A water pump data processing system and method based on a tunnel inclined shaft

Abstract

A water pump data processing system and method based on a tunnel inclined shaft, belonging to the technical field of data processing. It starts by reducing the key operation values, grasping the floating state of the value changes, and performing standardization processing through the reduction ratio, making the detection process more sensitive and more adaptable. Subsequently, through leveling processing and balanced operation, a separation point critical quantity for efficiently detecting outliers is obtained, thereby accurately locking the outliers in the key values. The advantage of this method is its good adaptability, compatibility, and accuracy, being suitable for several types of values, and simultaneously ensuring good detection performance even in the case of chaotic numerical distributions.

Description

Technical Field

[0001] The present invention belongs to the technical field of data processing, and particularly relates to a water pump data processing system and method based on a tunnel inclined shaft. Background Art

[0002] During the construction of a long tunnel, in order to shorten the construction period, a vertical shaft or an inclined shaft is set up to increase the working face, and some set up parallel adits, that is, the setting of vertical shafts or inclined shafts.

[0003] In the application of tunnel inclined shafts, as often mentioned in the prior art solution with the patent publication number "CN217001965U", the tunnel inclined shaft is arranged on the reverse slope section, and a water pump is arranged in the reverse slope section for drainage.

[0004] It can be seen that the water pump is an important tool to ensure the normal application of the tunnel inclined shaft and the reverse slope section. Therefore, key values of the water pump arranged in the reverse slope section need to be sampled to implement corresponding monitoring functions. Currently, generally, sensors are used to sample the key values of the water pump in the reverse slope section and transmit them to the controller. Then, the controller determines the condition of the water pump based on the key values of the water pump in the reverse slope section transmitted by the sampling.

[0005] And determining the condition of the water pump is to determine whether the water pump has a failure condition. Currently, the method for determining whether the water pump has a failure condition mainly depends on whether the key values of the water pump in the reverse slope section are outliers. Currently, the common outlier detection method is based on a predefined constant critical value. This method needs to define different critical values according to the key value characteristics of different water pumps to be suitable for high and low key values. However, because the number and specifications of water pumps in the reverse slope section are complex and the amount of key values is also large, the definition rules of the critical value are often very complex. Moreover, when this constant critical value method faces a scenario with large numerical fluctuations, the detection accuracy is insufficient. In addition, the current method also has the defect of poor applicability, that is, the outlier detection performance varies greatly for different numerical categories or numerical distributions. Summary of the Invention

[0006] To solve the defects in the prior art, the present invention proposes a water pump data processing system and method based on a tunnel inclined shaft, effectively avoiding the defects of complex definition rules of critical values, insufficient detection accuracy, and poor applicability of the constant critical value method in the prior art when facing a scenario with large numerical fluctuations.

[0007] The present invention adopts the following technical solutions.

[0008] A water pump data processing method based on a tunnel inclined shaft includes:

[0009] The sensor samples the key values of the water pump in the reverse slope section and transmits them to the controller. The controller determines the condition of the water pump based on the key values of the water pump in the reverse slope section transmitted through sampling.

[0010] Determining the condition of the water pump is to determine whether the water pump has a failure condition.

[0011] The method for determining whether the water pump has a failure condition includes:

[0012] Step 1: Receive the source key values of the water pump, remove the useless values, and perform sorting in ascending order of the key values to obtain a key value queue.

[0013] Step 2: Calculate the difference between the next key value and the previous key value in the key value queue in turn to obtain a difference queue, and calculate the ratio of each difference in the total sum of all differences respectively to obtain a difference ratio queue.

[0014] Step 3: Calculate the ratio of the standard deviation of the difference of the differences to the total sum of the standard deviations of all differences in turn to obtain a standard deviation ratio queue of differences.

[0015] Step 4: Multiply the key value queue starting from the first key value by each standard deviation ratio of the standard deviation ratio queue of differences in turn to obtain a starting point even separation point critical quantity. Multiply the key value queue starting from the second key value by each standard deviation ratio of the standard deviation ratio queue of differences in turn to obtain an ending point even separation point critical quantity.

[0016] Step 5: Calculate the average value of the starting point even separation point critical quantity and the ending point even separation point critical quantity to obtain an outlier separation point critical quantity.

[0017] Step 6: Obtain an outlier value queue by comparing the outlier separation point critical quantity with the key values. If there is no outlier value in the outlier value queue, it is determined that the water pump has no failure condition. If there is an outlier value in the outlier queue, it is determined that the corresponding water pump has a failure condition, and the controller transmits the outlier value and the message that the corresponding water pump has a failure condition to the liquid crystal screen for display.

