A method for predicting tool life cycle in intelligent tool cabinet
The reliability of environmental parameter data is evaluated through information entropy and extreme difference, the Pearson correlation coefficient is corrected, and the influence of multiple factors is quantified by the control variable method, which solves the problem of noise interference in the life cycle prediction of power tools and improves the accuracy of prediction and reliability provide scientific basis for the maintenance and resource management of power tools.
Patent Information
- Application Number
- CN202411190571.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-28
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2044-08-28
AI Technical Summary
In the prior art, when predicting the life cycle of power tools, the environmental parameter data in the smart tool cabinet is susceptible to noise interference, reducing the accuracy and reliability of the prediction results.
Through information entropy and extreme difference, the Pearson correlation coefficient is corrected, combined with historical data to predict, and the control variable method quantifies the influence of environmental parameters, usage frequency and maintenance frequency data to improve the life cycle of the power tool Prediction accuracy.
Effectively identify and eliminate noise data, improve the quality of environmental parameter data, enhance the accuracy and reliability of power tool life cycle prediction, and provide reliable support for maintenance planning and resource management.
Smart Images

Figure CN118709864B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of equipment life prediction, and in particular to a method for predicting the life cycle of tools in an intelligent tool cabinet. Background Art
[0002] Power tools are tools that are operated by hand or hanging and are driven by electric motors. They are divided into two types: handheld and portable. There are many types of power tools, including but not limited to electric drills, electric welding tongs, concrete vibrators, electric planers, electric grinders, etc. These tools are powered by electric motors to complete various tasks, such as compacting concrete foundations, planing wood, sanding operations, etc. In order to ensure that power tools can be used normally, preventive maintenance is required. That is, by predicting the life cycle of power tools, appropriate maintenance or replacement can be arranged before they fail, thereby reducing unexpected downtime and maintenance costs. Accurately predicting the life cycle of power tools helps to better plan resource allocation, ensure that tools are available at critical moments, and avoid waste of resources caused by over-purchase or premature scrapping.
[0003] A smart tool cabinet is a device used to store and manage power tools. Power tools are usually stored in a smart tool cabinet after use. The environmental conditions inside the smart tool cabinet have a direct impact on the performance and life cycle of power tools. For example, the humidity inside the smart tool cabinet will have a certain impact on the life of power tools and shorten their life cycle, because water vapor in a humid environment can penetrate into the interior of the electric equipment and react with the metal in the circuit, thereby accelerating the corrosion rate of the metal, which can cause damage to multiple components of the electronic equipment and even cause faults such as short circuits, thereby affecting the normal operation of the equipment. Therefore, the life cycle of power tools stored in the smart tool cabinet can be accurately predicted based on the environmental conditions inside the tool cabinet, which is conducive to the arrangement of power tool maintenance plans and the update and management of power tool resources.
[0004] When predicting the life cycle of a tool or equipment, the prior art usually comprehensively considers the impact of environmental conditions on the life cycle to obtain accurate prediction results. For example, a Chinese patent application document with publication number CN109598382A provides a life prediction method and device, by respectively obtaining a first value of an operating parameter of a hydrogen storage device and a second value of an operating environment parameter of the hydrogen storage device. If the first value of the operating parameter of the hydrogen storage device indicates that the hydrogen storage device completes a charging and discharging process, it is determined that the hydrogen storage device consumes one cycle life under the working parameters, based on the second value of the working environment parameter of the hydrogen storage device, the cycle life consumed by the hydrogen storage device under the working environment parameters is determined, and based on the cycle life of the current hydrogen storage device and the cycle life consumed by the hydrogen storage device under the working parameters and the working environment parameters, the remaining cycle life of the hydrogen storage device is obtained. Therefore, the remaining cycle life of the hydrogen storage device can be predicted by obtaining the first value of the working parameter of the hydrogen storage device and the second value of the working environment parameter.
