A method and system for remote fault diagnosis of intermediate frequency electrotherapy instrument

Abstract

The present application belongs to the technical field of equipment fault monitoring, and provides a remote fault diagnosis method and system for a medium-frequency electrotherapy instrument, comprising: in a preset monitoring period, real-time monitoring of output power obtains a fault period and an interval sequence, and it is judged whether the period is stable and presents a single gradual change trend; if yes, a fault period is locked and multiple types of cause parameter sequences are integrated, sequences presenting a single gradual change trend are extracted, and associated analysis is performed with the fault period sequence to determine a high-probability cause parameter sequence, an cause parameter-fault period model is fitted, a subsequent fault period is predicted based on the fault period and the interval sequence, a cause prediction fault period is output by combining the high-probability cause parameter prediction result and the cause parameter-fault period model, and finally a subsequent power monitoring interval is dynamically determined, so that the balance between missed detection avoidance and data redundancy reduction is achieved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of equipment fault monitoring, and in particular relates to a remote fault diagnosis method and system for a medium-frequency electrotherapy instrument. BACKGROUND

[0002] In the remote fault diagnosis of the medium-frequency electrotherapy instrument, abnormal fluctuation of the output power is a typical fault type, and often presents periodic characteristics. Precise capture of the periodic fault is the key to ensuring safe operation of the equipment. Current remote monitoring mostly adopts fixed collection intervals. However, in actual application, the fault period is easily affected by inducing factors such as power fluctuation, module heat accumulation, and load impedance aging, and presents a single gradual change trend (such as gradually decreasing or expanding) with the use time. When the fixed collection interval is out of sync with the fault period (for example, the fault period is 2 seconds and the collection interval is 3 seconds), the fault signal will be directly missed, resulting in missed detection. If the collection interval is shortened to avoid missed detection, a large amount of redundant monitoring data will be generated, increasing the transmission and storage load of the equipment, interfering with the fault diagnosis efficiency, and making it difficult to balance the monitoring accuracy and data redundancy problem.

[0003] At the same time, the prior art lacks a judgment mechanism for the stability of the fault period, cannot timely identify the periodic gradual change trend, and does not establish a correlation analysis model of the fault period and multiple types of inducing parameters, making it difficult to lock the high-probability inducing factor, and thus unable to predict the subsequent fault period and occurrence period based on the change of the inducing factor. The above problems make the existing monitoring scheme difficult to adapt to the actual scene of dynamic changes of the fault period of the medium-frequency electrotherapy instrument, affecting the accuracy and timeliness of fault diagnosis.

[0004] Therefore, the application provides a remote fault diagnosis method and system for a medium-frequency electrotherapy instrument. SUMMARY

[0005] In order to make up for the deficiencies of the prior art and solve at least one technical problem proposed in the background art.

[0006] The technical scheme adopted by the application to solve the technical problems is: a remote fault diagnosis method for a medium-frequency electrotherapy instrument, comprising:

[0007] In a preset monitoring period, the output power of the medium-frequency electrotherapy instrument is monitored in real time, and a fault period sequence and a fault period interval sequence are obtained by comparison.

[0008] Based on the fault period sequence, it is judged whether the fault period is stable. If it is not stable and the fault period presents a single gradual change trend, the fault period is determined based on the fault period sequence, multiple types of inducing parameter sequences are obtained by integrating multiple types of inducing parameters monitored in real time in the fault period.

[0009] Extract the cause parameter sequence with a single gradual trend, and respectively associate with the failure cycle sequence to determine the high probability cause parameter sequence.

[0010] Fitting the high probability cause parameter sequence and the failure cycle sequence obtains a cause parameter-failure cycle model.

[0011] Based on the failure cycle sequence and the failure cycle interval sequence, the failure cycle and the failure cycle interval are predicted to determine multiple predicted failure periods in the subsequent operation cycle.

[0012] Integrating the high probability cause parameter in the monitoring cycle, the high probability cause parameter in each predicted failure period is predicted, the cause parameter-failure cycle model is combined, the cause prediction failure cycle is output, and the power monitoring interval of the intermediate frequency electrotherapy instrument in the subsequent operation cycle is dynamically determined by combining the predicted failure period cycle and the predicted failure cycle interval.

[0013] As a further technical solution of the application, the process of obtaining the failure cycle sequence and the failure cycle interval sequence is:

[0014] The output power of the intermediate frequency electrotherapy instrument is monitored in real time in the monitoring period in the operation cycle of the intermediate frequency electrotherapy instrument, and compared with the preset power.

