Operation fault monitoring system for industrial sewage treatment equipment

Abstract

The invention belongs to the technical field of equipment fault monitoring, and particularly discloses and provides an industrial sewage treatment equipment operation fault monitoring system which comprises a primary fault monitoring module, a secondary monitoring judgment module, a secondary fault monitoring module and a fault early warning execution terminal. The primary fault monitoring module compares the real-time parameters of the sewage pump with a preset threshold value, determines the fault level when the parameters exceed the threshold value, and starts the early warning execution terminal; when the primary monitoring is normal, the secondary monitoring judgment module outputs a signal grade through conjoint analysis by combining historical health data of the equipment and multiple parameters of the water collecting well, and when the signal grade is high, the secondary fault monitoring module is started; the secondary fault monitoring module sets a time window according to a health index, calculates an operation parameter related index, adjusts a threshold value in combination with water quality, and counts a fault trend emergency degree; and finally, the fault early warning execution terminal realizes graded accurate monitoring and timely early warning, can adapt to complex working conditions, and reduces equipment damage and production loss.

Description

Technical Field

[0001] The present invention belongs to the technical field of equipment fault monitoring. Specifically, it relates to an operation fault monitoring system for industrial sewage treatment equipment. Background Art

[0002] With the rapid development of industry, the discharge of industrial sewage is increasing day by day. And the industrial sewage treatment process involves a variety of complex equipment and processes. In order to ensure the sewage treatment effect, it is necessary to monitor the operation faults of relevant equipment, such as sewage pumps.

[0003] The prior art, such as a real-time online monitoring and fault diagnosis system and method for a pumping station disclosed in a Chinese invention patent application with the application number 2017111832950, collects data such as vibration, temperature, and current in real time through a sensor network. After filtering and analyzing by a signal processing module, an intelligent diagnosis is realized by an industrial control computer software platform, and remote warning is combined with an alarm module, which solves the problems of low efficiency of traditional manual monitoring and lagging fault response, realizes automatic monitoring and rapid maintenance decision-making of the pumping station, reduces labor costs and improves equipment reliability.

[0004] The prior art, such as a method, system, computer storage medium and device for finding / processing pumping station faults disclosed in a Chinese invention patent application with the application number 2019102170828, captures abnormal signals by collecting data such as pressure and liquid level in real time and using a threshold comparison and trend analysis algorithm, locates the fault point by combining spectrum analysis or logic tree diagnosis, and judges the running steady state at the same time to reduce false alarms, which solves the problems of low efficiency of manual investigation and missed detection of potential risks, realizes accurate fault location and optimized allocation of maintenance resources, and reduces the incidence of safety accidents.

[0005] For the above technical solutions, obviously, for the first technical solution, its core is to realize a remote monitoring-diagnosis-decision closed loop through multi-module collaboration. For the second technical solution, its core is to improve the fault response accuracy through algorithm diagnosis + logic positioning, both of which replace manual experience with data-driven, but there are still the following deficiencies: 1. Currently, only the operation parameters of the pump body itself are used for fault judgment. When no fault is characterized, no extended fault prediction judgment is carried out, resulting in insufficient accuracy of fault monitoring.

[0006] 2. Currently, the fault monitoring only focuses on the pump body itself and does not carry out linkage monitoring with the sewage treatment process, such as the sediment in the collection tank and the water level change, resulting in weak fault root cause analysis and prediction capabilities.

[0007] 3. When making current fault judgments, the selection of fixed static thresholds makes it difficult to guarantee adaptability, and thus the false alarm rate and missed detection rate are both relatively high. Summary of the Invention

[0008] In view of this, to solve the problems raised in the above background technology, a system for monitoring the operation faults of industrial sewage treatment equipment is proposed.

[0009] The object of the present invention can be achieved by the following technical solutions: The present invention provides a system for monitoring the operation faults of industrial sewage treatment equipment, and the system includes the following modules: a primary fault monitoring module, which compares the real-time operation parameters of the sewage pump with the corresponding preset thresholds, and determines the fault level and activates the fault warning execution terminal when the corresponding preset thresholds are exceeded.

[0010] A secondary monitoring and judgment module, when the primary fault monitoring is normal, calculates the current health index by combining the historical health monitoring data of the equipment, and collects the influent flow rate, water level rising rate and bottom silt thickness in the sump within a preset time window, and outputs the secondary monitoring requirement signal level through multi-parameter joint analysis. If the signal level is high, activate the secondary fault monitoring module.

[0011] A secondary fault monitoring module, sets a secondary monitoring time window in combination with the current health index to calculate the change rate, standard deviation and trend slope of the operation parameters within the monitoring time window, and adjusts the fault judgment threshold by combining the water quality conditions in the sump. If any index exceeds the corresponding adjusted fault judgment threshold, count the fault tendency urgency.