[0005] However, when predicting the life cycle of power tools, the environmental parameters corresponding to the environmental conditions in the smart tool cabinet are mainly collected through various related sensors. Since the sensors are easily disturbed by the surrounding environment, such as the influence of electromagnetic fields, the collected environmental parameters fluctuate irregularly, that is, noise is introduced. These noises will reduce the accuracy and reliability of the prediction results of the life cycle of power tools. Summary of the invention
[0006] In order to solve the problem that when predicting the life cycle of a power tool, the noise in the environmental parameter data reduces the accuracy and reliability of the subsequent analysis and prediction results of the life cycle of the power tool, the present invention proposes a tool life cycle prediction method in an intelligent tool cabinet, which specifically adopts the following scheme:
[0007] A method for predicting tool life cycle in an intelligent tool cabinet, comprising:
[0008] Acquire environmental parameter data of the smart tool cabinet within a specific time period, and evaluate the first reliability of the environmental parameter data according to the product of information entropy and range of the environmental parameter data, wherein the first reliability is negatively correlated with the product;
[0009] According to the mean of the change amplitudes of all adjacent data in the environmental parameter data and the accumulated difference value between the maximum value in the environmental parameter data and its nearest neighboring data, the first reliability of the environmental parameter data is corrected to the second reliability, wherein the second reliability is negatively correlated with the mean of the change amplitude and the accumulated difference value, and positively correlated with the first reliability;
[0010] The second reliability is used to correct the Pearson correlation coefficient between the environmental parameter data and the life cycle data of the power tool, and the life cycle of the power tool to be predicted in the smart tool cabinet is predicted based on the corrected Pearson correlation coefficient combined with the historical life cycle average of the power tools.
[0011] The above technical solution measures the uncertainty of data through information entropy, and the range reflects the maximum fluctuation range of data. The result of multiplying the two can be used to quantify the instability of data. The larger the product, the greater the data fluctuation and the lower the reliability; vice versa. This operation helps to preliminarily screen out relatively stable and reliable environmental parameter data, and further refines the evaluation of data reliability. By calculating the average change amplitude between adjacent data and the cumulative value of the difference between the maximum value and the nearest data, those data points with large fluctuations and possible noise can be more accurately identified. At the same time, maintaining a positive correlation with the first reliability ensures that the data previously rated as highly reliable is still more reliable overall, and the Pearson correlation coefficient between the environmental parameter data and the life cycle data of the power tool is introduced to measure the strength of the correlation between the two. By correcting this coefficient through the second reliability, the impact of noise data on the correlation evaluation can be reduced, and the combination of historical data and the corrected correlation coefficient for prediction can more accurately estimate the life cycle of the power tool and obtain accurate and reliable prediction results.
[0012] Furthermore, predicting the life cycle of the electric tools to be predicted in the smart tool cabinet includes:
[0013] The control variable method is used to quantify the impact of the three dimensions of environmental parameter data, power tool usage frequency data, and power tool maintenance frequency data within the specific time period on the life cycle of the power tool. , , ;
[0014] The predicted value of the life cycle of the electric tool to be predicted satisfies the following relationship:
[0015] , is the average life cycle of all power tools in history, and They are the Pearson correlation coefficient between the use frequency of power tools and the life cycle of power tools, and the Pearson correlation coefficient between the maintenance frequency of power tools and the life cycle of power tools. for , and the sum of is the modified Pearson correlation coefficient.
[0016] The above technical solution quantifies the impact of environmental parameter data, usage frequency data and maintenance frequency data on the life cycle of power tools through the control variable method, and introduces the modified Pearson correlation coefficient, thereby ensuring that the prediction model can accurately reflect the correlation between each factor and the life cycle of power tools, and helps to identify which factors have a greater impact on the life of power tools. In addition, combined with the historical life cycle average of power tools, a more accurate prediction value of the life cycle of the power tool to be predicted can be obtained, providing reliable support for the maintenance plan and resource management of power tools.
[0017] Furthermore, the method for obtaining the degree of influence of the environmental parameter data, the use frequency data of the electric tool and the maintenance frequency data of the electric tool on the life cycle of the electric drill within the specific time period is:
[0018] The data of any one dimension among the environmental parameter data, the use frequency data of the electric tool, and the maintenance frequency data of the electric tool is used as the target dimension data, and the data other than the target dimension is used as the other dimension data;
[0019] The difference between the target dimension data in the specific time period and the mean of the target dimension data in all historical time periods is used as the data deviation of the target dimension data;
[0020] Select electric tools that have the same other dimensional data in their respective life cycles in history, and whose average difference in target dimensional data in their respective life cycles is consistent with the data deviation of the target dimensional data, and use the average difference in the life cycles corresponding to these electric tools as the degree of influence of the target dimensional data on the predicted life cycle of the electric tools.
[0021] The above technical solution calculates the deviation between the target dimension data and the mean of historical data in a specific time period, and finds historical samples with similar other dimension data but different target dimension data. This method can effectively isolate the impact of a single factor on the life of the electric drill. By comparing historical data in which only the target dimension changes under similar conditions, the actual impact of changes in the target dimension (such as frequency of use) on the life cycle of the electric drill can be more accurately estimated, thereby improving the accuracy of the prediction model. Since the solution adopts the control variable method, it can identify key factors in complex actual environments.