[0018] Further, in Step 1, the source key values of the water pump are the key values of the water pump in the reverse slope section transmitted through sampling; the key values of the water pump in the reverse slope section are the rotational speed value of the water pump in the reverse slope section, the working current value of the water pump in the reverse slope section, or the active power value of the water pump in the reverse slope section; the sensor is a rotational speed sensor, a current sensor, or an active power sensor provided on the water pump in the reverse slope section.

[0019] Further, in Step 1, receiving the source key values of the water pump is to receive the key values of the water pump transmitted through sampling during the set time interval when the water pump is working.

[0020] Further, in step 1, the useless values include NULL values, redundant values, illegal values, and set period values.

[0021] Further, in step 2, the decrement has the following operation equation: ; here, is the key value queue, is the sequence code of the key value in the key value queue, is the th key value in the key value queue, is the th key value in the key value queue, is the th decrement in the decrement queue.

[0022] Further, in step 2, the total decrement value has the following operation equation: ; here, is the number of key values in the key value queue.

[0023] Further, in step 2, the decrement ratio has the following operation equation: , is the decrement ratio of the th decrement in the decrement queue.

[0024] Further, in step 3, the decrement standard deviation has the following operation equation: , is the standard deviation of the th decrement in the decrement queue.

[0025] Further, in step 3, the total decrement standard deviation value has the following operation equation: .

[0026] Further, in step 3, the decrement standard deviation ratio has the following operation equation: , is the ratio of the th decrement in the total decrement standard deviation value of the decrement standard deviation ratio queue, that is, the th decrement standard deviation ratio in the decrement standard deviation ratio queue.

[0027] Further, in step 4, the starting point flat split point critical quantity has the following operation equation: .

[0028] Further, in step 4, the critical quantity of the end point flush with the separation point has an operation equation as follows: .

[0029] Further, in step 5, the critical quantity of the outlier separation point has an operation equation as follows: .

[0030] Further, in step 6, clustering is performed according to the critical quantity of the outlier separation point being lower than or not lower than. The number of key values in each cluster is totaled, and it is confirmed that the cluster with a lower number of values within a pair of clusters is an outlier value. Specifically:

[0031] Step 6-1: Treat the key values not lower than the critical quantity of the outlier separation point as cluster queue one;

[0032] Step 6-2: Treat the key values lower than the critical quantity of the outlier separation point as cluster queue two;

[0033] Step 6-3: Treat the cluster queue with a lower number of key values as the outlier value queue.

[0034] A water pump data processing system based on a tunnel inclined shaft, comprising:

[0035] The tunnel inclined shaft is arranged on the reverse slope section. A water pump for drainage is arranged in the reverse slope section. Both the sensor and the liquid crystal screen are connected to the controller. The sensor is used to sample the key values of the water pump in the reverse slope section and transmit them to the controller. The controller is used to determine the condition of the water pump based on the key values of the water pump in the reverse slope section transmitted through sampling;

[0036] The modules running on the controller include:

[0037] The removal module, which is used to receive the source key values of the water pump, remove the useless values, and then arrange them in ascending order of the key values to obtain the key value queue;

[0038] The operation module, which is used to sequentially calculate the difference between the next key value and the previous key value in the key value queue to obtain the difference queue, and respectively calculate the ratio of the difference in the total sum of all differences to obtain the difference ratio queue;

[0039] The ratio module, which is used to sequentially calculate the ratio of the standard deviation of the difference of the difference to its total sum of the standard deviation of all differences to obtain the standard deviation ratio queue of the difference;

[0040] A separation module, which is used to perform multiplication operations on each reduction standard deviation ratio of the critical value queue starting from the first critical value and the reduction standard deviation ratio queue of the critical value queue in turn to obtain the starting point flush separation point critical quantity, and perform multiplication operations on each reduction standard deviation ratio of the critical value queue starting from the second critical value and the reduction standard deviation ratio queue of the critical value queue in turn to obtain the ending point flush separation point critical quantity;

[0041] An average value module, which is used to calculate the average value of the starting point flush separation point critical quantity and the ending point flush separation point critical quantity to obtain the outlier separation point critical quantity;

[0042] An outlier module, which is used to obtain an outlier value queue by comparing the outlier separation point critical quantity and the critical value. If there is no outlier value in the outlier value queue, it is determined that the water pump has no failure condition. If there is an outlier value in the outlier queue, it is determined that the corresponding water pump has a failure condition, and the controller will transmit the outlier value and the message of the corresponding water pump having a failure condition to the liquid crystal screen for display.