[0012] A fault warning execution terminal, activates the corresponding warning instruction and conducts the corresponding warning based on the fault level and the fault tendency urgency.

[0013] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) The present invention solves the problem that the current only uses the operation parameters of the pump body itself without extended fault prediction and judgment by constructing a method of "primary real-time threshold monitoring + secondary health status and environmental coupling analysis", changing from the traditional single threshold comparison to a dynamic environment adaptive multi-dimensional analysis mode, and improving the accuracy of fault monitoring.

[0014] (2) The present invention determines the fault level by constructing a multi-dimensional feature evaluation system of parameter deviation coverage rate, interference parameter deviation coverage rate, parameter deviation degree, parameter deviation duration ratio and target overrun time difference. These dimensions describe the operation state of the equipment in an all-round and multi-level manner, greatly improving the determination accuracy of the fault level compared with single-dimensional judgment, providing accurate and reliable basis for equipment maintenance, and helping to reasonably arrange maintenance resources.

[0015] (3) The present invention calculates the current health index by integrating the historical health data of the equipment, and deeply digs out potential hidden dangers. And through multi-parameter joint analysis, it inputs the secondary monitoring requirement signal level, which not only avoids over-monitoring but also can accurately locate key anomalies, improves the depth and breadth of fault monitoring, makes up for the lack of current linkage monitoring with the sewage treatment process, is convenient for fault root cause analysis, and improves the prediction ability.

[0016] (4) The present invention adjusts the fault judgment threshold by combining the water quality conditions in the catch basin, ensuring the effectiveness of the threshold judgment result and the adaptability of the operation fault monitoring, thereby further reducing the false alarm rate and the undetected rate, and ensuring the stable and reliable operation of the sewage treatment process. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for describing the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0018] Figure 1 It is a schematic diagram of the connection of the system modules of the present invention.

[0019] Figure 2 It is a schematic diagram of the overall implementation steps flow of the present invention.

[0020] Figure 3 It is a schematic diagram of the process of the secondary monitoring and judgment module of the present invention.

[0021] Figure 4 It is a schematic diagram of the process of processing the secondary monitoring demand signal level of the present invention.

[0022] Figure 5 It is a schematic diagram of the overall process of the multi-parameter joint analysis of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0023] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0024] Embodiment 1

[0025] Please refer to Figure 1 and Figure 2 As shown, the present invention provides an industrial sewage treatment equipment operation fault monitoring system, which includes: a primary fault monitoring module, a secondary monitoring and judgment module, a secondary fault monitoring module, and a fault warning execution terminal.

[0026] Among the above, the secondary monitoring and judgment module is respectively connected to the primary fault monitoring module and the secondary fault monitoring module, and the fault warning execution terminal is respectively connected to the primary fault monitoring module and the secondary fault monitoring module.

[0027] The primary fault monitoring module compares the real-time operating parameters of the sewage pump with the corresponding preset thresholds. When the corresponding preset thresholds are exceeded, it determines the fault level and activates the fault warning execution terminal.

[0028] It should be added that the operating parameters include the flow rate, current, pressure, and vibration amplitude of the sewage pump. The preset thresholds are set according to the rated parameters of the equipment. Among them, the flow rate threshold range is ±15% of the rated value, the pressure threshold is ±20% of the rated value, the current threshold range is 20% of the rated value, and the vibration amplitude threshold ≤ 3.0 mm / s.

[0029] Understandably, during the operation of industrial sewage treatment equipment, the primary fault monitoring only relies on the comparison of real-time operating parameters with fixed thresholds, and it is difficult to capture equipment performance degradation, multi-parameter coupling anomalies, or progressive faults, such as hidden efficiency decline caused by silt accumulation, continuous deterioration of the health index, and other non-threshold over-limit problems, which are likely to result in missed early faults or false alarms for instantaneous fluctuations. Therefore, the following secondary monitoring and judgment are carried out to make up for the limitations of single-threshold judgment.

[0030] Furthermore, determining the fault level includes: A1. Calculate the parameter deviation coverage rate by dividing the total number of currently monitored operating parameter items by the number of items exceeding the corresponding thresholds.

[0031] A2. Mark the operating parameter items exceeding the thresholds as over-limit parameter items, count the number of over-limit parameter items whose over-limit duration exceeds the corresponding preset limit duration, and calculate the parameter deviation duration ratio by dividing it by the total number of over-limit parameter items.

[0032] A3. Calculate the difference between the over-limit duration of each over-limit parameter item and the corresponding preset limit duration, and take the maximum difference as the target over-limit time difference.