[0022] Furthermore, the modified Pearson correlation coefficient between the environmental parameter data and the life cycle data of the electric tool satisfies the following relationship:
[0023]
[0024] In the formula, is the corrected Pearson correlation coefficient between the environmental parameter data and the life cycle data of the power tool, The second reliability of the environmental parameter data, is the Pearson correlation coefficient between the environmental parameter data and the life cycle data of the power tool.
[0025] The above technical solution is to make the second reliability With the original Pearson correlation coefficient Multiplication ensures that the correlation evaluation is given higher weight only when the data is relatively stable. The modified Pearson correlation coefficient It can more accurately reflect the correlation between environmental parameter data and the life cycle of power tools.
[0026] Furthermore, the life cycle average of all the power tools in history is obtained by utilizing big data to obtain the total life cycle of multiple power tools, and the ratio of the total life cycle to the total number of power tools is used as the life cycle average of all the power tools in history.
[0027] Furthermore, when the target dimension data is environmental parameter data, the target dimension data of the electric tools within their respective life cycles are represented by the average value of the environmental parameter data.
[0028] Furthermore, the second reliability of the environmental parameter data satisfies the following relationship:
[0029]
[0030] In the formula, The second reliability of the environmental parameter data, is the first reliability of the environmental parameter data, is the mean value of the variation range of all adjacent data in the environmental parameter data, is the maximum value in the environmental parameter data, is the maximum value of The nearest neighbor value, is the number of all nearest neighbors of the maximum value, Preset hyperparameters.
[0031] The above technical solution calculates the mean of the change amplitude of adjacent data The difference between the maximum value and its nearest neighbor data can more finely identify abnormal fluctuations in the data and make the first reliability Including it in the calculation ensures that the second reliability takes into account both the uncertainty of the original data and the trend of data changes, thereby improving the comprehensiveness of the assessment. The lower the reliability, the more likely it is noise data, which helps to eliminate the impact of noise data.
[0032] Furthermore, the first reliability of the environmental parameter data satisfies the following relationship:
[0033]
[0034] In the formula, is the first reliability of the environmental parameter data, is the information entropy value of the environmental parameter data, and are the maximum and minimum values of the environmental parameter data, respectively. To preset hyperparameters, is a natural exponential function.
[0035] The above technical solution introduces information entropy and range ,Information entropy is a statistic used to measure the uncertainty of data.,The range reflects the fluctuation range of data. The uncertainty and fluctuation range of data are combined to objectively evaluate the reliability of data.
[0036] Furthermore, the variation range of the adjacent data is the absolute value of the difference between two adjacent data.
[0037] Furthermore, the specific time period is a time period from when the electric tool is first stored in the smart tool cabinet to the current moment.
[0038] The present invention has the following effects:
[0039] The present invention evaluates the first reliability of environmental parameter data through information entropy and range, effectively identifies the stability and uncertainty of data, corrects the first reliability by calculating the mean and maximum values of the change amplitudes of adjacent data and the cumulative value of the difference between their nearest data, further eliminates abnormal fluctuations, reduces the impact of noise data, and improves data quality. The Pearson correlation coefficient between environmental parameter data and the life cycle of power tools is adjusted using the corrected second reliability, thereby improving the accuracy and reliability of the life cycle prediction results of power tools and providing reliable support for subsequent maintenance plans and resource management. BRIEF DESCRIPTION OF THE DRAWINGS
[0040] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:
[0041] Figure 1 It is a schematic flow chart of the method of the present invention. DETAILED DESCRIPTION
[0042] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.
[0043] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.
[0044] Reference Figure 1 , a method for predicting tool life cycle in an intelligent tool cabinet, comprising steps S1-S4, specifically as follows:
[0045] S1: Acquire environmental parameter data of the smart tool cabinet within a specific time period, and evaluate the first reliability of the environmental parameter data.
