Stirrer fault monitoring method and system

Abstract

The invention provides a stirrer fault monitoring method and system, and belongs to the technical field of fault monitoring, and the method comprises the steps: carrying out the installation of a sensor group for a stirrer according to the equipment type of the stirrer; acquiring first data generated in the operation process of the stirrer based on the installed sensor group, and performing data cleaning on the first data to obtain second data; analyzing all the second data, determining initial fault coefficients of the stirrer based on different sensor types, and comparing the initial fault coefficients with standard fault coefficients under corresponding sensor types; and acquiring a comprehensive fault coefficient of the stirrer based on a comparison result, and performing comparative analysis on the comprehensive fault coefficient and a coefficient-early warning comparison table in combination with the initial fault coefficient so as to perform early warning on the current fault of the stirrer. Manual fault judgment is effectively avoided, real-time fault early warning is effectively achieved, then it is guaranteed that the stirrer is in a normal working state continuously and stably, and the experiment efficiency is effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of fault monitoring, and in particular to a method and system for monitoring faults of an agitator. Background Art

[0002] Laboratory blenders are versatile devices used for various mixing, grinding and homogenizing processes in scientific research and development. They are essential for preparing analytical samples, mixing different materials and performing chemical reactions. There are many types of laboratory blenders, each designed for a specific application. These devices are essential to ensuring the uniformity and quality of materials used in experiments and production processes, and therefore it is particularly important to judge abnormalities in blenders.

[0003] The common way to determine whether the agitator is abnormal is to do it manually. For example, when the agitator is in the starting state but does not run, the agitator will be manually determined to be faulty. This judgment method is too rough, that is, the agitator will only be judged to be faulty if it is obviously not running, which undoubtedly greatly reduces the efficiency of the experiment.

[0004] Therefore, the present invention proposes a stirrer fault monitoring method and system. Summary of the invention

[0005] The present invention provides a stirrer fault monitoring method and system, which are used to monitor the operation process of the stirrer by setting a sensor group, and intelligently determine the fault coefficient through data cleaning and data analysis, and then realize fault determination and early warning through comparative analysis of a reference table, effectively avoid human fault judgment, effectively realize real-time fault warning, and thus ensure that the stirrer is continuously and stably in a normal working state, effectively improving the experimental efficiency.

[0006] The present invention provides a stirrer fault monitoring method, comprising: Step 1: Install a sensor group on the agitator according to the device type of the agitator; Step 2: Based on the installed sensor group, first data generated by the agitator during operation are collected respectively, and the first data are cleaned to obtain second data; Step 3: Analyze all the second data to determine the initial failure coefficient of the agitator based on different sensor types, and compare it with the standard failure coefficient under the corresponding sensor type; Step 4: Based on the comparison result, the comprehensive fault coefficient of the agitator is obtained, and combined with the initial fault coefficient, a comparative analysis is performed with the coefficient-warning comparison table to issue a warning for the current fault of the agitator.

[0007] Preferably, according to the equipment type of the stirrer, installing the sensor group on the stirrer includes: Obtain a sensor deployment diagram matching the device type from a type-monitoring comparison table; The sensor group is installed on the agitator according to the sensor deployment diagram.