[0015] The starting time when the output power exceeds the preset power is marked as the failure starting point, the ending time when the output power returns to below the preset power is marked as the failure ending point, and the period length between the failure starting point and the failure ending point is the failure cycle. The failure cycles in the monitoring period are integrated in time sequence as the failure cycle sequence.

[0016] The interval period length between adjacent failure cycles is marked as the failure cycle interval, and the failure cycle intervals in the monitoring cycle are integrated in time sequence as the failure cycle interval sequence.

[0017] As a further technical solution of the application, the process of determining whether the failure cycle is stable is:

[0018] Based on the failure cycle sequence, the variation coefficient of the failure cycle sequence is calculated.

[0019] If the variation coefficient is greater than or equal to the variation coefficient threshold, it indicates that the failure cycle is unstable.

[0020] The process of determining whether the failure cycle presents a single gradual trend is:

[0021] If the failure cycle is unstable and the failure cycle sequence presents a continuous downward or upward trend, it indicates that the failure cycle is unstable and presents a single gradual trend.

[0022] As a further technical solution of the present application, the process of obtaining multiple different types of inducement parameter sequences is:

[0023] Mark the time period corresponding to each failure period in the failure period sequence as a failure period;

[0024] Extract multiple types of inducement parameters monitored in real time in each failure period, based on any one failure period and any one type of inducement parameter, the inducement parameters in the failure period are respectively processed by mean value and integrated in time sequence as inducement parameter sequence, and multiple different types of inducement parameter sequences are obtained based on multiple types of inducement parameters.

[0025] As a further technical solution of the present application, the process of determining the high-probability inducement parameter sequence is:

[0026] Extract the inducement parameter sequence with the same single gradual trend, based on any inducement parameter sequence;

[0027] Analyze the inducement parameter sequence and the failure period sequence to obtain a first inducement correlation value and a second inducement correlation value;

[0028] Sum the absolute values of the first inducement correlation value and the second inducement correlation value to obtain a failure inducement correlation value, and select the inducement parameter sequence corresponding to the maximum failure inducement correlation value as the high-probability inducement parameter sequence.

[0029] As a further technical solution of the present application, the acquisition method of the first inducement correlation value and the second inducement correlation value is:

[0030] After standardizing the inducement parameter sequence and the failure period sequence, calculate the Pearson correlation coefficient to obtain the first inducement correlation value;

[0031] Based on the inducement parameter sequence, calculate the inducement parameter change rate between each two adjacent inducement parameters in the inducement parameter sequence, and integrate the obtained inducement parameter change rate into an inducement parameter change rate sequence in time sequence;

[0032] Based on the failure period sequence, calculate the failure period change rate between each two adjacent failure periods in the failure period sequence, and integrate the obtained failure period change rate into a failure period change rate sequence in time sequence;

[0033] After standardizing the inducement parameter change rate sequence and the failure period change rate sequence, calculate the Pearson correlation coefficient to obtain the second inducement correlation value.

[0034] As a further technical solution of the present application, the inducement parameter-failure period model is obtained by fitting the high-probability inducement parameter in the high-probability inducement parameter sequence and the failure period in the failure period sequence by the least square method.

[0035] As a further technical solution of the present application, the process of determining a plurality of predicted failure time periods in the subsequent operation period is:

[0036] A moving window is set, and the moving window is slid on the sequence of failure periods or the sequence of failure period intervals to calculate the average value in the moving window to realize the prediction of the failure period and the failure period interval;

[0037] Based on the predicted failure period interval, the last failure time period in the monitoring period is obtained, the end time point of the failure time period is summed with the failure period interval to obtain the start time point of the predicted failure time period;

[0038] The start time point of the predicted failure time period is summed with the predicted failure period to obtain the end time point of the predicted failure time period, and the predicted failure time period is determined according to the start time point and the end time point of the predicted failure time period.

[0039] As a further technical solution of the present application, the process of dynamically determining the power monitoring interval of the intermediate frequency electrotherapy instrument in the subsequent operation period is:

[0040] The high-probability inducement parameters in the monitoring period are integrated according to the time sequence, and the high-probability inducement parameters are predicted by combining the moving average method, and the high-probability inducement parameters in each predicted failure time period are determined according to the predicted high-probability inducement parameter corresponding time point;

[0041] The high-probability inducement parameters in the predicted failure time period are input into the inducement parameter-failure period model after being averaged, and the inducement predicted failure period is output.

[0042] The inducement predicted failure period output by the inducement parameter-failure period model each time is summed with the predicted failure time period to obtain the power monitoring interval, and the power monitoring interval is dynamically adjusted after each predicted failure period interval.

[0043] A remote fault diagnosis method system for intermediate frequency electrotherapy instrument, comprising the following modules:

[0044] Data collection and arrangement module: in a preset monitoring period, the output power of the intermediate frequency electrotherapy instrument is monitored in real time, and the sequence of failure periods and the sequence of failure period intervals are obtained by comparison.

