Ultra-filtration device operation fault early warning method and ultra-filtration device operation fault early warning system

Abstract

The invention discloses an ultra-filtration device operation fault early warning method and system, belongs to the technical field of ultra-filtration device monitoring, and solves the problem that an ultra-filtration device cannot send out an early warning signal when encountering a fault during operation. Comprising the following steps: summarizing cases of serious pollution and blockage and flow overrun of the ultrafiltration device, collecting and storing historical operation data, collecting real-time operation data in the ultrafiltration device according to related parameters, constructing an ultrafiltration device mathematical model based on an One-Class-SVM classification algorithm, inputting the collected real-time operation data into the trained ultrafiltration device mathematical model, and calculating the flow overrun of the ultrafiltration device. Judging whether the running state of the ultrafiltration device is abnormal or not, constructing an ultrafiltration device mechanism model and acquiring mechanism data, outputting a switching value sequence in combination with the mathematical model of the ultrafiltration device, and performing early warning information analysis; according to the invention, abnormal information is pushed to related personnel at the early stage of serious pollution blockage or flow overrun of the ultrafiltration device, so that the system operation efficiency is improved, and the normal operation of the boiler water supply system is prevented from being influenced.

Description

Technical Field

[0001] The invention belongs to the technical field of ultrafiltration device monitoring, and relates to an ultrafiltration device operation fault early warning method and system. Background Art

[0002] The ultrafiltration device is a pressurized membrane separation device and an important part of the water treatment system. The ultrafiltration device uses the ultrafiltration membrane as the filtration medium. When the raw water flows through the membrane surface, under the pressure difference on both sides of the ultrafiltration membrane, the many tiny micropores densely distributed on the surface of the ultrafiltration membrane only allow water and small molecules to pass through and become permeate, while the macromolecules in the raw water that are larger than the micropores on the membrane surface are retained on the water inlet side of the membrane and become concentrated liquid. The ultrafiltration device can effectively remove particles, colloids, bacteria, heat sources and organic matter in the water, achieving the effect of purifying, separating and concentrating the raw water.

[0003] In the monitoring process of existing ultrafiltration water treatment devices, it is necessary to install multiple sensors to collect the internal conditions of the device, such as water pH value, conductivity, temperature, pressure, chemical oxygen demand and other monitoring data. For example, the invention patent application with application publication number CN116226484A discloses an ultrafiltration water treatment device monitoring data management system, which uses the changes in multi-dimensional sensor data to quantify the aging degree of the ultrafiltration water treatment device corresponding to the data, and characterize the current equipment status of the ultrafiltration water.

[0004] During the operation of the ultrafiltration system, as pollutants are trapped and accumulated on the membrane surface, the flux of the ultrafiltration membrane will gradually decrease, the amount of water treated will gradually decrease, and the operating pressure difference will gradually increase, so the ultrafiltration device needs to be backwashed regularly. Under current technology, the operation of the ultrafiltration system mainly relies on sequential control, that is, the ultrafiltration device will be shut down for backwashing only when the program runs to the set time. However, when serious fouling or flow exceeds the limit during the operation of the ultrafiltration system, the ultrafiltration device is in an overloaded operating state, which is prone to equipment damage, but the existing ultrafiltration system cannot issue a warning signal when the ultrafiltration device encounters an abnormal situation and causes an operating failure. Summary of the invention

[0005] The technical solution of the present invention is used to solve the problem that an ultrafiltration device cannot issue an early warning signal when encountering a fault during operation.