[0043] The beneficial effects of the present invention are that, compared with the prior art, the technical effects of the present invention include:

[0044] I. Appropriate value detection: The present invention can actively perform outlier detection according to the specific conditions of the critical values in the water pump, without relying on a predefined constant critical quantity. By automatically calculating the value reduction and reduction standard deviation, it actively forms a matching outlier separation point critical quantity, so as to achieve outlier detection for different value ranges and value categories;

[0045] II. Improved detection accuracy: By carefully analyzing the reduction and reduction standard deviation of the critical value, the present invention can accurately identify outlier values. Especially in the case of complex and diverse value distributions, the detection method of the present invention has higher accuracy and credibility compared with the current critical quantity method, and can effectively reduce the occurrence of misidentification and under-identification;

[0046] III. Enhanced compatibility: The outlier analysis method of the present invention is applicable to various types of critical values. Whether the critical value is high or low, above zero or below zero, it can be effectively processed by this method, so that the controller can be widely used in different water pumps and detection scenarios, with better compatibility;

[0047] IV. Simplified controller setting: The current method often needs to independently define critical quantities for different water pumps or value categories, and the setting is complex and often error-prone. The active analysis controller of the present invention simplifies the setting process of the controller by intelligently calculating the outlier separation point critical quantity, reduces the potential risks of manual intervention and definition errors, and improves the operation efficiency of the controller;

[0048] V. Immediate Numerical Disposal: The present invention can efficiently dispose of a lot of key numerical values formed on the water pump, immediately detect and select outliers, which is conducive to prior notification and prevention of the occurrence of hidden water pump failures. Description of the Drawings

[0049] Figure 1 is a flowchart of the water pump data processing method based on a tunnel inclined shaft described in the present invention;

[0050] Figure 2 is a partial structure diagram of the water pump data processing system based on a tunnel inclined shaft described in the present invention. Detailed Embodiment

[0051] To make the objectives, technical solutions, and advantages of the present invention clearer, the following will clearly and completely describe the technical solutions of the present invention in conjunction with the drawings in the embodiments of the present invention. The embodiments described in this application are only partial embodiments of the present invention, not all embodiments. According to the spirit of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

[0052] As Figure 1 shown, a water pump data processing method based on a tunnel inclined shaft described in the present invention includes:

[0053] The sensor samples the key numerical values of the water pump in the reverse slope section and transmits them to the controller. The controller determines the condition of the water pump based on the key numerical values of the water pump in the reverse slope section transmitted by the sampling.

[0054] Determining the condition of the water pump is to determine whether the water pump has a failure condition;

[0055] The method for determining whether the water pump has a failure condition (running on the controller) includes:

[0056] Step 1: Receive the source key numerical values of the water pump, remove the useless values, and then arrange them in ascending order of the key numerical values to obtain a key numerical value queue;

[0057] In a preferred but non-limiting embodiment of the present invention, in Step 1, the source key numerical values of the water pump are the key numerical values of the water pump in the reverse slope section transmitted by the sampling; the key numerical values of the water pump in the reverse slope section are the rotational speed value of the water pump in the reverse slope section, the working current value of the water pump in the reverse slope section, or the active power value of the water pump in the reverse slope section; the sensor is a rotational speed sensor, a current sensor, or an active power sensor provided on the water pump in the reverse slope section. The key numerical values in the arrangement in ascending order of the key numerical values after removing the useless values are the remaining values after removing the useless values from the source key numerical values.

[0058] In a preferred but non-limiting embodiment of the present invention, in step 1, the source key value of the receiving water pump is the key value of the water pump sampled and transmitted during the set time interval (the set time interval can be 1 h before the current time point, 30 min before the current time point, 15 min before the current time point, or 5 min before the current time point) during the operation of the receiving water pump.