[0045] Induced data integration module: based on the sequence of failure periods, it is judged whether the failure period is stable, if not stable and the failure period presents a single gradual change trend, the failure time period is determined based on the sequence of failure periods, and a plurality of different types of inducement parameter sequences are obtained by integrating a plurality of types of inducement parameters monitored in real time in the failure time period.

[0046] Inducing data determination module: extract the inducing parameter sequence with single gradual change trend, and respectively perform correlation change analysis with the fault period sequence to determine the high-probability inducing parameter sequence.

[0047] Correlation model construction module: fit the high-probability inducing parameter sequence and the fault period sequence to obtain an inducing parameter-fault period model.

[0048] Fault period prediction module: perform prediction of the fault period and the fault period interval based on the fault period sequence and the fault period interval sequence to determine multiple predicted fault periods in the subsequent operation period.

[0049] Dynamic monitoring adjustment module: integrate the high-probability inducing parameters in the monitoring period, predict the high-probability inducing parameters in each predicted fault period, combine the inducing parameter-fault period model to output the inducing predicted fault period, and combine the predicted fault period interval and the predicted fault period interval to dynamically determine the power monitoring interval of the intermediate frequency electrotherapy instrument in the subsequent operation period.

[0050] The beneficial effects of the present application are as follows: in the preset monitoring period, the output power of the intermediate frequency electrotherapy instrument is monitored in real time, the fault period sequence and the fault period interval sequence are obtained by comparison, whether the fault period is stable is judged based on the fault period sequence, if not, and the fault period presents a single gradual change trend, the fault period is determined based on the fault period sequence, multiple types of inducing parameter sequences are obtained by integrating the multiple types of inducing parameters monitored in real time in the fault period, the inducing parameter sequence with single gradual change trend is extracted, correlation change analysis is performed with the fault period sequence respectively to determine the high-probability inducing parameter sequence, the high-probability inducing parameter sequence and the fault period sequence are fitted to obtain an inducing parameter-fault period model, prediction of the fault period and the fault period interval is performed based on the fault period sequence and the fault period interval sequence to determine multiple predicted fault periods in the subsequent operation period, the high-probability inducing parameters in the monitoring period are integrated, the high-probability inducing parameters in each predicted fault period are predicted, the inducing parameter-fault period model is combined to output the inducing predicted fault period, and the power monitoring interval of the intermediate frequency electrotherapy instrument in the subsequent operation period is dynamically determined by combining the predicted fault period interval and the predicted fault period interval. The present application mainly solves the problems of missing detection of the gradual change period fault of the intermediate frequency electrotherapy instrument by fixed collection interval and the problem that fixed collection interval may cause monitoring data redundancy to affect the fault diagnosis of the intermediate frequency electrotherapy instrument. BRIEF DESCRIPTION OF DRAWINGS

[0051] The present application will be further described below with reference to the accompanying drawings.

[0052] Fig. 1 is a step flow chart of a method for remote fault diagnosis of an intermediate frequency electrotherapy instrument according to an embodiment of the present application.

[0053] Fig. 2 Figure 1 is a logic diagram of a method for remote fault diagnosis of a medium-frequency electrotherapy instrument according to an embodiment of the present application. DETAILED DESCRIPTION

[0054] In order to make the technical means, creative features, purposes and effects of the present application easy to understand, the present application is further described below in conjunction with specific embodiments.

[0055] Embodiment 1: Please refer to Figure 1, which shows a method for remote fault diagnosis of a medium-frequency electrotherapy instrument according to an embodiment of the present application, including the following steps: Figs. 1-2

[0056] Step 1: In a preset monitoring period, the output power of the medium-frequency electrotherapy instrument is monitored in real time, and the fault period sequence and the fault period interval sequence are obtained by comparison.

[0057] In step 1, the process of obtaining the fault period sequence and the fault period interval sequence is as follows:

[0058] In the monitoring period in the running period of the medium-frequency electrotherapy instrument, the output power of the medium-frequency electrotherapy instrument is monitored in real time, and compared with the preset power.

[0059] The starting time when the output power exceeds the preset power is marked as the fault starting point, the ending time when the output power returns to below the preset power is marked as the fault ending point, and the period length between the fault starting point and the fault ending point is the fault period. The fault periods in the monitoring period are integrated in time sequence as the fault period sequence.

[0060] The interval period length between adjacent fault periods is marked as the fault period interval, and the fault period intervals in the monitoring period are integrated in time sequence as the fault period interval sequence.