[0006] The present invention solves the above technical problems through the following technical solutions:

[0007] An ultrafiltration device operation failure early warning method, comprising:

[0008] Step 1: Summarize the cases of serious fouling and flow exceeding the limit of the ultrafiltration device, collect and store historical operation data, and propose corresponding solutions and treatment measures for different types of operation failures;

[0009] Step 2: Collect real-time operating data in the ultrafiltration device according to relevant parameters and perform relevant pretreatment measures;

[0010] Step 3: construct a mathematical model of the ultrafiltration device based on the One-Class-SVM classification algorithm, including:

[0011] (1) screening the historical operation data of the ultrafiltration device in step 1 and performing relevant pretreatment measures;

[0012] (2) dividing the training data set and the test data set of the ultrafiltration device;

[0013] (3) Set the support vector machine classification model parameters and train the mathematical model;

[0014] (4) Verify the accuracy of the mathematical model of the ultrafiltration device;

[0015] (5) Evaluate and adjust the mathematical model of the ultrafiltration device;

[0016] Step 4, input the real-time operation data collected in step 2 into the mathematical model of the ultrafiltration device trained in step 3, output the model calculation results, and determine whether the operation status of the ultrafiltration device is abnormal;

[0017] Step 5: Construct an ultrafiltration device mechanism model and obtain mechanism data, combine the ultrafiltration device mathematical model to output the switch quantity sequence, and perform early warning information analysis.

[0018] Furthermore, the relevant parameters in step 2 include the inlet flow rate of the ultrafiltration device, the inlet and outlet pressure difference of the ultrafiltration device, and the frequency feedback of the water pump of the ultrafiltration device.

[0019] Furthermore, the preprocessing measures in step 2 include operating condition division and principal component analysis.

[0020] Furthermore, the method of dividing the training data set and the test data set of the ultrafiltration device in step (2) in step 3 is as follows:

[0021] According to the historical operation data selected and screened in (1), the average value of the data at a certain time interval during operation is calculated as the modeling data set, and the training data set and the test data set of the mathematical model of the ultrafiltration device are divided, wherein the training data set is used to train the One-Class-SVM ultrafiltration device mathematical model, and the test data set is used to evaluate and adjust the One-Class-SVM ultrafiltration device mathematical model.

[0022] Furthermore, the calculation formula for verifying the accuracy of the mathematical model of the ultrafiltration device in step (4) of step 3 is as follows:

[0023]

[0024] Among them, accuracy is the model prediction accuracy, N accuracy The number of data samples predicted correctly by the model, N test is the total number of samples in the validation dataset.

[0025] Furthermore, the method for evaluating and adjusting the mathematical model of the ultrafiltration device (5) in step 4 includes: when the accuracy of the mathematical model of the ultrafiltration device verified in (4) is lower than the set value, returning to (3) to modify the parameters of the training model and repeating the training until the warning accuracy required by the mathematical model of the ultrafiltration device is reached.

[0026] Furthermore, the mechanism data in step 5 includes the ultrafiltration device water pump start-stop monitoring data, the ultrafiltration device start-stop monitoring data, and the number of abnormal states of the mathematical model.

[0027] Furthermore, the method for outputting the switching value sequence in step 5 is as follows:

[0028] a. If the analysis result of the mathematical model of the ultrafiltration device on the real-time operation data is "normal state", the mathematical model state output is 0, and the ultrafiltration device mechanism model is not enabled; if the analysis result of the mathematical model of the ultrafiltration device on the real-time operation data is "abnormal state", the mathematical model state output is 1;

[0029] b. If the ultrafiltration device water supply pump start-stop monitoring operation signal does not change, the ultrafiltration device water supply pump start-stop monitoring state outputs 0; if the ultrafiltration device water supply pump start-stop monitoring operation signal changes from 0 to 1 or the operation signal changes from 1 to 0, the ultrafiltration device water supply pump start-stop monitoring state outputs 1;

[0030] c. If the operating state of the water production valve of the ultrafiltration device changes from 0 to 1 or the operating state of the water production valve changes from 1 to 0 or the operating state of the water production valve remains at 0, the start-stop monitoring state of the ultrafiltration device outputs 0; if the operating state of the water production valve of the ultrafiltration device is 1, the start-stop monitoring state of the ultrafiltration device outputs 1;

[0031] d. If the number of consecutive abnormal states of the mathematical model does not exceed 3 times, the monitoring output of the number of consecutive abnormal states of the mathematical model is 0; if the number of consecutive abnormal states of the mathematical model exceeds 3 times, the monitoring output of the number of consecutive abnormal states of the mathematical model is 1.