[0046] A smart tool cabinet is a device used to store and manage power tools. The environmental conditions inside it have a direct impact on the performance and life of the tools. The following are several typical environmental conditions that a smart tool cabinet may face:
[0047] Temperature: High temperatures may cause the power tool to overheat and accelerate the aging of internal components, while low temperatures may affect battery performance and cause difficulty in starting. Humidity: High humidity may cause water vapor to adhere to metal parts, causing them to rust and corrode, or water vapor may enter the circuit board, accelerating the corrosion of the circuit board and affecting the normal use of the power tool. Dust / particles: They may block the vents and affect heat dissipation, or enter the inside of the power tool and cause wear. Electromagnetic interference: Electromagnetic interference generated by nearby electrical appliances may affect the electronic control system of the power tool. Vibration: Long-term mechanical vibration may cause fasteners to loosen and affect the structural stability of the tool.
[0048] Among these environmental conditions, humidity is a particularly important factor, which can easily cause oxidation and corrosion of the metal parts of the power tool, resulting in poor connection and affecting the use. Therefore, the environmental parameter data in this embodiment is taken as an example of humidity data. Therefore, the environmental parameter data referred to in this embodiment all represent humidity data, which are collected by corresponding humidity sensors. By monitoring the environmental parameter data, it can be used to evaluate the life cycle of the power tool and determine its remaining life. A more reasonable and effective maintenance plan can be formulated to avoid excessive or insufficient maintenance, and preventive maintenance can reduce unexpected failures, thereby reducing maintenance costs and downtime.
[0049] This embodiment uses sensors to continuously collect environmental parameter data in the smart tool cabinet, and the collection frequency is set to once per minute to ensure that the details of changes in environmental parameters can be captured. It also takes into account the feasibility of data storage and processing, and ensures that the collected environmental parameter data covers the power tool from the first time it is placed in the smart tool cabinet to the current analysis moment.
[0050] The smart tool cabinet can store multiple power tools. You can select any power tool to predict its life cycle, obtain the usage frequency data and maintenance frequency data of the power tool from the time it was first placed in the smart tool cabinet to the current time (specific time period), and the humidity data in the smart power cabinet. Since the environmental parameter data is collected by sensors, the sensors are easily affected by the environment during the collection process, resulting in noise data in the collected environmental parameter data, which affects the accuracy of subsequent prediction results. Therefore, it is necessary to evaluate the environmental parameter data, eliminate the influence of noise data, and obtain high-quality environmental parameter data for subsequent predictions.
[0051] Therefore, this step evaluates the first reliability of the environmental parameter data according to the product of the information entropy and the range of the environmental parameter data. The first reliability is negatively correlated with the product, which can be expressed as follows:
[0052]
[0053] In the formula, is the first reliability of the environmental parameter data, is the information entropy value of the environmental parameter data, and are the maximum and minimum values of the environmental parameter data, respectively. The default hyperparameter is set to 0.001 to avoid In the case of zero, is a natural exponential function.
[0054] This formula is calculated by calculating the information entropy and range , which can effectively evaluate the stability of environmental parameter data. The smaller the information entropy, the more concentrated the data distribution, that is, the more stable the data. The smaller the range, the smaller the data fluctuation range, which also reflects the better stability of the data. The natural exponential function is used to convert the numerical value into a value between 0 and 1. The closer the value is to 1, the more reliable the environmental parameter data is, that is, the higher the first reliability is. Conversely, the less reliable the environmental parameter data is, the lower the first reliability is. This quantitative method is helpful for intuitive preliminary evaluation of data quality.
[0055] When the environmental parameter data fluctuates greatly, the first reliability will be significantly reduced, indicating that these data may not be suitable for life cycle prediction. If they are used directly for life cycle prediction, the accuracy of the prediction results will be affected.
[0056] S2: Correct the first reliability of the environmental parameter data to the second reliability according to the average of the change amplitudes of all adjacent data in the environmental parameter data and the accumulated difference between the maximum value in the environmental parameter data and its nearest adjacent data.
[0057] Through the analysis of the above steps, the first reliability of the environmental parameter data within a specific time period is obtained. This indicator is based on the numerical performance of the environmental parameter data. In a stable environment, the air flow is relatively stable, and the environmental parameter data will not change suddenly. It often shows a gradual change feature, and the noise data often appears in the form of mutation points. Therefore, this step can obtain the second reliability of the environmental parameter data based on the gradual change of the environmental parameter data within a specific time period and the first reliability of the environmental parameter data. If the gradual change of the environmental parameter data is weaker, it means that there is a greater possibility of noise in the environmental parameter data, and its second reliability will be lower.