[0061] It can be understood that the role of step 1 is to: through real-time monitoring of the output power of the medium-frequency electrotherapy instrument, the abnormal power period (fault period) and the normal interval between the abnormal periods (fault period interval) are converted into structured sequence data, providing the fault period sequence for step 2 for judging whether the fault period is stable and the trend type; providing the fault period sequence and the fault period interval sequence for step 5 as the original data for predicting the subsequent fault period; the time sequence information of the fault period provides the time alignment basis for the correlation analysis of the cause parameter sequence and the fault period sequence in step 3 (ensuring that the cause parameters match the corresponding fault period).

[0062] ​Step two: based on the sequence of failure periods, determine whether the failure period is stable, if not stable and the failure period presents a single gradual trend, then based on the sequence of failure periods, determine the failure period, integrate the real-time monitored multi-type cause parameters in the failure period to obtain a plurality of different types of cause parameter sequences.

[0063] In step two, the process of determining whether the failure period is stable is:

[0064] Based on the sequence of failure periods, calculate the coefficient of variation of the sequence of failure periods.

[0065] If the coefficient of variation is greater than or equal to the coefficient of variation threshold, it indicates that the failure period is not stable.

[0066] If the coefficient of variation is less than the coefficient of variation threshold, it indicates that the failure period is stable.

[0067] It should be noted that if the failure period is stable, the post-failure period sequence is processed by averaging to obtain the power monitoring interval of the intermediate frequency electrotherapy instrument in the subsequent operation period.

[0068] In step two, the process of determining whether the failure period presents a single gradual trend is:

[0069] If the failure period is not stable, and the sequence of failure periods presents a continuous downward or upward trend, it indicates that the failure period is not stable and presents a single gradual trend.

[0070] In step two, the process of obtaining a plurality of different types of cause parameter sequences is:

[0071] Mark the time period corresponding to each failure period in the sequence of failure periods as the failure period.

[0072] Extract the real-time monitored multi-type cause parameters in each failure period, based on any one failure period and any one type of cause parameter, the cause parameters in the failure period are respectively processed by averaging and integrated in time sequence to obtain a cause parameter sequence, and based on the multi-type cause parameters, a plurality of different types of cause parameter sequences are obtained.

[0073] It should be noted that the cause parameter represents the inducing factor parameter of the intermediate frequency electrotherapy instrument in the historical past operation process, for example, the cause parameter includes but is not limited to filter capacitor capacity, voltage, and feedback resistance, etc.

[0074] It can be understood that the role of step two is to determine the stability of the failure period through the coefficient of variation, and then focus on the failure mode with unstable and single gradual trend (such failure mode is more likely to be caused by continuous influence of specific inducement), and extract the inducement parameter sequence of the corresponding period, to provide the single gradual trend failure period sequence and the corresponding inducement parameter sequence for step three, which is the prerequisite for correlation analysis; if the failure period is stable, the power monitoring interval is obtained through the mean value processing, which is complementary to the dynamic adjustment of the monitoring interval in step six.

[0075] Step three: extract the inducement parameter sequence with the same single gradual trend, and respectively perform correlation change analysis with the failure period sequence to determine the high probability inducement parameter sequence.

[0076] In step three, the process of extracting the inducement parameter sequence with the same single gradual trend is as follows:

[0077] Extract the inducement parameter sequence with the same single gradual trend, wherein the inducement parameter sequence with the same single gradual trend indicates that the inducement parameters in the inducement parameter sequence also present a continuous downward or upward trend.

[0078] In step three, the process of determining the high probability inducement parameter sequence is as follows:

[0079] Based on any inducement parameter sequence;

[0080] After standardizing the inducement parameter sequence and the failure period sequence, calculate the Pearson correlation coefficient to obtain a first inducement correlation value.

[0081] Based on the inducement parameter sequence, calculate the inducement parameter change rate between each two adjacent inducement parameters in the inducement parameter sequence, and integrate the obtained inducement parameter change rate into an inducement parameter change rate sequence according to the time sequence.

[0082] Wherein, the inducement parameter change rate YC i -YC i-1 , wherein YC i represents the i-th inducement parameter, and YC i-1 represents the i-1-th inducement parameter.

[0083] Based on the failure period sequence, calculate the failure period change rate between each two adjacent failure periods in the failure period sequence, and integrate the obtained failure period change rate into a failure period change rate sequence according to the time sequence.

[0084] Wherein, the failure period change rate GZ i -GZ i-1 , wherein GZ i represents the i-th failure period, and GZ i-1 represents the i-1-th failure period.

[0085] The Pearson correlation coefficient is calculated after the cause parameter change rate sequence and the fault cycle change rate sequence are standardized, and a second induced correlation value is obtained.