[0033] A4. Use the parameter deviation coverage rate, parameter deviation duration ratio, and target over-limit time difference as input parameters to match the preset fault level rule table to obtain the matched fault level.

[0034] Among them, when matching the preset fault level rule table, the input parameters of the fault level rule table are also obtained through the following steps: A41. Calculate the absolute deviation between the acquisition value of each over-limit parameter item and the corresponding threshold.

[0035] A42. Calculate the interference parameter deviation coverage rate by dividing the number of over-limit parameter items whose absolute deviation exceeds the preset interference deviation by the total number of over-limit parameter items.

[0036] A43. Analyze the absolute deviations of each over-limit parameter item through the preset deviation weight to obtain the parameter deviation degree.

[0037] The fault level rule table matches the fault level according to the combined characteristics of parameter deviation coverage rate, interference parameter deviation coverage rate, parameter deviation degree, parameter deviation duration ratio, and target overlimit time difference.

[0038] Understandably, the parameter deviation coverage rate reflects the proportion of overall operating parameters deviating from the normal range, and can quickly insight into the extent of parameters involved in faults. The interference parameter deviation coverage rate can effectively exclude interference factors, accurately focus on the parameters related to real faults, and improve the purity of fault judgment. The parameter deviation degree carefully quantifies the severity of faults through comprehensive analysis of deviation situations. The parameter deviation duration ratio presents the persistent and stubborn characteristics of faults clearly from the time dimension. The target overlimit time difference locks in the most severe overlimit situation and highlights the significant impact of key fault points on the equipment.

[0039] It should be noted that faults have multi-causality, gradualness, and suddenness. For example, affected by mechanical, electrical, and environmental coupling, a single indicator cannot comprehensively evaluate. Through the hierarchical verification of "breadth → depth → persistence", the accuracy of fault classification can be significantly improved, promoting the transformation of fault management from "post-failure maintenance" to "predictive maintenance".

[0040] In a specific embodiment, the matching preset fault level rule table is comprehensively set by combining equipment technical specifications, historical fault data, and industry standards, etc. Among them, the parameter deviation coverage rate, interference parameter deviation coverage rate, parameter deviation degree, parameter deviation duration ratio, and target overlimit time difference are respectively denoted as PF, GF, CP, CK, and ΔT, as shown in Table 1 for reference.

[0041] Table 1 Fault Level Rule Table

[0042]

[0043] The embodiment of the present invention determines the fault level by constructing a multi-dimensional feature evaluation system of parameter deviation coverage rate, interference parameter deviation coverage rate, parameter deviation degree, parameter deviation duration ratio, and target overlimit time difference. These dimensions describe the operating state of the equipment in an all-round and multi-level manner, greatly improving the determination accuracy of the fault level compared with single-dimensional judgment, providing accurate and reliable basis for equipment maintenance, and helping to reasonably arrange maintenance resources.

[0044] Please refer to Figure 3 As shown, for the secondary monitoring judgment module, when the primary fault monitoring is normal, it calculates the current health index in combination with the equipment historical health monitoring data, and collects the influent flow rate, water level rising rate, and bottom silt thickness of the sump well within a preset time window, and outputs the secondary monitoring requirement signal level through multi-parameter joint analysis. If the signal level is high, it activates the secondary fault monitoring module.

[0045] Understandably, the secondary monitoring and judgment module upgrades the traditional passive threshold alarm to active trend prediction through the progressive logic of "multi-source data fusion of water inlet flow, water level rise rate, bottom well sludge thickness and drainage flow → collaborative trend quantification → dynamic signal classification", thereby achieving early detection of hidden faults.

[0046] Specifically, the specific calculation process of the current health index is as follows: B1. Extract the time of each health monitoring, performance parameter health assessment score, operation parameter health assessment score and comprehensive health assessment score from the historical health monitoring data and form a three-dimensional data sequence.

[0047] B2. Scan adjacent time points to compare the evaluation scores and count the frequency of continuous decline in each dimension, and then compare it with the total number of health monitoring times to obtain the continuous decline ratio of each dimension.

[0048] B3. Based on time series analysis, linear fitting is performed on the data series of each dimension to obtain the change rate of each dimension.

[0049] B4. The continuous decline ratio and change rate of each dimension are combined into a feature matrix, and the features of each dimension are integrated through a preset weight matrix to output the health index correction coefficient.

[0050] B5. Correct the comprehensive health assessment score of the last monitoring using the health index correction coefficient.

[0051] Understandably, performance parameters are usually related to the core functions of the equipment, such as flow, head, and efficiency. For further understanding, the features of each dimension are fused through the preset weight matrix, and the specific example of outputting the health index correction coefficient is as follows: the decline ratio of performance parameters, operating parameters, and comprehensive health is recorded as , and .