[0058] The second reliability of the environmental parameter data satisfies the following relationship:
[0059]
[0060] In this formula, The second reliability of the environmental parameter data, is the first reliability of the environmental parameter data, is the mean of the variation range of all adjacent data in the environmental parameter data, reflecting the gradual change of the environmental parameter data. If the value is small, it means that the environmental parameter data changes slowly and the gradual change is good, so the second reliability is high; on the contrary, if If it is large, it means that the environmental parameter data changes dramatically, there may be noise interference, and the reliability is low. is the maximum value in the environmental parameter data, is the maximum value of The nearest neighbor value, represents the preset hyperparameters, , which exists to prevent The situation occurs, that is, the maximum value in the environmental parameter data is equal to the value of its nearest neighbor data. is the number of all the nearest neighbor data of the maximum value. The nearest neighbor here refers to the nearest value. For example, if the environmental parameter data are 20, 23, 40, 39, and 36, and the maximum value is 40, then , then the two nearest neighbors of 40 are 39 and 36. , then the three nearest neighbors of 40 are 39, 36 and 23.
[0061] In this formula, Represents the maximum value of environmental parameter data and its nearest neighbor The difference accumulation value of the environmental parameter data. The larger the value, the more likely the maximum value of the environmental parameter data is caused by noise. If the gradual change of the environmental parameter data is weaker ( The larger the value of ), the more likely it is that there is noise in the environmental parameter data, and the lower its second reliability.
[0062] In this formula, is the number of all nearest neighbor data of the maximum value. If the number of environmental parameter data is less than 1000, set If the number of environmental parameter data collected is set between 1000 and 10000 If the number of collected environmental parameter data is greater than 10,000, set .
[0063] In this formula, the change range of adjacent data is the absolute value of the difference between two adjacent data. For example, if the environmental parameter data are 20, 23, 24, 18, 16, 21, 19, 22, 17, and 20, the change ranges of the adjacent data are 3, 1, 6, 2, 5, 2, 3, 5, and 3 respectively. The average of the change ranges of all adjacent data is , ,like , then select The five nearest environmental parameter data are: 23, 22, 21, 20, 20. In this formula, .
[0064] S3: Using the second reliability, correct the Pearson correlation coefficient between the environmental parameter data and the life cycle data of the electric tool.
[0065] The purpose of calculating the Pearson correlation coefficient is to quantify the strength of the linear correlation between two variables. In this scenario, the Pearson correlation coefficient reflects the strength of the correlation between the environmental parameter data and the life cycle of the power tool. Calculating the Pearson correlation coefficient can identify which environmental conditions have the greatest impact on the life cycle of the power tool.
[0066] The calculation method of the Pearson correlation coefficient between the environmental parameter data and the life cycle data of the power tool is as follows: using big data technology to obtain the life cycle data of all power tools stored in the smart tool cabinet in history, as well as the mean value of the environmental parameter data corresponding to the life cycle of each power tool in history, and based on these data, calculate the Pearson correlation coefficient between the environmental parameter data and the life cycle data of the power tool. .
[0067] For example, historically, there are 10 power tools stored in the smart tool cabinet, and the sequence of their life cycle data (in years) is , the sequence of the mean values of the environmental parameter data corresponding to these 10 power tools during their respective life cycles (humidity data is used as an example here) Based on these two sets of data, the Pearson correlation coefficient between the environmental parameter data and the life cycle data of the power tool is calculated. , which means that there is a negative correlation between the environmental parameter data and the life cycle of the power tool, that is, when the environmental parameter data becomes larger, the life of the power tool will be slightly shortened. This is a reasonable correlation coefficient, indicating that the environmental parameter data has a certain impact on the life cycle of the power tool.
[0068] According to the second reliability of the environmental parameter data in a specific time period, the Pearson correlation coefficient between the environmental parameter data and the life cycle of the power tool is corrected, and the corrected Pearson correlation coefficient is calculated, and the corrected Pearson correlation coefficient satisfies the following relationship:
[0069]
[0070] In the formula, is the corrected Pearson correlation coefficient between the environmental parameter data and the life cycle data of the power tool, The second reliability of environmental parameter data, is the Pearson correlation coefficient between the environmental parameter data and the life cycle data of the power tool.
[0071] The modified Pearson correlation coefficient takes into account the secondary reliability of the environmental parameter data and more accurately reflects the actual correlation between the environmental parameter data and the life cycle of the electric drill. Using a more accurate correlation coefficient for prediction can help improve the accuracy of the power tool life cycle prediction, thereby better formulating maintenance plans and resource management strategies.
[0072] S4: Predicting the life cycle of the electric tools to be predicted in the smart tool cabinet according to the modified Pearson correlation coefficient and the historical life cycle average of the electric tools.