[0059] However, the presence of useless values in the source key value will weaken the performance of the value and is not conducive to the accuracy and reliability of subsequent parsing and processing. To remove the useless values in the source key value, in a preferred but non-limiting embodiment of the present invention, in step 1, the useless values include NULL values, redundant values, illegal values, and set period values. Here, the NULL value, that is, no value, such values cannot perform numerical parsing processing, which is generally caused by incorrect sensor sampling or failed value reading, so it needs to be removed. Redundant values are not conducive to the correctness of the parsed value, so they also need to be removed. Illegal values are values that significantly do not meet the numerical compliance. Just like for key values required to be higher than zero, values lower than zero need to be removed. Just like the rotational speed value of the water pump in the reverse slope section, the working current value of the water pump in the reverse slope section, or the active power value of the water pump in the reverse slope section. The set period value is like the period of maintaining the water pump within the set time interval. The values during the period of maintaining the water pump need to be removed to prevent it from being unfavorable to the accuracy of the operation.

[0060] Step 2: Calculate the decrement of the next key value and the previous key value in the key value queue in sequence to obtain a decrement queue, and calculate the ratio of the decrement in the total sum of all decrements respectively to obtain a decrement ratio queue;

[0061] In a preferred but non-limiting embodiment of the present invention, in step 2, the decrement is calculated by the equation: ; here, is the key value queue, is the sequence code of the key value in the key value queue (the sequence code is a code configured by incrementing one by one in sequence for the key values in the key value queue starting from 1), is the th key value in the key value queue, is the th key value in the key value queue, is the th decrement in the decrement queue. And if the number of key values is , the decrement is obtained by subtracting the th key value from the )th item of the key value. Therefore, the number of decrement values is ones.

[0062] In a preferred but non-limiting embodiment of the present invention, in step 2, the total reduction value has an operation equation as follows: ; where is the number of key values in the key value queue.

[0063] In a preferred but non-limiting embodiment of the present invention, in step 2, the reduction ratio has an operation equation as follows: , is the reduction ratio of the th reduction in the reduction queue.

[0064] Step 3: Calculate the ratio of the reduction standard deviation of each reduction to the total value of the overall reduction standard deviation in sequence to obtain a reduction standard deviation ratio queue;

[0065] In a preferred but non-limiting embodiment of the present invention, in step 3, the reduction standard deviation has an operation equation as follows: , is the standard deviation of the th reduction in the reduction queue.

[0066] In a preferred but non-limiting embodiment of the present invention, in step 3, the total value of the reduction standard deviation has an operation equation as follows: .

[0067] In a preferred but non-limiting embodiment of the present invention, in step 3, the reduction standard deviation ratio has an operation equation as follows: , is the ratio of the th reduction in the overall reduction standard deviation total value in the reduction standard deviation ratio queue, that is, the th reduction standard deviation ratio in the reduction standard deviation ratio queue.

[0068] Step 4: Perform multiplication operations on each reduction standard deviation ratio in the key value queue starting from the first key value and the reduction standard deviation ratio queue in sequence to obtain a starting point flush separation point critical quantity, and perform multiplication operations on each reduction standard deviation ratio in the key value queue starting from the second key value and the reduction standard deviation ratio queue in sequence to obtain an ending point flush separation point critical quantity;

[0069] Just as if the number of key values is , then the number of reduction ratio data and reduction standard deviation ratio is both , it will cause the key value queue, the decrement ratio queue, and the decrement standard deviation ratio queue not to be aligned one-to-one. Therefore, it is necessary to perform alignment operations on the key value queue, the decrement ratio queue, and the decrement standard deviation ratio queue. In this application, two types of alignment methods can be used. One is start point alignment, and the other is end point alignment. If only one type of numerical alignment method is used for calculation, then there will definitely be a highest or lowest value that cannot be calculated. Therefore, this application uses the method in step 5 below to achieve alignment.

[0070] In a preferred but non-limiting embodiment of the present invention, in step 4, the critical quantity of the start point alignment separation point The operation equation is: .

[0071] In a preferred but non-limiting embodiment of the present invention, in step 4, the critical quantity of the end point alignment separation point The operation equation is: .

[0072] Step 5: Calculate the average value of the critical quantity of the start point alignment separation point and the critical quantity of the end point alignment separation point to obtain the outlier separation point critical quantity;

[0073] In a preferred but non-limiting embodiment of the present invention, in step 5, the outlier separation point critical quantity The operation equation is: .

[0074] Step 6: Obtain the outlier value queue by comparing the outlier separation point critical quantity with the key values. If there is no outlier value in the outlier value queue, it is determined that the water pump has no failure condition. If there is an outlier value in the outlier queue, it is determined that the corresponding water pump has a failure condition, and the controller will transmit the outlier value and the message that the corresponding water pump has a failure condition to the liquid crystal screen for display. After obtaining the outlier separation point critical quantity, the relationship between the outlier value and the outlier separation point critical quantity can be predefined according to the specific situation to achieve the detection of the outlier value.