[0008] Preferably, performing data cleaning on the first data to obtain the second data includes: Extract several historical maintenance information of the agitator consistent with the equipment type from the historical database and construct a historical maintenance matrix ,in, A historical maintenance vector constructed by representing the first historical maintenance information of the agitator of the corresponding equipment type; A historical maintenance vector constructed by representing the nth historical maintenance information of the agitator of the corresponding equipment type; Lock the historical maintenance matrix The number of occurrences of non-zero parameters in each row vector in , and the row maintenance error is calculated; ; in, represents the maintenance error of row i1; represents the sum of maintenance weights for non-zero parameters in the historical maintenance vector of row i1; represents the sum of the precision loss for non-zero parameters in the historical maintenance vector of row i1; represents the number of occurrences of non-zero parameters in the historical maintenance vector of row i1; ln represents the sign of the logarithmic function; Indicates the allowable error conversion factor of the maintenance type corresponding to the i1th historical maintenance information; Represents the historical maintenance matrix Each row in involves the total number of maintenance parameters; Lock the historical maintenance matrix The number of occurrences of non-zero parameters in each column vector in , and the column maintenance error is calculated; ; in, represents the maintenance error of column i2; Represents the sum of maintenance weights for non-zero parameters in the i2th column vector; represents the sum of the precision loss for non-zero parameters in the i2th column vector; Represents the number of occurrences of non-zero parameters in the i2th column vector; It represents the allowable error conversion factor of the maintenance type corresponding to the row vector of the maximum weight extracted from the maintenance weights of different historical maintenance information involving the non-zero parameters in the i2th column; Represents the sum of the maintenance weights of all parameters in the i2th column vector; Extract the historical maintenance matrix The maximum value in each row in , construct the first maintenance parameter array, and extract the historical maintenance matrix The maintenance parameter in each row in the maintenance parameter array has the largest maintenance weight, and a second maintenance parameter array is constructed; Determining main maintenance parameters based on the first maintenance parameter array and the second maintenance parameter array; Locking the column vector corresponding to each main maintenance parameter, calculating the maintenance mean and maintenance variance of the corresponding column vector, and combining all row maintenance errors and column maintenance errors to determine the cleaning accuracy of the first data; ; in, Indicates the corresponding cleaning accuracy; Indicates the number of main maintenance parameters; represents the maintenance mean of the column vector corresponding to the j1th main maintenance parameter; represents the maintenance variance of the column vector corresponding to the j1th main maintenance parameter; A cleaning method matching the cleaning accuracy is selected from a cleaning database to perform cleaning to obtain second data.

[0009] Preferably, determining the main maintenance parameters based on the first maintenance parameter array and the second maintenance parameter array includes: Perform frequency statistics and parameter value sum calculation on the same maintenance parameter in the first maintenance parameter array to construct a first parameter function corresponding to the maintenance parameter ,in, Indicates the statistical frequency of the corresponding maintenance parameters; represents the sum of all parameter values ​​involved in the first maintenance parameter array for the corresponding maintenance parameter; considering the maintenance parameter corresponding to the first parameter function satisfying the first main criterion as the first maintenance parameter; The frequency of the non-zero and zero cases of the same maintenance parameter in the second maintenance parameter array is counted, and the weight and ratio of the non-zero and zero cases of the same maintenance parameter are recorded to construct a second parameter function corresponding to the maintenance parameter. ,in, Indicates the statistical frequency of non-zero conditions of the corresponding maintenance parameters; Indicates the statistical frequency of the 0 situation of the corresponding maintenance parameter; Indicates based on The weights and sums of the maintenance parameters involved; Indicates based on The weights and sums of the maintenance parameters involved; considering the maintenance parameter corresponding to the second parameter function satisfying the second main criterion as the second maintenance parameter; The first maintenance parameter and the second maintenance parameter are combined to obtain a main maintenance parameter.

[0010] Preferably, all the second data are analyzed to determine the initial failure coefficient of the agitator based on different sensor types, including: Extracting sub-data of the same sensor type from all the second data respectively, and drawing a single curve of the same sensor type according to the arrangement position sequence of the sensor group on the stirrer; Comparing the single curve with a corresponding standard curve range, and calculating a corresponding initial failure coefficient; ; in, is the corresponding initial failure coefficient; is the upper boundary excess range area of ​​the single curve determined based on the comparison result; is the lower boundary excess range area of ​​the single curve determined based on the comparison result; is the enclosed area corresponding to the standard curve range; is the length of the curve that is within the range of the corresponding standard curve; is the total length of the corresponding standard curve range; exp represents the exponential function symbol.

[0011] Preferably, obtaining the comprehensive failure coefficient of the agitator based on the comparison result includes: Compare each initial fault coefficient with the corresponding standard fault coefficient to determine the coefficient deviation; Calculate the variance of all coefficient deviations; Counting the weight array of the main maintenance parameters in the corresponding column vector, and selecting the maximum weight from the weight array and calculating the average weight, and then averaging again to obtain the initial weight; According to max (initial weight, set weight), the required weight of the sensor type corresponding to the main maintenance parameter is obtained; Redistribute the weights based on all the required weights to get the current weight of the corresponding sensor type.

[0012] The comprehensive failure coefficient is obtained by relying on the current weight and coefficient deviation and combining the variance.