[0086] The first induced correlation value and the second induced correlation value are summed after being absolute valued, and a fault induced correlation value is obtained.

[0087] The cause parameter sequence corresponding to the maximum fault induced correlation value is selected as the high-probability cause parameter sequence.

[0088] It can be understood that the role of step three is to accurately locate the most likely cause of the fault by screening the cause parameters with the same trend as the fault cycle and calculating the double correlation value (parameter value correlation, change rate correlation), and to evaluate the correlation between the cause and the fault from two dimensions of static value and dynamic change rate through the Pearson correlation coefficient, to provide the high-probability cause parameter sequence for step four as the input of the cause-fault model construction; the result of the correlation analysis (such as the high-probability cause being the filter capacitor capacity) provides a clear object for the cause prediction of step six, ensuring that the prediction direction is focused.

[0089] Step four: fitting the high-probability cause parameter sequence and the fault cycle sequence to obtain a cause parameter-fault cycle model.

[0090] In step four, the cause parameter-fault cycle model is obtained by fitting the high-probability cause parameters in the high-probability cause parameter sequence and the fault cycles in the fault cycle sequence through the least squares method.

[0091] It can be understood that the role of step four is to fit the relationship between the high-probability cause and the fault cycle through the least squares method, to convert the correlation between the cause and the fault into a quantifiable model relationship, to provide the calculation basis for the cause prediction of step six, and to make the dynamic monitoring interval adjustment of step six more reasonable.

[0092] Step five: predicting the fault cycle and the fault cycle interval based on the fault cycle sequence and the fault cycle interval sequence to determine a plurality of predicted fault periods in the subsequent operation cycle.

[0093] It should be noted that the monitoring cycle is set in the operation cycle, and the monitoring cycle length is less than the operation cycle length. The subsequent cycle in the operation cycle other than the monitoring cycle is the subsequent operation cycle.

[0094] In step five, the process of determining a plurality of predicted fault periods in the subsequent operation cycle is:

[0095] based on the fault cycle sequence or the fault cycle interval sequence;

[0096] A1, set a moving window, wherein the moving window is set in such a manner that a plurality of windows (such as 3-10) are tested, the mean square error (MSE) of the moving average and the original data in each window is calculated, and the window with the minimum mean square error (MSE) is selected as the moving window (indicating that the smoothing effect of the window is most consistent with the true trend of the data).

[0097] A2, slide the moving window on the sequence of fault periods or the sequence of fault period intervals, calculate the average value in the moving window to realize the prediction of the fault period and the fault period interval.

[0098] Based on the predicted fault period interval, the last fault period in the monitoring period is obtained, the end time point of the fault period is summed with the fault period interval to obtain the start time point of the predicted fault period.

[0099] The start time point of the predicted fault period is summed with the predicted fault period to obtain the end time point of the predicted fault period, and the start time point and the end time point of the predicted fault period are determined to determine the predicted fault period in the subsequent operation period.

[0100] It can be understood that the function of step five is to predict the fault period and interval based on the moving average method, determine the specific period in which the fault may occur in the future, and provide a time range for the cause parameter prediction and monitoring interval adjustment in step six.

[0101] Step 6: Integrate high-probability cause parameters in the monitoring period, predict high-probability cause parameters in each predicted fault period, combine the cause parameter-fault period model, output the cause prediction fault period, and dynamically determine the power monitoring interval of the medium frequency electrotherapy instrument in the subsequent operation period in combination with the predicted fault period and the predicted fault period interval.

[0102] In step 6, the process of predicting high-probability cause parameters in each predicted fault period is as follows:

[0103] Integrate the high-probability cause parameters in the monitoring period in time sequence, and predict the high-probability cause parameters in combination with the moving average method, determine the high-probability cause parameters in each predicted fault period according to the corresponding time point of the predicted high-probability cause parameters.

[0104] It should be noted that the prediction of the high-probability cause parameters in combination with the moving average method is the same as the prediction of the fault period and the fault period interval in step 5 above, which will not be described here.

[0105] In step 6, the process of outputting the cause prediction fault period is as follows:

[0106] The high-probability cause parameter in the predicted failure period is input into the cause parameter-failure period model after being averaged, and a cause predicted failure period is output.

[0107] In step six, the process of dynamically determining the power monitoring interval of the intermediate frequency electrotherapy instrument in the subsequent operation period is:

[0108] The cause predicted failure period output by the cause parameter-failure period model each time is summed and averaged with the predicted failure period cycle to obtain the power monitoring interval, and the power monitoring interval is dynamically adjusted after each predicted failure period interval.