[0075] In a preferred but non-limiting embodiment of the present invention, in step 6, clustering is performed according to being lower than and not lower than the outlier separation point critical quantity, and the number of key values in each cluster is counted. According to the principle that the outlier value is often a lower quantity value, it is confirmed that the cluster of values with a lower number in a pair of clusters is the outlier value. Specifically:

[0076] Step 6-1: Treat the key values not lower than the outlier separation point critical quantity as cluster queue one;

[0077] Step 6-2: Treat the key values lower than the outlier separation point critical quantity as cluster queue two;

[0078] Step 6-3: Treat the cluster queue with a lower number of key values as the outlier value queue.

[0079] In step 6, based on the principle that outliers are often lower-valued, outliers can also be extracted and displayed through a clearer method, specifically as follows:

[0080] Mark the key values in the key value queue on the rectangular coordinate system according to the sampling time points of the key values to form coordinate points. The abscissa of the coordinate point is the sampling time point of the key value, and the ordinate of the coordinate point is the value of the key value;

[0081] Draw a horizontal line with the ordinate being the outlier separation point critical quantity on the rectangular coordinate system;

[0082] Compare the number of coordinate points on both sides of the horizontal line, and take the ordinate of the coordinate points on the side with fewer numbers as the outlier value.

[0083] Non-outlier values and outlier values can also be marked with different colors to make the coordinate points more distinguishable and easier to identify.

[0084] The various equations in the method for determining whether a water pump has a failure condition can be integrated into:

[0085] ;

[0086] Here, is the decrement standard deviation , which represents the variation amplitude of the values. By performing quadratic processing on the decrements of all key values, the discrimination amount is expanded, highlighting the positions and amplitudes of the value discrimination amounts, which is conducive to better locking of outliers.

[0087] is the ratio of the decrement standard deviation to the total value of the decrement standard deviation, which transforms the original value amplitude into a ratio amplitude, that is, performs standardization processing, reflecting the outlier effect amount ratio of the decrements at each value position, which is conducive to calculating the critical quantity of the outlier points later.

[0088] and are the starting point flush separation point critical quantity and the ending point flush separation point critical quantity respectively. Since the decrement value is the quantity obtained by subtracting a pair of adjacent values, the number of decrements will always be one less than the source key value. Both the starting point flush and ending point flush methods need to be taken into account, and the decrement represents the relationship of the quantity obtained by subtracting the next value from the previous value in the queue. When this decrement is not low, the probability that at least one of the next value and the previous value is an outlier is not low. However, at this time, it is impossible to accurately identify which one of them is the outlier. Therefore, construct and To overcome this defect, within the next value is regarded as the previous value within At this time, and the values of respectively represent the amplitudes with outliers when a value is regarded as the previous value and the next value.

[0089] Subsequently, the mean value of the starting point flush separation point critical quantity and the ending point flush separation point critical quantity is obtained, and the outlier separation point critical quantity is obtained. The idea of such mean value processing is to balance the calculated values under the two types of flush methods and ensure the smoothness and credibility of outlier detection.

[0090] This application starts by subtracting the critical value of the key value, grasping the floating situation of the value change, and performing standardization processing through the subtraction ratio, making the detection process more sensitive and more suitable. Subsequently, through flush processing and balanced calculation, a separation point critical quantity that can efficiently detect outliers is obtained, so as to accurately lock the outliers in the key value. The advantage of this method is its good suitability, compatibility and accuracy, which is suitable for several types of values, and at the same time ensures good detection performance even in the case of chaotic value distribution.

[0091] This application provides two methods:

[0092] One method is Method 1. In this method, the controller initially assumes that there are often outliers in the key value. Therefore, it is necessary to fully execute the entire detection process (that is, the method of determining whether the water pump has a failure condition). This method assumes that all values have the probability of hidden outliers. Therefore, regardless of the initial attributes of the values, the controller will perform comprehensive outlier detection according to the pre-defined equations and methods. Steps 1 to 6 elaborate on the complete process of this method. Through step-by-step calculation of the value subtraction, subtraction standard deviation and corresponding separation point critical quantity, and finally determine and send out the outlier queue. This method is suitable for scenarios where the values are very sensitive and it is necessary to ensure that hidden outliers are not missed.