[0013] Preferably, a comparative analysis is performed with a coefficient-warning comparison table to warn of the current fault of the agitator, including: Based on the comprehensive fault coefficient and all initial fault coefficients, a coefficient array is constructed; Comparing the coefficient array with each result in the coefficient-warning comparison table in turn; According to the consistent comparison result, the current fault and warning information of the agitator is obtained, and a warning is issued.

[0014] The present invention provides a stirrer fault monitoring system, comprising: An installation module, used for installing a sensor group on the agitator according to the device type of the agitator; A data cleaning module, used for respectively collecting first data generated by the agitator during operation based on the installed sensor group, and performing data cleaning on the first data to obtain second data; A coefficient determination module, used for analyzing all the second data, determining the initial fault coefficient of the agitator based on different sensor types, and comparing it with the standard fault coefficient under the corresponding sensor type; The early warning analysis module is used to obtain the comprehensive fault coefficient of the agitator based on the comparison result, and to compare and analyze it with the coefficient-early warning comparison table in combination with the initial fault coefficient to issue an early warning for the current fault of the agitator.

[0015] Compared with the prior art, the present invention has the following beneficial effects: The operation process of the agitator is monitored by setting up a sensor group, and the fault coefficient is intelligently determined through data cleaning and data analysis. Then, through comparative analysis of the reference table, fault determination and early warning are realized, which effectively avoids human judgment of faults and effectively realizes real-time fault warning, thereby ensuring that the agitator is continuously and stably in normal working condition, effectively improving experimental efficiency.

[0016] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0017] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1 is a flow chart of a method for monitoring agitator failure in an embodiment of the present invention; Figure 2 A structural diagram of a stirrer fault monitoring system according to an embodiment of the present invention; Figure 3 It is a structural diagram of a single curve and a corresponding standard curve range in an embodiment of the present invention. DETAILED DESCRIPTION

[0020] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0021] The present invention provides a method for monitoring agitator failure. Figure 1 As shown, including: Step 1: Install a sensor group on the agitator according to the device type of the agitator; Step 2: Based on the installed sensor group, first data generated by the agitator during operation are collected respectively, and the first data are cleaned to obtain second data; Step 3: Analyze all the second data to determine the initial failure coefficient of the agitator based on different sensor types, and compare it with the standard failure coefficient under the corresponding sensor type; Step 4: Based on the comparison result, the comprehensive fault coefficient of the agitator is obtained, and combined with the initial fault coefficient, a comparative analysis is performed with the coefficient-warning comparison table to issue a warning for the current fault of the agitator.

[0022] In this embodiment, the equipment type refers to the type of agitator, such as paddle type, stirring tank type, spiral type, high shear type, horizontal type, vertical type and screw type, and the parameters measured corresponding to different types are different. The sensor group includes a combination of one or more temperature sensors, humidity sensors, vibration sensors, voltage meters, current meters, etc. It should be noted that the sensor groups on the same agitator are the same. At the same time, the sensors in the sensor group are intelligent sensors.

[0023] In this embodiment, the first data refers to the temperature, humidity, vibration amplitude, voltage, current, etc. of the measured part of the agitator. Since the sensor may have abnormal measurement during the measurement process, such as missing measurement data, abnormal measurement data, etc., these abnormal conditions are eliminated to obtain the second data. In this embodiment, the same sensor type refers to a sensor for temperature measurement type, a sensor for humidity measurement type, and so on.

[0024] In this embodiment, the initial fault coefficient is the calculation result for the same sensor type, and the standard fault coefficient is pre-set. In this embodiment, the comprehensive fault coefficient is calculated based on different initial fault coefficients and the weights of the corresponding sensor types. In this embodiment, the coefficient-early warning comparison table includes different coefficient arrays and fault information and early warning methods corresponding to the arrays, so the corresponding faults and early warnings can be directly obtained. The beneficial effects of the above technical solution are: by setting a sensor group to monitor the operation process of the agitator, and by intelligently determining the fault coefficient through data cleaning and data analysis, and then by comparative analysis of the comparison table, fault determination and early warning are realized, effectively avoiding human judgment of faults, and effectively realizing real-time fault early warning, thereby ensuring that the agitator is continuously and stably in normal working state, and effectively improving experimental efficiency.