[0109] For example, the predicted failure period intervals are 5 minutes, 6 minutes, 8 minutes and 10 minutes respectively, and the power monitoring interval is dynamically adjusted at intervals of 5 minutes, 6 minutes, 8 minutes and 10 minutes respectively after the end of the last failure period in the monitoring period.

[0110] It can be understood that the function of step six is to dynamically adjust the monitoring interval in combination with the cause prediction and the failure period model, so as to achieve the balance between accurate monitoring and resource optimization.

[0111] The technical scheme of the embodiment of the application is: in a preset monitoring period, the output power of the intermediate frequency electrotherapy instrument is monitored in real time, the failure period sequence and the failure period interval sequence are obtained by comparison, whether the failure period is stable is judged based on the failure period sequence, if it is not stable and the failure period presents a single gradual change trend, the failure period is determined based on the failure period sequence, multiple types of cause parameter sequences are obtained by integrating multiple types of cause parameters monitored in real time in the failure period, the cause parameter sequences with the same single gradual change trend are extracted and are respectively associated with the failure period sequence for change analysis, the high-probability cause parameter sequence is determined, the cause parameter-failure period model is obtained by fitting the high-probability cause parameter sequence and the failure period sequence, the failure period and the failure period interval are predicted based on the failure period sequence and the failure period interval sequence, multiple predicted failure periods in the subsequent operation period are determined, the high-probability cause parameter in each predicted failure period is predicted, the cause predicted failure period is output in combination with the cause parameter-failure period model, and the power monitoring interval of the intermediate frequency electrotherapy instrument in the subsequent operation period is dynamically determined in combination with the predicted failure period cycle and the predicted failure period interval. The application mainly solves the problems of missing the gradual change period failure of the intermediate frequency electrotherapy instrument caused by fixed collection intervals and the monitoring data redundancy caused by fixed collection intervals affecting the failure diagnosis of the intermediate frequency electrotherapy instrument.

[0112] Embodiment 2: A remote failure diagnosis system for an intermediate frequency electrotherapy instrument according to the embodiment of the application comprises the following modules:

[0113] Data collection and arrangement module: in the preset monitoring period, the output power of the intermediate frequency electrotherapy instrument is monitored in real time, and the fault period sequence and the fault period interval sequence are obtained by comparison.

[0114] Induced data integration module: based on the fault period sequence, it is judged whether the fault period is stable, if not stable and the fault period presents a single gradual change trend, the fault period is determined based on the fault period sequence, and the multiple types of induced parameter sequences are obtained by integrating the multiple types of induced parameters monitored in real time in the fault period.

[0115] Induced data determination module: extract the induced parameter sequences with a single gradual change trend, and respectively perform correlation change analysis with the fault period sequence to determine the high-probability induced parameter sequence.

[0116] Correlation model construction module: fitting the high-probability induced parameter sequence and the fault period sequence to obtain the induced parameter-fault period model.

[0117] Fault period prediction module: based on the fault period sequence and the fault period interval sequence, the fault period and the fault period interval are predicted, and multiple predicted fault periods in the subsequent running period are determined.

[0118] Dynamic monitoring and adjustment module: integrating the high-probability induced parameters in the monitoring period, predicting the high-probability induced parameters in each predicted fault period, combining the induced parameter-fault period model, outputting the induced predicted fault period, and combining the predicted fault period and the predicted fault period interval, dynamically determining the power monitoring interval of the intermediate frequency electrotherapy instrument in the subsequent running period.

[0119] The above shows and describes the basic principles, main features and advantages of the present application. It should be understood by those skilled in the art that the present application is not limited to the above embodiments, and the above embodiments and descriptions in the specification are only to illustrate the principles of the present application. Without departing from the spirit and scope of the present application, various changes and improvements can be made to the present application, and these changes and improvements all fall within the scope of the claimed present application. The scope of protection of the present application is defined by the appended claims and their equivalents.