[0073] When evaluating and predicting the life cycle of power tools, the traditional method is to obtain the life cycle average of multiple historical power tools to obtain the life cycle of the power tool to be predicted. Although this method is easy to operate, it has significant limitations because it ignores the impact of environmental parameter data on the life cycle of power tools. In addition, the frequency of use and maintenance of power tools are also important factors in determining their life cycle. The frequency of use of power tools will significantly affect their lifespan. High-frequency use usually means more frequent wear and damage. Therefore, considering the frequency of use of power tools can more accurately predict their life cycle. In addition, regular maintenance and maintenance can help extend the service life of power tools.
[0074] In summary, in order to improve the accuracy of power tool life cycle prediction, it is necessary to comprehensively consider environmental parameter data, usage frequency, maintenance frequency and other data. By establishing a comprehensive prediction model, the actual service life of the tool can be evaluated more scientifically, thereby formulating more effective maintenance plans and resource management strategies.
[0075] The method for predicting the life cycle of the electric tools to be predicted in the smart tool cabinet is as follows:
[0076] S41: quantifying the influence of the environmental parameter data, the usage frequency data and the maintenance frequency data in the specific time period on the life cycle of the electric tool by using a control variable method.
[0077] In order to facilitate the calculation and quantification of data, when analyzing the environmental parameter data within a specific time period or the environmental parameter data of each power tool in its life cycle in history, the mean value of the environmental parameter data is used to represent it, so that each power tool corresponds to one environmental parameter data (represented by the mean value of the environmental parameter data within its life cycle).
[0078] The overall logic of the control variable method is:
[0079] The data of any one dimension among the environmental parameter data, the use frequency data of the electric tool, and the maintenance frequency data of the electric tool is used as the target dimension data, and the data other than the target dimension is used as the other dimension data;
[0080] The difference between the target dimension data in the specific time period and the mean value of the target dimension data of the historical electric tools in their respective life cycles is taken as the data deviation of the target dimension data. Calculating the data deviation of the target dimension data helps to determine which electric tools in history are closest to the electric tool to be predicted in this dimension. By selecting electric tools with similar data deviations, the impact of the target dimension data on the life cycle of the electric tool to be predicted can be more accurately estimated.
[0081] Select electric tools whose other dimensional data are consistent in history, and whose average difference in target dimensional data within their respective life cycles is consistent with the data deviation of the target dimensional data. At this time, the average difference in the life cycles corresponding to these electric tools is used as the degree of influence of the target dimensional data on the predicted life cycle of the electric tools.
[0082] For example, the environmental parameter data (expressed as mean) of the power tool to be predicted in a specific time period is 46, the usage frequency is 0.1, and the maintenance frequency is 0.5. Historically, 10 power tools were stored in the smart tool cabinet. The sequence of life cycle data is , the life cycle is measured in years.
[0083] The sequence of environmental parameter data (in the smart tool cabinet) corresponding to these 10 power tools during their respective life cycles: ; Sequence of usage frequency data of these 10 power tools: ; The sequence of maintenance frequency data of these 10 power tools: , and the units of usage frequency and maintenance frequency are both years / times.
[0084] First, the environmental parameter data is used as the target dimension data, and the power tool usage frequency data and the power tool maintenance frequency data are used as other dimension data;
[0085] The environmental parameter data of the power tool to be predicted in a specific time period and the average of the environmental parameter data of multiple power tools in their respective life cycles in history The difference is taken as the data deviation of the environmental parameter data of the power tool to be predicted in a specific time period, that is, ;
[0086] Select the power tools with the same historical usage frequency data and maintenance frequency, and the average difference between the environmental parameter data of these power tools in their respective life cycles is equal to the data deviation of the environmental parameter data, and then use the average difference in the life cycle of these power tools as the degree of influence of the environmental parameter data on the life cycle. In addition, if the number of these power tools is 2, the difference value of the life cycle of these two power tools is directly used as the degree of influence of the environmental parameter data on the life cycle of the power tools. In this case, there is only one difference value, and there is no need to calculate the mean value.
[0087] For example, in the above data sequence, the usage frequency of the 4th power tool and the 7th power tool are both 0.3, and the maintenance frequency is 0.4. The environmental parameter data of the 4th power tool during its life cycle is 45, and the environmental parameter data of the 7th power tool during its life cycle is 46. The difference between the environmental parameter data of the 4th power tool during its life cycle and the environmental parameter data of the 7th power tool during its life cycle is 1, which is exactly equal to the deviation of the environmental parameter data. Then, the difference between the life cycle of the 4th power tool and the life cycle of the 7th power tool is taken as the influence of the environmental parameter data on the life cycle of the power tool. ,even though .