[0025] The present invention provides a method for monitoring a stirrer fault, and according to the equipment type of the stirrer, a sensor group is installed on the stirrer, comprising: Obtain a sensor deployment diagram matching the device type from a type-monitoring comparison table; The sensor group is installed on the agitator according to the sensor deployment diagram.

[0026] In this embodiment, the type-monitoring comparison table includes different equipment types of agitators and a sensor deployment diagram corresponding to the type, and the deployment diagram includes the installation positions of the sensors in the agitator, for example, sensor groups are respectively installed at parts A1 and B2 of the agitator.

[0027] The beneficial effect of the above technical solution is that matching the sensor deployment diagram from the comparison table facilitates the effective installation of the sensor group, providing a basis for subsequent monitoring of the agitator and determining the existing faults.

[0028] The present invention provides a method for monitoring agitator faults, wherein the first data is cleaned to obtain second data, comprising: Extract several historical maintenance information of the agitator consistent with the equipment type from the historical database and construct a historical maintenance matrix ,in, A historical maintenance vector constructed by representing the first historical maintenance information of the agitator of the corresponding equipment type; A historical maintenance vector constructed by representing the nth historical maintenance information of the agitator of the corresponding equipment type; Lock the historical maintenance matrix The number of occurrences of non-zero parameters in each row vector in , and the row maintenance error is calculated; ; in, represents the maintenance error of row i1; represents the sum of maintenance weights for non-zero parameters in the historical maintenance vector of row i1; represents the sum of the precision loss for non-zero parameters in the historical maintenance vector of row i1; represents the number of occurrences of non-zero parameters in the historical maintenance vector of row i1; ln represents the sign of the logarithmic function; Indicates the allowable error conversion factor of the maintenance type corresponding to the i1th historical maintenance information; Represents the historical maintenance matrix Each row in involves the total number of maintenance parameters; Lock the historical maintenance matrix The number of occurrences of non-zero parameters in each column vector in , and the column maintenance error is calculated; ; in, represents the maintenance error of column i2; Represents the sum of maintenance weights for non-zero parameters in the i2th column vector; represents the sum of the precision loss for non-zero parameters in the i2th column vector; Represents the number of occurrences of non-zero parameters in the i2th column vector; It represents the allowable error conversion factor of the maintenance type corresponding to the row vector of the maximum weight extracted from the maintenance weights of different historical maintenance information involving the non-zero parameters in the i2th column; Represents the sum of the maintenance weights of all parameters in the i2th column vector; Extract the historical maintenance matrix The maximum value in each row in , construct the first maintenance parameter array, and extract the historical maintenance matrix The maintenance parameter in each row in the maintenance parameter array has the largest maintenance weight, and a second maintenance parameter array is constructed; Determining main maintenance parameters based on the first maintenance parameter array and the second maintenance parameter array; Locking the column vector corresponding to each main maintenance parameter, calculating the maintenance mean and maintenance variance of the corresponding column vector, and combining all row maintenance errors and column maintenance errors to determine the cleaning accuracy of the first data; ; in, Indicates the corresponding cleaning accuracy; Indicates the number of main maintenance parameters; represents the maintenance mean of the column vector corresponding to the j1th main maintenance parameter; represents the maintenance variance of the column vector corresponding to the j1th main maintenance parameter; A cleaning method matching the cleaning accuracy is selected from a cleaning database to perform cleaning to obtain second data.

[0029] In this embodiment, the historical database includes the historical maintenance conditions of different agitators, which are all recorded in advance. It should be noted that since the faults repaired each time may be different, the maintenance parameters involved are also different and the values ​​of the parameters involved are also different. However, in order to facilitate the calculation and statistics of all possible fault parameters and standardize the values ​​of the fault parameters involved in the historical maintenance information, a historical maintenance matrix is ​​constructed, and the weight of each maintenance parameter in each row vector is determined in advance based on the corresponding maintenance condition.

[0030] In this embodiment, the non-zero parameter refers to the case where the value of the parameter in the matrix is ​​not 0.

[0031] In this embodiment, the first maintenance parameter array = { The maximum value in each row of}.

[0032] In this embodiment, the second maintenance parameter array = { The value with the largest maintenance weight for the maintenance parameter in each row.