Claims

1. A method for remote fault diagnosis of a medium-frequency electrotherapy device, characterized in that: Comprise: In a preset monitoring period, by real-time monitoring of the output power of the intermediate frequency electrotherapy instrument, the failure cycle sequence and the failure cycle interval sequence are obtained by comparison; Based on the failure cycle sequence, it is judged whether the failure cycle is stable, if not stable and the failure cycle presents a single gradual change trend, the failure period is determined based on the failure cycle sequence, and the multiple types of inducement parameter sequences are obtained by integrating the multiple types of inducement parameters monitored in real time in the failure period; Extract the inducement parameter sequences with the same single gradual change trend, and respectively analyze the change with the failure cycle sequence to determine the high probability inducement parameter sequence; The inducement parameter-failure cycle model is obtained by fitting the high probability inducement parameter sequence and the failure cycle sequence; Based on the failure cycle sequence and the failure cycle interval sequence, the failure cycle and the failure cycle interval are predicted, and multiple predicted failure periods in the subsequent running period are determined; Integrate the high probability inducement parameters in the monitoring period, predict the high probability inducement parameters in each predicted failure period, combine the inducement parameter-failure cycle model, output the inducement prediction failure cycle, and dynamically determine the power monitoring interval of the intermediate frequency electrotherapy instrument in the subsequent running period combined with the predicted failure period and the predicted failure cycle interval; Wherein, the process of obtaining the failure cycle sequence and the failure cycle interval sequence is: In the intermediate frequency electrotherapy instrument running period, a preset monitoring period is set, the output power of the intermediate frequency electrotherapy instrument is monitored in real time, and compared with the preset power; The starting time of the output power exceeding the preset power is marked as the failure starting point, the ending time of the output power returning to below the preset power is marked as the failure ending point, and the period length between the failure starting point and the failure ending point is the failure cycle. The failure cycles in the monitoring period are integrated in time sequence as the failure cycle sequence; The interval period length between adjacent failure cycles is marked as the failure cycle interval, and the failure cycle intervals in the monitoring period are integrated in time sequence as the failure cycle interval sequence; The process of judging whether the failure cycle is stable is: Based on the failure cycle sequence, the coefficient of variation of the failure cycle sequence is calculated; If the coefficient of variation is greater than or equal to the coefficient of variation threshold, it indicates that the failure cycle is unstable; The process of judging whether the failure cycle presents a single gradual change trend is: If the failure cycle is unstable, and the failure cycle sequence presents a continuous downward or continuous upward trend, it indicates that the failure cycle is unstable and presents a single gradual change trend; The process of obtaining multiple different types of inducement parameter sequences is: Mark the time period corresponding to each failure cycle in the failure cycle sequence as the failure period; Extract multiple types of inducement parameters monitored in real time in each failure period, based on any one failure period and any one type of inducement parameter, the inducement parameters in the failure period are respectively processed by mean value and integrated in time sequence as the inducement parameter sequence, and multiple different types of inducement parameter sequences are obtained based on multiple types of inducement parameters; The process of determining the high probability inducement parameter sequence is: Extract the inducement parameter sequences with the same single gradual change trend, based on any inducement parameter sequence; Analyze the inducement parameter sequence and the failure cycle sequence to obtain a first inducement correlation value and a second inducement correlation value; The first induced correlation value and the second induced correlation value are summed after being absolute valued to obtain a fault induced correlation value, and a cause parameter sequence corresponding to the maximum fault induced correlation value is selected as a high-probability cause parameter sequence; The first induced correlation value and the second induced correlation value are obtained in the following manner: After the cause parameter sequence and the fault period sequence are standardized, a Pearson correlation coefficient is calculated to obtain the first induced correlation value; Based on the cause parameter sequence, a cause parameter change rate between each two adjacent cause parameters in the cause parameter sequence is calculated, and the obtained cause parameter change rates are integrated in time sequence to obtain a cause parameter change rate sequence; Based on the fault period sequence, a fault period change rate between each two adjacent fault periods in the fault period sequence is calculated, and the obtained fault period change rates are integrated in time sequence to obtain a fault period change rate sequence; After the cause parameter change rate sequence and the fault period change rate sequence are standardized, a Pearson correlation coefficient is calculated to obtain the second induced correlation value.

2. The method according to claim 1, wherein: The cause parameter-fault period model is obtained by fitting the high-probability cause parameters in the high-probability cause parameter sequence and the fault periods in the fault period sequence by the least square method.

3. The method according to claim 1, wherein: The process of determining a plurality of predicted fault periods in the subsequent operation period is as follows: A moving window is set, the moving window is slid on the fault period sequence or the fault period interval sequence, and the average value in the moving window is calculated to realize the prediction of the fault period and the fault period interval; Based on the predicted fault period interval, the last fault period in the monitoring period is obtained, the end time point of the fault period is summed with the fault period interval to obtain the start time point of the predicted fault period; The start time point of the predicted fault period is summed with the predicted fault period to obtain the end time point of the predicted fault period, and the predicted fault period is determined according to the start time point and the end time point of the predicted fault period.

4. The method according to claim 1, wherein: The process of dynamically determining the power monitoring interval of the intermediate frequency electrotherapy instrument in the subsequent operation period is as follows: The high-probability cause parameters in the monitoring period are integrated in time sequence, and the high-probability cause parameters are predicted by the moving average method, and the high-probability cause parameters in each predicted fault period are determined according to the time point corresponding to the predicted high-probability cause parameters; The high-probability cause parameters in the predicted fault period are averaged and input into the cause parameter-fault period model to output the cause predicted fault period; The cause predicted fault period output by the cause parameter-fault period model each time is summed with the predicted fault period interval to obtain the power monitoring interval, and the power monitoring interval is dynamically adjusted after each predicted fault period interval.