[0088] Then, following the same method as above, control the variables:
[0089] Using frequency data as target dimension data, environmental parameter data and maintenance frequency as other dimension data, the difference between the usage frequency of the power tool to be predicted in a specific time period and the average usage frequency of multiple power tools in their respective life cycles in history is used as the data deviation of the usage frequency of the power tool to be predicted;
[0090] Select the power tools with the same environmental parameter data and maintenance frequency in their respective life cycles in history, and the average difference of the usage frequency data of these power tools in their respective life cycles is the same as the data deviation of the usage frequency of the power tools to be predicted. Then, the average difference of the life cycle of these power tools is used as the impact degree of usage frequency data on the life cycle. For example, if there are 3 power tools with life cycles of 3, 5, and 2, the average difference in the life cycles of these power tools is calculated as follows: first calculate the mean as 3.33, then calculate the deviation of each life cycle from the mean (the difference from the mean is then calculated as the absolute value), respectively. , , , and finally the mean of 0.33, 1.67, and 1.33 is calculated as 1.11, which is the average difference in the life cycle of these three power tools. The average differences of other dimensional data are also calculated using the same method.
[0091] Finally, taking maintenance frequency data as the target dimension data, environmental parameter data and usage frequency data as other dimension data, the difference between the maintenance frequency of the power tool to be predicted within a specific time period and the average maintenance frequency of multiple power tools in history during their respective life cycles is used as the data deviation of the maintenance frequency of the power tool to be predicted.
[0092] Select the power tools with the same environmental parameter data and the same usage frequency in their respective life cycles in history, and the average difference of the maintenance frequency data of these power tools in their respective life cycles is the same as the data deviation of the maintenance frequency of the power tool to be predicted, and then use the average difference of the life cycle of these power tools as the impact degree of the maintenance frequency data on the life cycle .
[0093] In summary, this step uses the control variable method to take environmental parameter data, usage frequency data and maintenance frequency data as target dimension data respectively, and keeps other dimension data unchanged, which can effectively quantify the impact of each factor on the life cycle of power tools. This method helps to clearly distinguish the impact of each factor and avoids the complexity caused by the simultaneous changes of multiple factors, thereby improving the accuracy and interpretability of the prediction.
[0094] S42: Calculate the predicted value of the life cycle of the electric tool to be predicted:
[0095]
[0096] In the formula, is the average life cycle of all power tools in history, and They are the Pearson correlation coefficient between the usage frequency data and the life cycle of the power tool, and the Pearson correlation coefficient between the maintenance frequency data and the life cycle of the power tool. is the corrected Pearson correlation coefficient between the environmental parameter data and the life cycle of the power tool, express , , the sum of , , They respectively represent the weight of the impact of environmental parameter data (expressed as mean) on the life cycle of power tools in a specific time period, the weight of the impact of usage frequency data on the life cycle of power tools, and the weight of the impact of maintenance frequency data on the life cycle of power tools.
[0097] In this formula It indicates the comprehensive influence of various influencing factors on the life cycle of the power tool to be predicted in a specific time period, through the modified Pearson correlation coefficient and the original Pearson correlation coefficient , To quantify the correlation between different factors and the life cycle of power tools, by considering the correlation coefficient, it can be ensured that the correlation between each factor and the life cycle of power tools is fully considered when calculating the degree of impact, thereby improving the accuracy of the prediction.
[0098] The calculation method of the Pearson correlation coefficient between the usage frequency data and the life cycle of the power tool is as follows: use big data technology to obtain the life cycle data of all power tools stored in the smart tool cabinet in history, as well as the usage frequency data of each power tool in history during its life cycle, and calculate the Pearson correlation coefficient between the usage frequency data and the life cycle data of the power tool based on these data. .
[0099] The calculation method of the Pearson correlation coefficient between the maintenance frequency data and the life cycle of the power tool is as follows: use big data technology to obtain the life cycle data of all power tools stored in the smart tool cabinet in history, as well as the maintenance frequency data of each power tool in history during its life cycle, and calculate the Pearson correlation coefficient between the maintenance frequency data and the life cycle data of the power tool based on these data. .