[0052] If the change rate of a dimension is greater than or equal to 0, 0 is used as the normalization result of the change rate of the dimension. Otherwise, the change rate is mapped to between 0 and 1 through a mapping function. For example, the change rate is recorded as , select the exponential decay function as the mapping function for mapping, that is, select As a mapping function, is a variable, , is a natural constant.

[0053] The normalized processing results of the corresponding change rates of each dimension are recorded as , and , construct the feature matrix , , the preset weight matrix is ​​recorded as , , ~ respectively represent , , , , and the corresponding weights, and the sum is 1.

[0054] Multiply the feature matrix and the preset weight matrix to calculate the product, and take the output result as the health index correction coefficient. Among them, , , , , and The specific values of the corresponding weights can be set in combination with empirical data, etc.

[0055] It should also be added that the specific correction formula of the health index correction coefficient is as follows: , represents the comprehensive health assessment score of the last monitoring, represents the health index correction coefficient, represents the current health index.

[0056] Specifically, please refer to Figure 5 shown, the specific analysis process of the multi-parameter joint analysis includes: C1. Extract the drainage flow from the sewage pump operation parameters, calculate the planned water level rise rate in combination with the cross-sectional area of the sump, the influent flow rate within the preset time window, and the thickness of the bottom sludge, and then compare it with the collected water level rise rate to obtain the water level rise rate difference.

[0057] C2. Obtain the allowable rate difference corresponding to the current health index by matching in the preset health index - actual characterization deviation table.

[0058] C3. If the water level rise rate difference is less than the allowable rate difference, mark the secondary monitoring requirement signal level as low, and calculate and analyze the water level rise rate deviation degree in combination with the water level rise rate difference and the allowable rate difference.

[0059] Among them, the water level rise rate deviation degree is obtained by dividing the difference between the water level rise rate difference and the allowable rate difference by the allowable rate difference.

[0060] C4. If the water level rise rate deviation degree is lower than the corresponding preset threshold, mark the secondary monitoring requirement signal level as medium, otherwise as high.

[0061] Among them, the specific analysis process of the planned water level rising rate is as follows: C11. Calculate the growth rate of the bottom sludge thickness within the preset time window, and screen out the maximum bottom sludge thickness.

[0062] C12. After normalizing the growth rate and the maximum bottom sludge thickness, input them into the Sigmoid function to generate the sludge thickness influence factor.

[0063] C13. Use the time window weighted average algorithm to calculate the average flow difference between the corresponding influent flow and effluent flow within the preset time window.

[0064] C14. After correcting the cross-sectional area of the sump through the sludge thickness influence factor, obtain the effective cross-sectional area of the sump, and take the ratio of the average flow difference to the effective cross-sectional area of the sump as the planned water level rising rate.

[0065] It should be noted that in actual operation, sludge will accumulate at the bottom of the well, which actually reduces the effective water flow cross-sectional area. The increase in sludge thickness means that the available vertical space decreases, so the effective cross-sectional area will decrease. This will in turn affect the water flow velocity and rising rate, so it is necessary to correct the cross-sectional area.

[0066] Specifically, the specific calculation formula of the time window weighted average algorithm is: , represents the average flow difference within the preset time window, is the set time window weight, represents time, represents the influent and effluent flow difference at the moment, represents the time window length,

[0067] It should also be noted that the correction method for correcting the cross-sectional area of the sump through the sludge thickness influence factor is the same as the correction method of the health index correction coefficient, and its specific formula will not be shown here.

[0068] Further, please refer to Figure 4 As shown, when the secondary monitoring demand signal level is low, it outputs that the sewage pump is operating normally.

[0069] When the secondary monitoring demand signal level is medium, match and compare the water level rising rate deviation degree with the water level rising rate deviation degrees corresponding to the preset extended monitoring ratios to output the matching extended monitoring ratio.

[0070] Adjust the prediction time window through the extended monitoring ratio, and repeatedly execute the sump data collection and secondary monitoring demand signal level analysis until the signal level is determined to be low or high.

[0071] It should be added that the deviation degree of the water level rising rate corresponding to each extended monitoring ratio

[0072] In the embodiment of the present invention, the current health index is calculated by integrating the historical health data of the device to deeply explore potential hidden dangers. And by performing multi-parameter joint analysis to input the secondary monitoring requirement signal level, it not only avoids over-monitoring but also can accurately locate key anomalies, improves the depth and breadth of fault monitoring, makes up for the lack of linkage monitoring with the sewage treatment process, facilitates root cause analysis of faults, and improves the prediction ability.