[0093] Another type of method is Method 2. In this method, before the controller executes the complete detection process, it initially performs an initial outlier confirmation on the key values. This initial confirmation aims to quickly evaluate the overall situation of the values and confirm whether they contain outliers. If the initial confirmation value indicates the presence of outliers, the controller then executes the subsequent detection method and then analyzes the values to accurately lock in the outlier points; otherwise, if the initial confirmation indicates that the values have no outliers, the controller can terminate the subsequent detection process according to specific requirements, thereby saving computational effort and processing time. Additionally, the operator can also execute the subsequent process even when the initial confirmation shows no outliers to ensure the comprehensiveness of the detection and detect hidden subtle outlier values. This type of method is particularly suitable for scenarios where the amount of data is large, the computational power is limited, or the requirements for immediacy are relatively high.

[0094] Regarding Method 2, it can be as follows:

[0095] Before step S3 for calculating the reduced standard deviation, it also includes:

[0096] Step 3-1: Set the numerical reduction ratio. The numerical reduction ratio can be set by analyzing the corresponding attributes of historical outliers through a cumulative method to ensure that the set value can accurately reflect the outlier characteristics of the values. The following gives the implementation method of step 3-1:

[0097] Step 3-1-1: Historical value collection: Collect a set amount of historical key values stored in the controller's memory, which includes the detected outliers and non-outliers, to ensure that the collected values are typical and can cover different pumps, different time intervals, and various outlier situations;

[0098] Step 3-1-2: Cumulative analysis: Through the cumulative distribution of historical outliers, determine the variation range and distribution norm of outliers in the case of outliers, that is, achieve it through methods such as histogram analysis and maximum likelihood estimation methods;

[0099] Step 3-1-3: Ratio critical value setting: Based on the above cumulative analysis value, determine an appropriate reduction ratio critical value. This critical value is usually based on the cumulative attributes of the outlier variation and can define the outlier critical value by setting a bit number (such as the first to fifth values in the key value queue) to ensure that the variation within this critical value range is recognized as an outlier;

[0100] Step 3-1-4: Setting value judgment: Apply the set reduction ratio critical value to the new value set for verification. By comparing the detected outliers with the actual situation, determine the usability of this critical value and perform relevant adjustments according to the verification results.

[0101] Step 3-2: Calculate the ratio of the highest value in the operation decrement queue to the total decrement value, and obtain the highest numerical decrement ratio.

[0102] Step 3-3: If the numerical decrement ratio is lower than the highest numerical decrement ratio, then proceed with the subsequent process. Otherwise, abort the process. Taking the numerical decrement ratio as twenty as an example, when the highest numerical decrement ratio is not less than twenty, then execute Step 4 later to find out the specific outlier values. When the highest numerical decrement ratio is lower than twenty, then abort the subsequent process.

[0103] As Figure 2 shown, a water pump data processing system based on a tunnel inclined shaft according to the present invention includes:

[0104] The tunnel inclined shaft is arranged on the reverse slope section. A water pump for drainage is arranged in the reverse slope section. Both the sensor and the liquid crystal display screen are connected to the controller. The sensor is used to sample the key values of the water pump in the reverse slope section and transmit them to the controller. The controller is used to determine the condition of the water pump based on the key values of the water pump in the reverse slope section transmitted through sampling. The reverse slope section can be the first reverse slope section.

[0105] The modules running on the controller include:

[0106] The removal module is used to receive the source key values of the water pump, remove the useless values, and then perform sorting in ascending order of the key values to obtain the key value queue.

[0107] The operation module is used to calculate the decrement of the next key value and the previous key value in the key value queue in sequence to obtain the decrement queue, and calculate the ratio of the decrement to the total decrement value respectively to obtain the decrement ratio queue.

[0108] The ratio module is used to calculate the ratio of the standard deviation of the decrement of the decrement to the total standard deviation of the decrement in sequence to obtain the decrement standard deviation ratio queue.

[0109] The separation module is used to perform multiplication operations on the key value queue starting from the first key value and each decrement standard deviation ratio in the decrement standard deviation ratio queue in sequence to obtain the starting point flush separation point critical quantity, and perform multiplication operations on the key value queue starting from the second key value and each decrement standard deviation ratio in the decrement standard deviation ratio queue in sequence to obtain the ending point flush separation point critical quantity.

[0110] The mean value module is used to calculate the mean value of the starting point flush separation point critical quantity and the ending point flush separation point critical quantity to obtain the outlier separation point critical quantity.