[0032] Furthermore, the method for analyzing the warning information in step 5 is as follows:

[0033] a. When the state output of the mathematical model of the ultrafiltration device is 0, no warning information is issued;

[0034] b. When the state output of the UF device mathematical model is 1 and the start / stop monitoring state output of the UF device water pump is 1, no warning information is issued;

[0035] c. When the state output of the UF device mathematical model is 1, the start / stop monitoring state output of the UF device water pump is 0, and the start / stop monitoring state output of the UF device is 0, no warning information is issued;

[0036] d. When the state output of the UF device mathematical model is 1, the start and stop monitoring state output of the UF device water pump is 0, the start and stop monitoring state output of the UF device is 1, and the number of continuous abnormal state monitoring state output of the mathematical model is 0, no warning information is issued;

[0037] e. When the status output of the mathematical model of the ultrafiltration device is 1, the start and stop monitoring status output of the ultrafiltration device water pump is 0, the start and stop monitoring status output of the ultrafiltration device is 1, and the monitoring status output of the mathematical model for the number of consecutive abnormal states is 1, the ultrafiltration device is in an overloaded operation state and issues a warning message.

[0038] The present invention also provides an ultrafiltration device operation failure early warning system, comprising:

[0039] Historical data storage module, mathematical model training module, data acquisition and processing module, operation status diagnosis module, and early warning information analysis module;

[0040] The historical data storage module is used to summarize cases where the ultrafiltration device has serious fouling and flow rate exceeds the limit, collect and store historical operation data, and propose corresponding solutions and treatment measures for different types of operation failures;

[0041] The data acquisition and processing module is used to collect real-time operating data in the ultrafiltration device according to relevant parameters and perform relevant pre-processing measures;

[0042] The mathematical model training module is used to construct a mathematical model of an ultrafiltration device based on a One-Class-SVM classification algorithm, including:

[0043] Screen the historical operation data of the ultrafiltration device in the historical data storage module and perform relevant pre-processing measures;

[0044] Dividing the ultrafiltration device into a training data set and a test data set;

[0045] Set the support vector machine classification model parameters and train the mathematical model;

[0046] Verify the accuracy of the mathematical model of the ultrafiltration device;

[0047] Evaluate and adjust the mathematical model of ultrafiltration device;

[0048] The operation status diagnosis module is used to input the real-time operation data collected in the data collection and processing module into the trained ultrafiltration device mathematical model in the mathematical model training module, output the model calculation results, and judge whether the operation status of the ultrafiltration device is abnormal;

[0049] The early warning information analysis module is used to construct an ultrafiltration device mechanism model and obtain mechanism data, and output a switching quantity sequence in combination with the ultrafiltration device mathematical model to perform early warning information analysis.

[0050] The advantages of the present invention are: the present invention adopts the One-Class-SVM ultrafiltration device mathematical model of unsupervised learning method to judge the abnormal situation of the operation state of the ultrafiltration device, and obtains the mechanism data and generates the switch quantity sequence by constructing the ultrafiltration device mechanism model, and performs the ultrafiltration device early warning analysis in combination with the mathematical model. In the early stage of serious fouling or flow exceeding the limit of the ultrafiltration device, the abnormal information is pushed to the relevant personnel, so as to buy time for the relevant personnel to adjust the system operation condition in time, improve the system operation efficiency, avoid affecting the normal operation of the boiler feed water system, and increase the service life of the fiber component of the ultrafiltration device. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a flow chart of an ultrafiltration device operation failure early warning method and system according to an embodiment of the present invention;

[0052] Figure 2 It is a flow chart of a mathematical model of an ultrafiltration device according to an embodiment of the present invention;

[0053] Figure 3 This is a confusion matrix diagram of a test set according to an embodiment of the present invention;

[0054] Figure 4 This is a flow chart of the mechanism model of the ultrafiltration device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in combination with the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0056] The technical solution of the present invention is further described below in conjunction with the accompanying drawings and specific embodiments:

[0057] Embodiment 1

[0058] like Figure 1 Specifically, a method for early warning of an ultrafiltration device operation failure is disclosed, comprising:

[0059] Step 1: Summarize the cases of serious fouling and flow exceeding the limit of the ultrafiltration device, collect and store historical operation data, and propose corresponding solutions and treatment measures for different types of operation failures;

[0060] In this embodiment, a summary is made based on the relevant faults and causes that have occurred in the history of an ultrafiltration device for boiler feed water in a thermal power plant; by summarizing the cases of severe fouling and flow exceeding the limit in the ultrafiltration device, the most relevant parameter set is analyzed as the relevant parameters, and the historical operating data of the relevant parameters are collected and stored; corresponding solutions and optimal handling measures are proposed for different fault types in the ultrafiltration device operation failure cases.

[0061] Step 2: Collect real-time operating data in the ultrafiltration device according to relevant parameters and perform relevant pretreatment measures;

[0062] In this embodiment, the relevant parameters include parameters such as the inlet flow rate of the ultrafiltration device, the inlet and outlet pressure difference of the ultrafiltration device, and the frequency feedback of the water pump of the ultrafiltration device. The real-time operating status data selected for the ultrafiltration device include the inlet flow rate of the A ultrafiltration device, the inlet and outlet pressure difference of the A ultrafiltration device, the inlet flow rate of the B ultrafiltration device, the inlet and outlet pressure difference of the B ultrafiltration device, the frequency feedback of the water pump of the A ultrafiltration device, and the frequency feedback of the water pump of the B ultrafiltration device;

[0063] Among them, the preprocessing methods mainly include working condition division and principal component analysis.

[0064] Step 3, constructing a mathematical model of the ultrafiltration device based on the One-Class-SVM classification algorithm;

[0065] (1) screening the historical operation data of the ultrafiltration device in step 1 and performing relevant pretreatment measures;

[0066] Figure 2 This is a flow chart of the mathematical model of the ultrafiltration device of this embodiment. In this embodiment, by extracting the actual operation records of some measuring points in the feed water system of a thermal power plant for one year, the actual operation data of the ultrafiltration device with serious fouling and flow exceeding the limit are screened out as historical operation data.

[0067] (2) dividing the training data set and the test data set of the ultrafiltration device;

[0068] In this embodiment, the selected historical operation data is selected, and the average value of the data at every twenty seconds of operation is calculated as the modeling data set, and the training data set and the test data set of the mathematical model of the ultrafiltration device are divided, wherein the training data set is used to train the One-Class-SVM ultrafiltration device mathematical model, and the test data set is used to evaluate and adjust the One-Class-SVM ultrafiltration device mathematical model;

[0069] In this embodiment, a total of 85,457 ultrafiltration device samples (a set of data and labels is 1 sample) are generated, of which the number of training data set samples of the One-Class-SVM ultrafiltration device mathematical model is 64,093, and the number of test data set samples is 21,364.

[0070] (3) Setting the mathematical model parameters of the One-Class-SVM ultrafiltration device and training the mathematical model;

[0071] By analyzing the divided ultrafiltration device training set data, one or more early warning mathematical models of ultrafiltration device operation failures are established; different algorithms or multiple algorithms can be used for data processing, such as a mixture of one or more algorithms such as support vector machine algorithm, DBSCN algorithm, random forest algorithm, etc.;

[0072] The mathematical logic of support vector machine is to map the vector to a higher dimensional space, establish a maximum margin decision hyperplane, build two parallel hyperplanes on both sides of the hyperplane that separates the data, and maximize the distance between the two parallel hyperplanes by establishing a separation hyperplane with a suitable direction. The larger the distance or gap between the parallel hyperplanes, the smaller the total error of the classifier.

[0073] In this embodiment, a one-class support vector machine (One-Class-SVM) classifier is used to construct a mathematical model of an ultrafiltration device; the logic of the One-Class-SVM classifier is to use a hypersphere, hoping to minimize the volume of the hypersphere, rather than using a hyperplane for division as in the traditional SVM algorithm; the algorithm obtains a spherical boundary around the data in the feature space, which can solve high-dimensional problems in large feature spaces; compared to the commonly used random forest model in the field and other traditional classification and regression support vector machine models based on supervised learning, the One-Class-SVM classifier uses an unsupervised learning method and does not need to mark the output labels of the training set;

[0074] The One-Class-SVM classifier can require that the distance from all training data points xi to the center is strictly less than r. After solving the Lagrangian duality, it determines whether the new data point z is inside. If the distance from z to the center is less than or equal to the radius r, it is not an outlier. If the distance from z to the center is outside the hypersphere, it is identified as an outlier.