[0033] In this embodiment, for example, the locked parameter A1 corresponds to the second column. At this time, the maintenance mean of the second column is determined, which is the average value of all values ​​involved in the column.

[0034] In this embodiment, the cleaning database includes cleaning accuracy and cleaning methods matching the cleaning accuracy, and the cleaning methods include deletion method, replacement method, interpolation method, etc.

[0035] The beneficial effects of the above technical solution are: extracting historical maintenance information from the historical database to construct a matrix, and then calculating the row maintenance error and the column maintenance error, and determining the main maintenance parameters through maximum value extraction and maximum weight extraction, and obtaining the cleaning accuracy through comprehensive processing from two aspects, making it convenient to match the cleaning method to achieve data cleaning, providing an accurate basis for subsequent fault analysis and early warning, and facilitating timely repair of agitator faults.

[0036] The present invention provides a method for monitoring agitator faults, which determines main maintenance parameters based on the first maintenance parameter array and the second maintenance parameter array, including: Perform frequency statistics and parameter value sum calculation on the same maintenance parameter in the first maintenance parameter array to construct a first parameter function corresponding to the maintenance parameter ,in, Indicates the statistical frequency of the corresponding maintenance parameters; represents the sum of all parameter values ​​involved in the first maintenance parameter array for the corresponding maintenance parameter; considering the maintenance parameter corresponding to the first parameter function satisfying the first main criterion as the first maintenance parameter; The frequency of the non-zero and zero cases of the same maintenance parameter in the second maintenance parameter array is counted, and the weight and ratio of the non-zero and zero cases of the same maintenance parameter are recorded to construct a second parameter function corresponding to the maintenance parameter. ,in, Indicates the statistical frequency of non-zero conditions of the corresponding maintenance parameters; Indicates the statistical frequency of the 0 situation of the corresponding maintenance parameter; Indicates based on The weights and sums of the maintenance parameters involved; Indicates based on The weights and sums of the maintenance parameters involved; considering the maintenance parameter corresponding to the second parameter function satisfying the second main criterion as the second maintenance parameter; The first maintenance parameter and the second maintenance parameter are combined to obtain a main maintenance parameter.

[0037] In this embodiment, the first maintenance parameter array = {a1-0.1 b1-0.2 a1-0.6 b2-0.3}. At this time, the statistical frequency of the maintenance parameter a1 is 2, and the sum of the parameter values ​​is 0.1+0.6=0.7, and so on.

[0038] In this embodiment, the first main criterion refers to that the statistical frequency needs to be greater than or equal to 2, and the parameter value sum needs to be greater than or equal to 0.5. In this case, a1 is regarded as the first maintenance parameter.

[0039] In this embodiment, the second maintenance parameter array = {a1-0.1 b2-0.6 a1-0.6 b1-0.2 a1-0}. At this time, the statistical frequency of the maintenance parameter a1 when it is not zero is 2, and the statistical frequency when it is zero is 1. The weight sum ratio is: the weight sum of the maintenance parameter a1 when it is not zero / (the weight sum of the maintenance parameter a1 when it is not zero + the weight sum of the maintenance parameter a1 when it is zero), and so on.

[0040] In this embodiment, the second main criterion refers to the statistical frequency of the non-zero situation of the same maintenance parameter being greater than or equal to the statistical frequency of the zero situation. , at this time, a1 is regarded as the second maintenance parameter.

[0041] At this time, the main maintenance parameter obtained after the union processing is a1.

[0042] The beneficial effect of the above technical solution is: by analyzing the first maintenance parameter array and the second maintenance parameter array respectively, the corresponding parameter functions are constructed respectively, and then by comparing and analyzing with the standard, the main maintenance parameters are effectively obtained, providing an analytical basis for determining the cleaning method.

[0043] The present invention provides a stirrer fault monitoring method, which analyzes all second data to determine the initial fault coefficient of the stirrer based on different sensor types, including: Extracting sub-data of the same sensor type from all the second data respectively, and drawing a single curve of the same sensor type according to the arrangement position sequence of the sensor group on the stirrer; Comparing the single curve with a corresponding standard curve range, and calculating a corresponding initial failure coefficient; ; in, is the corresponding initial failure coefficient; is the upper boundary excess range area of ​​the single curve determined based on the comparison result; is the lower boundary excess range area of ​​the single curve determined based on the comparison result; is the enclosed area corresponding to the standard curve range; is the length of the curve that is within the range of the corresponding standard curve; is the total length of the corresponding standard curve range; exp represents the exponential function symbol.