[0111] An outlier module, which is used to obtain an outlier value queue by comparing the control outlier separation point threshold and the key value. If there is no outlier in the outlier value queue, it is determined that the water pump has no failure condition. If there is an outlier in the outlier queue, it is determined that the corresponding water pump has a failure condition, and the controller will transmit the outlier value and the message that the corresponding water pump has a failure condition to the liquid crystal screen for display. The controller can be an industrial control computer, a single-chip microcomputer or a PLC.

[0112] The beneficial effects of the present invention are that, compared with the prior art, the technical effects of the present invention include:

[0113] I. Appropriate value detection: The present invention can actively perform outlier detection according to the specific conditions of the key values in the water pump, without relying on a pre-defined constant threshold. By automatically calculating the value reduction and the reduction standard deviation, the corresponding outlier separation point threshold is automatically formed, so as to achieve outlier detection for different value ranges and value categories;

[0114] II. Improved detection accuracy: By carefully analyzing the value reduction and the reduction standard deviation of the key values, the present invention can accurately identify outliers. Especially in the case of complex and diverse value distributions, the detection method of the present invention is more accurate and reliable than the current threshold method, and can effectively reduce the occurrence of misidentification and under-identification;

[0115] III. Enhanced compatibility: The outlier analysis method of the present invention is applicable to various types of key values. Whether the key value is high or low, above zero or below zero, it can be effectively processed by this method, so that the controller can be widely used in different water pumps and detection scenarios, with better compatibility;

[0116] IV. Simplified controller settings: The current method often needs to independently define thresholds for different water pumps or value categories, and the settings are complex and often error-prone. The active analysis controller of the present invention simplifies the controller setting process by intelligently calculating the outlier separation point threshold, reduces the risk of manual intervention and definition errors, and improves the application efficiency of the controller;

[0117] V. Instantaneous value processing: The present invention can effectively process a large number of key values generated on the water pump, instantaneously detect and select outliers, which is suitable for early warning and preventing the occurrence of hidden water pump failures.

[0118] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that: still can modify or equivalently replace the specific implementation manners of the present invention, and any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the protection scope of the claims of the present invention.

Claims

1. A method for processing water pump data based on tunnel inclined shaft, characterized in that: include: The sensor samples the key values ​​of the water pump in the reverse slope section and transmits them to the controller. The controller determines the status of the water pump by sampling the key values ​​of the water pump in the reverse slope section. To determine the condition of the water pump is to determine whether the water pump has failed; Methods for determining whether a water pump has failed include: Step 1: Receive the source key values ​​of the water pump, remove useless values, and sort them in descending order of the key values ​​to obtain a key value queue; Step 2: sequentially calculate the decrements of the next key value and the previous key value in the key value queue to obtain a decrement queue, and respectively calculate the ratio of the decrements in the total decrement value to obtain a decrement ratio queue; Step 3: Calculate the ratio of the standard deviation of the reduction to the total value of the standard deviation of the reduction in turn, and obtain the standard deviation ratio queue of the reduction; Step 4: Multiply the key value queue starting from the first key value and each reduction standard deviation ratio of the reduction standard deviation ratio queue in turn to obtain the critical amount of the starting point leveling separation point, and multiply the key value queue starting from the second key value and each reduction standard deviation ratio of the reduction standard deviation ratio queue in turn to obtain the critical amount of the ending point leveling separation point; Step 5: Calculate the mean of the critical value of the starting point and the critical value of the ending point to obtain the critical value of the outlier separation point; Step 6: Obtain an outlier queue by comparing the outlier separation point critical quantity and the key value. If there is no outlier in the outlier queue, it is determined that the water pump has no failure. If there is an outlier in the outlier queue, it is determined that the corresponding water pump has a failure. The controller transmits the outlier and the corresponding water pump failure message to the LCD screen for display; In step 1, the source key value of the water pump is the key value of the water pump in the reverse slope section transmitted via sampling; the key value of the water pump in the reverse slope section is the speed value of the water pump in the reverse slope section, the working current value of the water pump in the reverse slope section, or the active value of the water pump in the reverse slope section; The sensor is a rotation speed sensor, a current sensor or an active power sensor arranged on the water pump in the reverse slope section.

2. The method for processing water pump data based on tunnel inclined shaft according to claim 1 is characterized in that: In step 1, the source key value of the receiving water pump is the key value of the receiving water pump sampled and transmitted in the set time interval during the working period; In step 1, useless values ​​include NULL values, redundant values, illegal values, and set-period values.