[0073] The secondary fault monitoring module sets a secondary monitoring time window in combination with the current health index, calculates the change rate, standard deviation and trend slope of the operating parameters within the monitoring time window, and adjusts the fault judgment threshold by combining the water quality situation in the sump. If any index exceeds the corresponding adjusted fault judgment threshold, the fault tendency urgency is statistically counted.

[0074] Understandably, the secondary fault monitoring module upgrades the traditional threshold alarm to a predictive decision-making system through the multi-dimensional data fusion and dynamic logic chain of "health index - sump water quality - operating parameters", captures hidden faults, and improves the foresight of alarms.

[0075] Specifically, the setting of the secondary monitoring time window includes: D1. Query the preset monitoring time setting table based on the secondary monitoring requirement signal level and the type of sewage pump to obtain the corresponding reference time window.

[0076] D2. Synthesize the deviation degree of the water level rising rate and the current health index, calculate the monitoring adjustment coefficient through weighted fusion, and adjust the reference time window based on the monitoring adjustment coefficient to obtain the secondary monitoring time window.

[0077] Understandably, the length of the secondary monitoring time window is set to be adjustable for the purpose of responding to potential faults faster when the requirement signal level is high.

[0078] The monitoring time setting table is comprehensively set by combining historical data, industry standards, expert experience, etc. For example, analyze the fault records of the same type of sewage pumps in the past 3 years, such as blockage, overload, and wear, count the average response time of different faults, and combine the monitoring requirements of the outdoor drainage design specification, the centrifugal pump technical conditions, and the recommended maintenance cycle and key parameter sampling frequency in the pump machine technical manual, etc., and can be manually imported by personnel.

[0079] It should be added that the deviation degree of the water level rising rate is denoted as , and the current device health index is denoted as , and the weighted fusion formula is: , is the monitoring adjustment coefficient, and respectively represent the weight corresponding to the water level rising rate deviation degree and the current health index. The higher the water level rising rate deviation degree and the lower the health index, the higher the need to shorten the monitoring time window. To avoid excessive adjustment of the time window, it is set that the maximum value of is 0.5.

[0080] Among them, the specific calculation formula for the secondary monitoring time window is: , represents the reference time window, represents the secondary monitoring time window. Exemplarily, assume that the reference time window is 30 minutes and the monitoring adjustment coefficient is 0.5, that is, the secondary monitoring time window is set to 15 minutes.

[0081] Specifically, adjusting the fault judgment threshold includes: E1. Real-time monitor the water quality parameters at different depths in the sump, including at least one of turbidity and suspended solids.

[0082] E2. Calculate the average value of the water quality parameters at different depths. If the calculation result is within the preset normal range of the water quality parameters, maintain the preset original fault judgment threshold; otherwise, calculate the water quality parameter influence coefficient based on the average value of the water quality parameters.

[0083] E3. Adjust the preset original fault judgment threshold through the water quality parameter influence coefficient to generate a dynamically adjusted fault judgment threshold.

[0084] It can be understood that the real-time monitoring of the water quality parameters at different depths in the sump is achieved by installing suspended solid sensors and turbidity sensors at different depth positions in the sump to real-time monitor the content of suspended solids and turbidity in the water quality.

[0085] It should be noted that the adjustment of the original fault judgment threshold through the water quality parameter influence coefficient is the same as the correction method of the health index correction coefficient, and its specific formula will not be shown here.

[0086] It can also be understood that the specific calculation process of calculating the water quality parameter influence coefficient based on the average value of the water quality parameters is as follows: Take the median of the preset normal range of the water quality parameters as the reference value, and take the difference between the upper limit and the lower limit of the preset normal range of the water quality parameters to obtain the fluctuation range. Calculate the difference between the average value of the water quality parameters and the reference value, take the absolute value of the difference, and divide it by the fluctuation range to obtain the water quality parameter influence coefficient. Specifically, to ensure the rationality and scientificity of the setting of the fault judgment threshold, the maximum value of the water quality parameter influence coefficient is set to 0.3.

[0087] Another specifically, the statistical fault tendency urgency includes: L1. Mark the indicators that exceed the corresponding fault judgment threshold as the exceeded indicators, and calculate the fault trigger ratio by dividing the exceeded indicators by the total number of indicators.

[0088] L2. If the fault trigger ratio is 1, mark the fault tendency urgency as 100%. Otherwise, calculate the threshold exceedance degree for the exceeded indicators.

[0089] L3. Analyze the fault trigger ratio and the threshold exceedance degrees of each exceeded indicator through rule - based logic fusion technology to obtain the fault tendency urgency.

[0090] It should be added that the specific calculation process of calculating the threshold exceedance degree is as follows: Calculate the difference between the actual calculated value of the indicator and the corresponding fault judgment threshold, divide the calculated difference by the corresponding fault judgment threshold to obtain the corresponding threshold exceedance degree. And to facilitate analysis and ensure the rationality of the data range, the maximum value of the threshold exceedance degree is set to 1.