[0100] Furthermore, based on the predicted life cycle of the electric tool to be predicted, relevant measures can be taken to extend the life of the electric tool, such as optimizing storage conditions, adjusting the frequency of use, and performing regular maintenance.
[0101] In the description of this specification, "plurality" or "several" means at least two, such as two, three or more, etc., unless otherwise clearly defined.
[0102] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.
Claims
1. A method for predicting tool life cycle in an intelligent tool cabinet. It is characterized in that include: The environmental parameter data of the smart tool cabinet within a specific time period is obtained, and the first reliability of the environmental parameter data is evaluated according to the product of the information entropy and the range of the environmental parameter data, satisfying the following relationship: ; In the formula, is the first reliability of the environmental parameter data, is the information entropy value of the environmental parameter data, and are the maximum and minimum values of the environmental parameter data, respectively. To preset hyperparameters, is a natural exponential function; the first reliability is negatively correlated with the product; According to the mean of the change amplitudes of all adjacent data in the environmental parameter data and the accumulated difference between the maximum value in the environmental parameter data and its nearest neighboring data, the first reliability of the environmental parameter data is corrected to the second reliability, satisfying the following relationship: ; In the formula, The second reliability of the environmental parameter data, is the mean value of the variation range of all adjacent data in the environmental parameter data, is the maximum value of The nearest neighbor value, is the number of all the nearest data of the maximum value; the second reliability is negatively correlated with the mean of the variation range and the accumulated difference value, and is positively correlated with the first reliability; The Pearson correlation coefficient between the environmental parameter data and the life cycle data of the electric tool is corrected by using the second reliability, and the corrected Pearson correlation coefficient satisfies the following relationship: ; In the formula, is the corrected Pearson correlation coefficient between the environmental parameter data and the life cycle data of the power tool, is the Pearson correlation coefficient between the environmental parameter data and the life cycle data of the electric tool; predicting the life cycle of the electric tool to be predicted in the smart tool cabinet according to the modified Pearson correlation coefficient combined with the life cycle average of the historical electric tools, including: The control variable method is used to quantify the impact of the three dimensions of environmental parameter data, power tool usage frequency data, and power tool maintenance frequency data within the specific time period on the life cycle of the power tool. , , , the acquisition method is: take the data of any dimension among the environmental parameter data, the use frequency data of the electric tool, and the maintenance frequency data of the electric tool as the target dimension data, and take the data other than the target dimension as the other dimension data; take the difference between the target dimension data in the specific time period and the mean of the target dimension data in all historical time periods as the data deviation of the target dimension data; select the electric tools with the same other dimension data in their respective life cycles in history, and the average difference of the target dimension data of these electric tools in their respective life cycles is consistent with the data deviation of the target dimension data, and take the average difference of the life cycles corresponding to these electric tools as the influence of the target dimension data on the life cycle of the electric tool to be predicted; The predicted value of the life cycle of the electric tool to be predicted satisfies the following relationship: , is the average life cycle of all power tools in history, and They are the Pearson correlation coefficient between the use frequency of power tools and the life cycle of power tools, and the Pearson correlation coefficient between the maintenance frequency of power tools and the life cycle of power tools. for , and sum.
2. The method for predicting tool life cycle in an intelligent tool cabinet according to claim 1, It is characterized in that The average life cycle of all the power tools in history is obtained by using big data to obtain the total life cycle of multiple power tools, and the ratio of the total life cycle to the total number of power tools is used as the average life cycle of all the power tools in history.
3. The method for predicting tool life cycle in an intelligent tool cabinet according to claim 1, It is characterized in that When the target dimension data is environmental parameter data, the target dimension data of the electric tools within their respective life cycles are represented by the average value of the environmental parameter data.
4. The method for predicting tool life cycle in an intelligent tool cabinet according to claim 1, It is characterized in that The variation range of the adjacent data is the absolute value of the difference between two adjacent data.
5. The method for predicting tool life cycle in an intelligent tool cabinet according to claim 1, It is characterized in that The specific time period is a time period from when the power tool to be predicted is first stored in the smart tool cabinet to the current moment.
Citation Information
Patent Citations
Method for determining service life and reliability of high-temperature high-pressure component by utilizing ultrasonic
CN101509898A
Rolling bearing service life state same-scale characterization and recognition method under different rotating speeds
CN105956574A
A life prediction method and device
CN109598382A
Vision-based online life prediction method for high-temperature component
CN114997273A
Photovoltaic module life prediction method, storage medium and electronic equipment
CN118395869A