5. A method for remote fault diagnosis of intermediate frequency electrotherapy apparatus, characterized in that: The method comprises the following modules: The data collection and arrangement module: in the preset monitoring period, the output power of the intermediate frequency electrotherapy instrument is monitored in real time, and the fault cycle sequence and the fault cycle interval sequence are obtained by comparison; The induced data integration module: based on the fault cycle sequence, it is judged whether the fault cycle is stable, if not stable and the fault cycle presents a single gradual change trend, the fault period is determined based on the fault cycle sequence, the multiple types of induced parameter sequences are obtained by integrating the multiple types of induced parameters monitored in real time in the fault period. The induced data determination module: extract the induced parameter sequences with single gradual change trend, and respectively analyze the correlation change with the fault cycle sequence to determine the high probability induced parameter sequence. The correlation model construction module: fitting the high probability induced parameter sequence and the fault cycle sequence to obtain the induced parameter-fault cycle model. The fault period prediction module: based on the fault cycle sequence and the fault cycle interval sequence, the fault cycle and the fault cycle interval are predicted to determine multiple predicted fault periods in the subsequent running period. The dynamic monitoring adjustment module: integrate the high probability induced parameters in the monitoring period, predict the high probability induced parameters in each predicted fault period, combine the induced parameter-fault cycle model, output the induced predicted fault cycle, and combine the predicted fault period and the predicted fault cycle interval to dynamically determine the power monitoring interval of the intermediate frequency electrotherapy instrument in the subsequent running period. The process of obtaining the fault cycle sequence and the fault cycle interval sequence is as follows: In the running period of the intermediate frequency electrotherapy instrument, a preset monitoring period is set, the output power of the intermediate frequency electrotherapy instrument is monitored in real time, and the output power is compared with the preset power. The starting time of the output power exceeding the preset power is marked as the fault starting point, the ending time of the output power returning to below the preset power is marked as the fault ending point, and the period length between the fault starting point and the fault ending point is the fault cycle. The fault cycles in the monitoring period are integrated in sequence as the fault cycle sequence. The interval period length between adjacent fault cycles is marked as the fault cycle interval, and the fault cycle intervals in the monitoring period are integrated in sequence as the fault cycle interval sequence. The process of judging whether the fault cycle is stable is as follows: Based on the fault cycle sequence, the coefficient of variation of the fault cycle sequence is calculated. If the coefficient of variation is greater than or equal to the coefficient of variation threshold, it indicates that the fault cycle is unstable. The process of judging whether the fault cycle presents a single gradual change trend is as follows: If the fault cycle is unstable, and the fault cycle sequence presents a continuous downward or upward trend, it indicates that the fault cycle is unstable and presents a single gradual change trend. The process of obtaining multiple different types of induced parameter sequences is as follows: The time period corresponding to each fault cycle in the fault cycle sequence is marked as the fault period. Extract the multiple types of induced parameters monitored in real time in each fault period, based on any one fault period and any one type of induced parameter, the induced parameters in the fault period are respectively processed by mean value and integrated in sequence as the induced parameter sequence, and based on multiple types of induced parameters, multiple different types of induced parameter sequences are obtained. The process of determining the high probability induced parameter sequence is as follows: Extracting the cause parameter sequence as a single gradual trend, based on any cause parameter sequence; Analyzing the cause parameter sequence and the failure period sequence to obtain a first induced correlation value and a second induced correlation value; Summing the absolute values of the first induced correlation value and the second induced correlation value to obtain a failure induced correlation value, and selecting the cause parameter sequence corresponding to the maximum failure induced correlation value as a high-probability cause parameter sequence; The first induced correlation value and the second induced correlation value are obtained in the following manner: Standardizing the cause parameter sequence and the failure period sequence and then calculating the Pearson correlation coefficient to obtain the first induced correlation value; Based on the cause parameter sequence, calculating the cause parameter change rate between each two adjacent cause parameters in the cause parameter sequence, and integrating the obtained cause parameter change rates in time sequence into a cause parameter change rate sequence; Based on the failure period sequence, calculating the failure period change rate between each two adjacent failure periods in the failure period sequence, and integrating the obtained failure period change rates in time sequence into a failure period change rate sequence; Standardizing the cause parameter change rate sequence and the failure period change rate sequence and then calculating the Pearson correlation coefficient to obtain the second induced correlation value.

Images