[0091] Understandably, there is a logical hierarchy between the threshold exceedance degree and the fault trigger ratio. The trigger ratio is the pre - condition for "whether a fault is triggered", and the exceedance degree is the refinement of "the severity of the fault". And the rules can clearly define the logical hierarchy between the indicators, facilitating accurate and rapid decision - making. And in the calculation of the fault trigger urgency, the rule - based logic fusion technology avoids the deficiencies of statistical technologies such as weighted fusion in non - linear logic, decision transparency, and real - time performance through hard - condition priority, logical layering, and interpretable decision - making.

[0092] In a preferred embodiment of the present invention, for the convenience of analysis and understanding, the rule - based logic fusion technology can specifically adopt Boolean rules. Exemplarily, if the fault trigger ratio is greater than 0.8 or the threshold exceedance degree of any exceeded indicator is greater than 0.9, then the urgency is 100%. If the fault trigger ratio is greater than 0.5 and less than or equal to 0.8, and at the same time the threshold exceedance degree of at least 1 exceeded indicator is greater than 0.7, or the fault trigger ratio is greater than 0.5 and less than or equal to 0.7, and at the same time the threshold exceedance degrees of at least two exceeded indicators are both greater than 0.6, then the urgency is 80%. If the fault trigger ratio is less than 0.5 or the threshold exceedance degrees of all exceeded indicators are less than 0.6, then the urgency = 30% + average threshold exceedance degree × 70%.

[0093] The embodiment of the present invention adjusts the fault judgment threshold by combining the water quality situation in the sump, ensuring the effectiveness of the threshold judgment result and the adaptability of the operation fault monitoring, thereby further reducing the false alarm rate and the missed detection rate, and ensuring the stable and reliable operation of the sewage treatment process.

[0094] The fault warning execution terminal starts the corresponding warning instruction and conducts corresponding warnings based on the fault level and the fault tendency urgency.

[0095] Understandably, the warning instructions triggered based on the fault level are as follows: for a first-level fault, immediately cut off the power supply of the current sewage pump and switch to the standby pump; for a second-level fault, execute frequency reduction operation on the sewage pump, such as reducing the motor frequency to 70%-80% of the rated value, restricting the drainage flow to a safe range, and generating a maintenance work order for automatic push. For example, send an instruction of "Second-level warning: the vibration of pump No. 2 has increased significantly. It is recommended to check the bearing lubrication status within 48 hours." to the relevant operation and maintenance management personnel; for a third-level fault, record the current operating parameters of the sewage pump to generate a record log for storage.

[0096] The warning instructions triggered based on the urgency of the fault trend are as follows: if the urgency of the fault trend is greater than or equal to the set first threshold, mark the current as a first-level fault and trigger the set warning instructions for the first-level fault; if the urgency of the fault trend is greater than or equal to the set second threshold and less than the set first threshold, mark the current as a second-level fault and trigger the set warning instructions for the second-level fault; if the urgency of the fault trend is less than the set second threshold, mark the current as a third-level fault and trigger the set warning instructions for the third-level fault.

[0097] Exemplarily, for the convenience of analysis, the first threshold and the second threshold can be respectively set to 0.8 and 0.5.

[0098] In the embodiment of the present invention, by constructing a method of "first-level real-time threshold monitoring + second-level coupling analysis of health status and environment", the problem of only making non-extended fault prediction and judgment based on the operating parameters of the pump body itself is solved. It is transformed from the traditional single threshold comparison to a dynamic environment adaptive multi-dimensional analysis mode, improving the accuracy of fault monitoring.

[0099] Embodiment 2

[0100] To verify the necessity of the second-level monitoring and judgment module and the second-level monitoring and judgment settings, taking the sewage treatment plant in a chemical industrial park as an example, a comparison is made with a static threshold as the comparison scheme. The static threshold for flow is set to 60 cubic meters per hour, and the pressure threshold is set to 0.4 MPa.

[0101] During a daily operation, the flow rate of the sewage pump suddenly climbed from 60 cubic meters per hour to 68 cubic meters per hour briefly and then quickly dropped back to the normal range, while the pressure remained at 0.4 MPa. According to the comparison with the static threshold, the comparison scheme determines that the equipment is operating normally.

