the maximum value from the minimum recorded mean time before failure, and adjust the configuration of each component of the equipment to the configuration corresponding to that maximum value.

2. The equipment control method based on fault prediction as described in claim 1, characterized in that, The mean time before failure of the component is obtained by the probability density function of the component's reliability.

3. The equipment control method based on fault prediction as described in claim 2, characterized in that, The probability density function of the reliability of the component is obtained by using the reliability data curve of the component at the time of manufacture.

4. The equipment control method based on fault prediction as described in claim 3, characterized in that, The probability density function of the reliability of the component is expressed by the formula. The calculation shows that R(t) is a function of the reliability data curve, and t is the working time of the component.

5. The equipment control method based on fault prediction as described in claim 4, characterized in that, The mean time before failure is expressed by the formula Calculated.

6. The equipment control method based on fault prediction as described in claim 1, characterized in that, Predicting the failure probability of equipment includes: Acquire vibration data and / or temperature rise data of each component of the equipment; Plot the vibration data and / or temperature rise data as vibration curves and temperature rise rate curves, respectively. Based on the changing trends of vibration curves and / or temperature rise rate curves, the failure probability of the equipment is predicted.

7. The equipment control method based on fault prediction as described in claim 1, characterized in that, The improvement or maintenance of the operating status of the equipment includes: The adjusted output power meets user requirements, the failure probability is less than or equal to the danger threshold, and the minimum mean time to failure among all components is greater than or equal to MTTF1.

8. The equipment control method based on fault prediction as described in claim 1, characterized in that, The improvement or maintenance of the operating status of the equipment includes: The adjusted output power meets user requirements, the failure probability is less than or equal to the safety threshold, and the minimum mean time to failure (MTTF) of all components is greater than or equal to MTTF1.

9. A device comprising components and a control module for controlling the components, characterized in that, The control module uses the fault prediction-based equipment control method as described in any one of claims 1 to 8 to control each component.

10. The device as claimed in claim 9, characterized in that, The equipment includes an air conditioner.

Images