3. The method for processing water pump data based on tunnel inclined shaft according to claim 2 is characterized in that: In step 2, reduce The operational equation is: ; Here, is the key value queue, is the sequence code of the key value in the key value queue, Is the first in the key value queue Key values, Is the first in the key value queue Key values, Is the first in the reduction queue A reduction; In step 2, decrement the total value The operational equation is: ; Here, is the number of key values ​​in the key value queue; In step 2, the reduction ratio The operational equation is: , It is the first The reduction ratio of a reduction.

4. The method for processing water pump data based on tunnel inclined shaft according to claim 3 is characterized in that: In step 3, decrement the standard deviation The operational equation is: , The first The standard deviation of the reduction; In step 3, decrement the total standard deviation The operational equation is: ; In step 3, decrement the standard deviation ratio The operational equation is: , is the standard deviation ratio of the reduction in the first The reduction is within the total standard deviation of all reductions, that is, the first in the reduction standard deviation ratio queue. The ratio of the standard deviation of the reduction.

5. The method for processing water pump data based on tunnel inclined shaft according to claim 4 is characterized in that: In step 4, the starting point is aligned with the separation point critical amount The operational equation is: .

6. The method for processing water pump data based on tunnel inclined shaft according to claim 5 is characterized in that: In step 4, the end point is aligned with the separation point critical amount The operational equation is: .

7. The method for processing water pump data based on tunnel inclined shaft according to claim 6 is characterized in that: In step 5, the outlier separation point critical volume The operational equation is: .

8. The method for processing water pump data based on tunnel inclined shaft according to claim 7 is characterized in that: In step 6, clustering is performed based on the critical value of being below or not below the outlier separation point, the number of key values ​​in each cluster is summed up, and the cluster value with the lower number in a pair of clusters is confirmed to be an outlier, which is specifically: Step 6-1: Take the key value that is not less than the outlier separation point as clustering queue 1; Step 6-2: Take the critical value below the outlier separation point as clustering queue 2; Step 6-3: Treat the clustered queue with a lower number of key values ​​as the outlier queue.

9. A water pump data processing system based on tunnel inclined shaft, characterized in that: include: The tunnel inclined shaft is set on the reverse slope section, and a water pump is set in the reverse slope section for drainage. The sensor and the LCD screen are connected to the controller. The sensor is used to sample the key values ​​of the water pump in the reverse slope section and transmit them to the controller. The controller is used to identify the condition of the water pump through the key values ​​of the water pump in the reverse slope section transmitted by the sampling. The modules running on the controller include: A removal module is used to receive the source key values ​​of the water pump, remove useless values, and perform arrangement in order from low to high of the key values ​​to obtain a key value queue; A calculation module, which is used to calculate the decrements of the next key value and the previous key value in the key value queue in sequence to obtain a decrement queue, and respectively calculate the ratio of the decrement in the total decrement value to obtain a decrement ratio queue; A ratio module, which is used to sequentially calculate the ratio of the standard deviation of the reduction to the total value of the standard deviation of the reduction, and obtain a reduction standard deviation ratio queue; A separation module is used to sequentially perform multiplication operations on the key value queue starting from the first key value and each reduction standard deviation ratio of the reduction standard deviation ratio queue to obtain the critical amount of the start point level separation point, and sequentially perform multiplication operations on the key value queue starting from the second key value and each reduction standard deviation ratio of the reduction standard deviation ratio queue to obtain the critical amount of the end point level separation point; A mean value module is used to calculate the mean of the critical value of the starting point and the critical value of the end point to obtain the critical value of the outlier separation point; An outlier module is used to obtain an outlier queue by comparing the outlier separation point critical quantity and the key value. If there is no outlier in the outlier queue, it is determined that the water pump has no failure condition. If there is an outlier in the outlier queue, it is determined that the corresponding water pump has a failure condition. The controller transmits the outlier and the corresponding water pump failure condition to the LCD screen for display; The source key value of the water pump is the key value of the water pump in the reverse slope section transmitted through sampling; the key value of the water pump in the reverse slope section is the speed value of the water pump in the reverse slope section, the working current value of the water pump in the reverse slope section or the active value of the water pump in the reverse slope section; the sensor is a speed sensor, current sensor or active power sensor arranged on the water pump in the reverse slope section.

Images