[0102] For the technical solution of the present invention, after the primary monitoring is judged to be normal, the secondary monitoring is entered. Based on the historical health monitoring data of the equipment in the past 8 months, the system obtains the current health index as 75, with a full score of 100. At the same time, it is collected that within the past 3 hours in the sump, the influent flow rate has increased from 110 cubic meters per hour to 130 cubic meters per hour, the water level rising rate is 0.3 meters per hour, and the bottom sludge thickness reaches 0.45 meters. Through the combined analysis of multiple parameters, the output signal level of the secondary monitoring requirement is medium. The system further compares the water level rising rate deviation degree with the water level rising rate deviation degrees corresponding to the preset extended monitoring ratios, and matches the extended monitoring ratio to 40%, thereby extending the prediction time window from 3 hours to 4.2 hours. After re-monitoring, the signal level of the secondary monitoring requirement is upgraded to high, and the secondary fault monitoring module is started. Through the monitoring and analysis of the secondary fault monitoring module, the fault tendency urgency is obtained as 0.6, reaching the secondary fault level, and the warning instruction of the secondary fault level setting is triggered.

[0103] Comparing the two technical solutions, it is obvious that the missed detection rate and false detection rate of the technical solution adopted by the present invention have been significantly reduced, and the accuracy of fault judgment has been strongly guaranteed. It can well capture short-term data anomalies, successfully avoid possible production stoppage losses, strongly guarantee the stable operation of production, and also reduce the maintenance cost.

[0104] By integrating the historical health data of the equipment, the influent flow rate of the sump, the water level deviation, the sludge thickness and the real-time operation status, multi-dimensional and progressive fault symptom recognition is realized, and the rough judgment defect of "either black or white" in the primary monitoring is solved, so as to realize early fault warning and reduce the false alarm and missed alarm rates.

[0105] The above formula is a formula obtained by collecting a large amount of data for software simulation to approximate the real situation recently. The preset parameters in the formula are set by those skilled in the art according to the actual situation.

[0106] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product.

[0107] Those of ordinary skill in the art can realize that the modules and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0108] In addition, in each embodiment of the present application, each functional module may be integrated into one processing module, may exist physically alone for each module, or two or more modules may be integrated into one module.

[0109] As described above, it is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

[0110] Finally, the above are only the preferred embodiments of the present invention and are not used to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An operation fault monitoring system for industrial sewage treatment equipment, characterized in that, The system includes the following modules: The primary fault monitoring module compares the real-time operation parameters of the sewage pump with the corresponding preset thresholds. When the parameters exceed the corresponding preset thresholds, it determines the fault level and activates the fault warning execution terminal; The secondary monitoring and judgment module calculates the current health index by combining the historical health monitoring data of the equipment when the primary fault monitoring is normal, and collects the influent flow rate, water level rising rate, and bottom silt thickness in the sump within a preset time window. It outputs the secondary monitoring requirement signal level through multi-parameter joint analysis. If the signal level is high, it activates the secondary fault monitoring module; The secondary fault monitoring module sets a secondary monitoring time window based on the current health index, calculates the change rate, standard deviation, and trend slope of the operation parameters within the monitoring time window, and adjusts the fault judgment threshold by combining the water quality situation in the sump. If any index exceeds the corresponding adjusted fault judgment threshold, it counts the fault tendency urgency; The fault warning execution terminal activates the corresponding warning instruction and conducts the corresponding warning based on the fault level and the fault tendency urgency.

2. The operation fault monitoring system for an industrial sewage treatment device according to claim 1, characterized in that: The determination of the fault level includes: Calculating the parameter deviation coverage rate by dividing the total number of currently monitored operation parameter items by the number of items exceeding the corresponding threshold; Marking the operation parameter items exceeding the threshold as over-limit parameter items, counting the number of over-limit parameter items whose over-limit duration exceeds the corresponding preset limit duration, and calculating the parameter deviation duration ratio by dividing it by the total number of over-limit parameter items; Calculating the difference between the over-limit duration of each over-limit parameter item and the corresponding preset limit duration, and taking the maximum difference as the target over-limit time difference; Taking the parameter deviation coverage rate, parameter deviation duration ratio, and target over-limit time difference as input parameters, and obtaining the matched fault level after matching the preset fault level rule table.

3. An industrial sewage treatment equipment operation fault monitoring system according to claim 2, characterized in that: When matching the preset fault level rule table, the input parameters of the fault level rule table are also obtained through the following steps: Calculating the absolute deviation between the collected value of each over-limit parameter item and the corresponding threshold; Calculating the interference parameter deviation coverage rate by dividing the number of over-limit parameter items whose absolute deviation exceeds the preset interference deviation by the total number of over-limit parameter items; Analyzing the absolute deviation of each over-limit parameter item through a preset deviation weight to obtain the parameter deviation degree; The fault level rule table performs fault level matching according to the combined characteristics of the parameter deviation coverage rate, interference parameter deviation coverage rate, parameter deviation degree, parameter deviation duration ratio, and target over-limit time difference.