[0044] In this embodiment, if Figure 3 As shown, A1 is a single curve, the standard curve range is obtained by drawing B1 and B2, the upper boundary exceeds the range refers to the area where the single curve exceeds the upper boundary of the standard curve range, that is, C1, and the lower boundary exceeds the range refers to the area where the single curve exceeds the lower boundary of the table conversion curve range, that is, C2.

[0045] In this embodiment, the closed area of ​​the standard curve range is the area of ​​the region corresponding to the standard curve range.

[0046] The beneficial effect of the above technical solution is that by comparing a single curve with the corresponding standard curve range, the initial fault coefficient is effectively obtained, providing an analysis basis for subsequent early warning.

[0047] The present invention provides a stirrer fault monitoring method, which obtains a comprehensive fault coefficient of the stirrer based on a comparison result, comprising: Compare each initial fault coefficient with the corresponding standard fault coefficient to determine the coefficient deviation; Calculate the variance of all coefficient deviations; Counting the weight array of the main maintenance parameters in the corresponding column vector, and selecting the maximum weight from the weight array and calculating the average weight, and then averaging again to obtain the initial weight; According to max (initial weight, set weight), the required weight of the sensor type corresponding to the main maintenance parameter is obtained; Redistribute the weights based on all the required weights to get the current weight of the corresponding sensor type.

[0048] The comprehensive failure coefficient is obtained by relying on the current weight and coefficient deviation and combining the variance.

[0049] In this embodiment, the coefficient deviation is the absolute value of the difference between the initial fault coefficient and the standard fault coefficient.

[0050] In this embodiment, the variance is calculated based on the variance formula, which is common knowledge.

[0051] In this embodiment, the weight array refers to the maintenance weight of each parameter in the column vector corresponding to the main maintenance parameters, and the average weight refers to the average value of all weights in the array.

[0052] In this embodiment, the initial weight obtained by averaging again = (maximum weight + average weight) / 2.

[0053] In this embodiment, the current weight = the required weight of the corresponding sensor type / the sum of the required weights of all sensor types. It should be noted that the required weight for the sensor type that is not the main maintenance parameter is the average value of the maintenance weights of the column vectors involved in the corresponding maintenance parameters in the matrix.

[0054] In this embodiment, the comprehensive fault coefficient=current weight×sum of coefficient deviations+variance.

[0055] The beneficial effect of the above technical solution is: by comparing the coefficients to determine the deviation and calculate the variance, and combining the weight array to comprehensively obtain the current weight, and then calculate the comprehensive fault coefficient, providing a basis for comparative analysis.

[0056] The present invention provides a stirrer fault monitoring method, which performs comparative analysis with a coefficient-warning comparison table to warn of the current fault of the stirrer, including: Based on the comprehensive fault coefficient and all initial fault coefficients, a coefficient array is constructed; Comparing the coefficient array with each result in the coefficient-warning comparison table in turn; According to the consistent comparison result, the current fault and warning information of the agitator is obtained, and a warning is issued.

[0057] In this embodiment, the coefficient array = {all initial fault coefficients combined fault coefficients}.

[0058] In this embodiment, the coefficient-warning comparison table includes different coefficient arrays and warning results matching the arrays, and the warning results include faults and warnings.

[0059] Agitator failures include: insufficient voltage, missing phase of motor coil or power cable, excessive operating temperature, impeller damage, etc.

[0060] The warning can be a text display result, or it can be a corresponding warning for different faults, making it unique and convenient to directly identify the fault of the agitator.

[0061] The beneficial effect of the above technical solution is that by comparing and analyzing the coefficient array and the comparison table, it is easy to obtain faults and early warnings.

[0062] The present invention provides a stirrer fault monitoring system, such as Figure 2 As shown, including: An installation module, used for installing a sensor group on the agitator according to the device type of the agitator; A data cleaning module, used for respectively collecting first data generated by the agitator during operation based on the installed sensor group, and performing data cleaning on the first data to obtain second data; A coefficient determination module, used for analyzing all the second data, determining the initial fault coefficient of the agitator based on different sensor types, and comparing it with the standard fault coefficient under the corresponding sensor type; The early warning analysis module is used to obtain the comprehensive fault coefficient of the agitator based on the comparison result, and to compare and analyze it with the coefficient-early warning comparison table in combination with the initial fault coefficient to issue an early warning for the current fault of the agitator.

[0063] The beneficial effects of the above technical solution are: by setting up a sensor group to monitor the operation process of the agitator, and through data cleaning and data analysis to intelligently determine the fault coefficient, and then through comparative analysis of the reference table, fault determination and early warning are realized, effectively avoiding human fault judgment, and effectively realizing real-time fault warning, thereby ensuring that the agitator is continuously and stably in normal working condition, effectively improving experimental efficiency.

[0064] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for monitoring agitator failure, characterized in that: include: Step 1: Install a sensor group on the agitator according to the device type of the agitator; Step 2: Based on the installed sensor group, first data generated by the agitator during operation are collected respectively, and the first data are cleaned to obtain second data; Step 3: Analyze all the second data to determine the initial failure coefficient of the agitator based on different sensor types, and compare it with the standard failure coefficient under the corresponding sensor type; Step 4: Based on the comparison result, the comprehensive fault coefficient of the agitator is obtained, and combined with the initial fault coefficient, a comparative analysis is performed with the coefficient-warning comparison table to issue a warning for the current fault of the agitator.

2. The stirrer fault monitoring method according to claim 1, characterized in that: According to the equipment type of the agitator, the sensor group is installed on the agitator, including: Obtain a sensor deployment diagram matching the device type from a type-monitoring comparison table; The sensor group is installed on the agitator according to the sensor deployment diagram.

3. The stirrer fault monitoring method according to claim 1, characterized in that: Cleaning the first data to obtain second data includes: Extract several historical maintenance information of the agitator consistent with the equipment type from the historical database and construct a historical maintenance matrix ,in, A historical maintenance vector constructed by representing the first historical maintenance information of the agitator of the corresponding equipment type; A historical maintenance vector constructed by representing the nth historical maintenance information of the agitator of the corresponding equipment type; Lock the historical maintenance matrix The number of occurrences of non-zero parameters in each row vector in , and the row maintenance error is calculated; ； in, represents the maintenance error of row i1; represents the sum of maintenance weights for non-zero parameters in the historical maintenance vector of row i1; represents the sum of the precision loss for non-zero parameters in the historical maintenance vector of row i1; represents the number of occurrences of non-zero parameters in the historical maintenance vector of row i1; ln represents the sign of the logarithmic function; Indicates the allowable error conversion factor of the maintenance type corresponding to the i1th historical maintenance information; Represents the historical maintenance matrix Each row in involves the total number of maintenance parameters; Lock the historical maintenance matrix The number of occurrences of non-zero parameters in each column vector in , and the column maintenance error is calculated; ； in, represents the maintenance error of column i2; Represents the sum of maintenance weights for non-zero parameters in the i2th column vector; represents the sum of the precision loss for non-zero parameters in the i2th column vector; Represents the number of occurrences of non-zero parameters in the i2th column vector; It represents the allowable error conversion factor of the maintenance type corresponding to the row vector of the maximum weight extracted from the maintenance weights of different historical maintenance information involving the non-zero parameters in the i2th column; Represents the sum of the maintenance weights of all parameters in the i2th column vector; Extract the historical maintenance matrix The maximum value in each row in , construct the first maintenance parameter array, and extract the historical maintenance matrix The maintenance parameter in each row in the maintenance parameter array has the largest maintenance weight, and a second maintenance parameter array is constructed; Determining main maintenance parameters based on the first maintenance parameter array and the second maintenance parameter array; Locking the column vector corresponding to each main maintenance parameter, calculating the maintenance mean and maintenance variance of the corresponding column vector, and combining all row maintenance errors and column maintenance errors to determine the cleaning accuracy of the first data; ； in, Indicates the corresponding cleaning accuracy; Indicates the number of main maintenance parameters; represents the maintenance mean of the column vector corresponding to the j1th main maintenance parameter; represents the maintenance variance of the column vector corresponding to the j1th main maintenance parameter; A cleaning method matching the cleaning accuracy is selected from a cleaning database to perform cleaning to obtain second data.