4. An operation fault monitoring system for an industrial sewage treatment device according to claim 1, characterized in that: The specific calculation process of the current health index is as follows: Extracting the time, performance parameter health assessment score, operation parameter health assessment score, and comprehensive health assessment score of each health monitoring from the historical health monitoring data and forming a three-dimensional data sequence; Scanning adjacent time points for assessment score comparison and counting the continuous decline frequency of each dimension, and then calculating the continuous decline ratio of each dimension by dividing it by the total number of health monitoring times; Performing linear fitting on the data sequence of each dimension based on time series analysis to obtain the change rate of each dimension; Forming a feature matrix with the continuous decline ratio and change rate of each dimension, and fusing the features of each dimension through a preset weight matrix to output the health index correction coefficient; Correcting the comprehensive health assessment score of the last monitoring through the health index correction coefficient.

5. The operation fault monitoring system of an industrial sewage treatment device according to claim 1, characterized in that: The specific analysis process of the multi-parameter joint analysis includes: Extract the drainage flow rate from the operating parameters of the sewage pump, calculate the planned water level rising rate by combining the cross-sectional area of the sump well, the influent flow rate within a preset time window, and the thickness of the bottom sludge, and then compare it with the collected water level rising rate to obtain the water level rising rate difference. Obtain the permitted rate difference corresponding to the current health index by matching in the preset health index - actual characterization deviation table. If the water level rising rate difference is less than the permitted rate difference, mark the secondary monitoring requirement signal level as low, and calculate and analyze the water level rising rate deviation degree by combining the water level rising rate difference and the permitted rate difference. If the water level rising rate deviation degree is lower than the corresponding preset threshold, mark the secondary monitoring requirement signal level as medium, otherwise as high.

6. An industrial sewage treatment equipment operation fault monitoring system according to claim 5, characterized in that: The specific analysis process of the planned water level rising rate is as follows: Calculate the growth rate of the bottom sludge thickness within a preset time window and screen out the maximum bottom sludge thickness. After standardizing the growth rate and the maximum bottom sludge thickness, input them into the Sigmoid function to generate the sludge thickness influence factor. Use the time window weighted average algorithm to calculate the average flow rate difference between the influent flow rate and the drainage flow rate within a preset time window. After correcting the cross-sectional area of the sump well by the sludge thickness influence factor to obtain the effective cross-sectional area of the sump well, take the ratio of the average flow rate difference to the effective cross-sectional area of the sump well as the planned water level rising rate.

7. An operation fault monitoring system for an industrial sewage treatment device according to claim 5, characterized in that: Output that the sewage pump is operating normally when the secondary monitoring requirement signal level is low. When the secondary monitoring requirement signal level is medium, match and compare the water level rising rate deviation degree with the water level rising rate deviation degrees corresponding to the preset extended monitoring ratios to output the matching extended monitoring ratio. Adjust the prediction time window according to the extended monitoring ratio, and repeat the execution of the sump well data collection and the secondary monitoring requirement signal level analysis until the signal level is determined to be low or high.

8. An operation fault monitoring system for an industrial sewage treatment device according to claim 5, characterized in that: The setting of the secondary monitoring time window includes: Query the preset monitoring time setting table based on the secondary monitoring requirement signal level and the type of the sewage pump to obtain the corresponding reference time window. Based on the water level rising rate deviation degree and the current health index, calculate the monitoring adjustment coefficient through weighted fusion, and adjust the reference time window based on the monitoring adjustment coefficient to obtain the secondary monitoring time window.

9. The operation fault monitoring system for an industrial sewage treatment device according to claim 1, characterized in that: The adjustment of the fault judgment threshold includes: Real-time monitor the water quality parameters at different depths in the sump well, including at least one of turbidity and suspended solids. Calculate the average value of the water quality parameters at different depths. If the calculation result is within the preset normal range of the water quality parameters, maintain the preset original fault judgment threshold, otherwise calculate the water quality parameter influence coefficient based on the average value of the water quality parameters. Adjust the preset original fault judgment threshold by the water quality parameter influence coefficient to generate the dynamically adjusted fault judgment threshold.

10. The operation fault monitoring system for an industrial sewage treatment device according to claim 1, wherein: The statistics of the fault trend urgency includes: Mark the indicators exceeding the corresponding fault judgment threshold as exceeding indicators, and calculate the fault trigger ratio by taking the ratio of the exceeding indicators to the total number of indicators. If the fault trigger ratio is 1, mark the fault trend urgency as 100%, otherwise calculate the threshold exceeding degree for the exceeding indicators. The fault tendency urgency is obtained by analyzing the fault trigger ratio and the threshold exceedance degree of each exceeded index through a rule-based logic fusion technique.

Images