4. The stirrer fault monitoring method according to claim 1, characterized in that: Determining main maintenance parameters based on the first maintenance parameter array and the second maintenance parameter array includes: Perform frequency statistics and parameter value sum calculation on the same maintenance parameter in the first maintenance parameter array to construct a first parameter function corresponding to the maintenance parameter ,in, Indicates the statistical frequency of the corresponding maintenance parameters; represents the sum of all parameter values ​​involved in the first maintenance parameter array for the corresponding maintenance parameter; considering the maintenance parameter corresponding to the first parameter function satisfying the first main criterion as the first maintenance parameter; The frequency of the non-zero and zero cases of the same maintenance parameter in the second maintenance parameter array is counted, and the weight and ratio of the non-zero and zero cases of the same maintenance parameter are recorded to construct a second parameter function corresponding to the maintenance parameter. ,in, Indicates the statistical frequency of non-zero conditions of the corresponding maintenance parameters; Indicates the statistical frequency of the 0 situation of the corresponding maintenance parameter; Indicates based on The weights and sums of the maintenance parameters involved; Indicates based on The weights and sums of the maintenance parameters involved; considering the maintenance parameter corresponding to the second parameter function satisfying the second main criterion as the second maintenance parameter; The first maintenance parameter and the second maintenance parameter are combined to obtain a main maintenance parameter.

5. The stirrer fault monitoring method according to claim 1, characterized in that: All the second data are analyzed to determine the initial failure coefficient of the agitator based on different sensor types, including: Extracting sub-data of the same sensor type from all the second data respectively, and drawing a single curve of the same sensor type according to the arrangement position sequence of the sensor group on the stirrer; Comparing the single curve with a corresponding standard curve range, and calculating a corresponding initial failure coefficient; ； in, is the corresponding initial failure coefficient; is the upper boundary excess range area of ​​the single curve determined based on the comparison result; is the lower boundary excess range area of ​​the single curve determined based on the comparison result; is the enclosed area corresponding to the standard curve range; is the length of the curve that is within the range of the corresponding standard curve; is the total length of the corresponding standard curve range; exp represents the exponential function symbol.

6. The stirrer fault monitoring method according to claim 4, characterized in that: Obtaining a comprehensive failure coefficient of the agitator based on the comparison result includes: Compare each initial fault coefficient with the corresponding standard fault coefficient to determine the coefficient deviation; Calculate the variance of all coefficient deviations; Counting the weight array of the main maintenance parameters in the corresponding column vector, and selecting the maximum weight from the weight array and calculating the average weight, and then averaging again to obtain the initial weight; According to max (initial weight, set weight), the required weight of the sensor type corresponding to the main maintenance parameter is obtained; Redistribute the weights based on all the required weights to get the current weight of the corresponding sensor type; The comprehensive failure coefficient is obtained by relying on the current weight and coefficient deviation and combining the variance.

7. The stirrer fault monitoring method according to claim 1, characterized in that: Comparative analysis is performed with the coefficient-warning comparison table to warn of the current fault of the agitator, including: Based on the comprehensive fault coefficient and all initial fault coefficients, a coefficient array is constructed; Comparing the coefficient array with each result in the coefficient-warning comparison table in turn; According to the consistent comparison result, the current fault and warning information of the agitator is obtained, and a warning is issued.

8. A stirrer fault monitoring system, characterized in that: include: An installation module, used for installing a sensor group on the agitator according to the device type of the agitator; A data cleaning module, used for respectively collecting first data generated by the agitator during operation based on the installed sensor group, and performing data cleaning on the first data to obtain second data; A coefficient determination module, used for analyzing all the second data, determining the initial fault coefficient of the agitator based on different sensor types, and comparing it with the standard fault coefficient under the corresponding sensor type; The early warning analysis module is used to obtain the comprehensive fault coefficient of the agitator based on the comparison result, and to compare and analyze it with the coefficient-early warning comparison table in combination with the initial fault coefficient to issue an early warning for the current fault of the agitator.

Images