[0075] In this embodiment, the One-Class-SVM classifier model uses a Gaussian kernel function, the outlier point ratio is set to 0.005, the residual convergence condition is 0.001, and the maximum number of iterations is not limited.

[0076] (4) Verify the accuracy of the mathematical model of the ultrafiltration device;

[0077] The test data set is input into the trained mathematical model for verification and the accuracy of the model is calculated using the following formula:

[0078]

[0079] Among them, accuracy is the model prediction accuracy, N accuracy The number of data samples predicted correctly by the model, N test is the total number of samples in the validation dataset.

[0080] (5) Evaluate and adjust the mathematical model of the ultrafiltration device;

[0081] When (4) verifies that the accuracy of the mathematical model of the ultrafiltration device is lower than the set value, return to (3) to modify the parameters of the training model and repeat the training until the early warning accuracy required by the mathematical model of the ultrafiltration device is reached;

[0082] In this embodiment, the accuracy of the mathematical model of the ultrafiltration device is set to 90%; when the accuracy is less than 90%, the model parameters are modified and retrained;

[0083] Figure 3 This is the confusion matrix diagram of the test set of this embodiment. The ultrafiltration device test set contains 2 types of labels and 21,364 samples;

[0084] In the confusion matrix, 0 represents no abnormality, and 1 represents abnormal conditions such as fouling or overflow. Figure 3 As shown, the number of samples predicted correctly is 21219, the number of samples predicted incorrectly is 145, and the test accuracy of the ultrafiltration device support vector machine classification model can reach 99.32%. Therefore, the mathematical model of the ultrafiltration device has a good predictive value.

[0085] Step 4, input the real-time operation data collected in step 2 into the mathematical model of the ultrafiltration device trained in step 3, output the model calculation results, and determine whether the operation status of the ultrafiltration device is abnormal;

[0086] In this embodiment, if the analysis result of the mathematical model of the ultrafiltration device on the real-time operation data is "normal state", the mathematical model state output is 0; if the analysis result of the mathematical model of the ultrafiltration device on the real-time operation data is "abnormal state", the mathematical model state output is 1.

[0087] Step 5: construct an ultrafiltration device mechanism model and obtain mechanism data, combine the ultrafiltration device mathematical model to output the switch quantity sequence, and perform early warning information analysis;

[0088] In this embodiment, due to data transmission fluctuations, the real-time data collected has large noise, and there is a possibility of false alarm when it is transmitted to the mathematical model of the ultrafiltration device. Therefore, it is necessary to analyze the early warning information in combination with the operating status of the relevant equipment in the current ultrafiltration device. By constructing the mechanism model of the ultrafiltration device, obtaining the mechanism data and generating the sequence switch quantity, it is determined whether it is necessary to issue an early warning information;

[0089] In this embodiment, the mechanism data includes the ultrafiltration device water pump start-stop monitoring data, the ultrafiltration device start-stop monitoring data, and the number of abnormal state times of the ultrafiltration device mathematical model.

[0090] In this embodiment, the analysis logic of the ultrafiltration device mathematical model and mechanism model is as follows:

[0091] If the analysis result of the ultrafiltration device mathematical model on the real-time operation data is "normal state", the mathematical model state output is 0, and the ultrafiltration device mechanism model is not enabled; if the analysis result of the ultrafiltration device mathematical model on the real-time operation data is "abnormal state", the mathematical model state output is 1;

[0092] Ultrafiltration device water pump start and stop monitoring: monitor the ultrafiltration device water pump operation signal. When the operation signal does not change, the ultrafiltration device water pump start and stop monitoring status outputs 0; when the operation signal changes from 0 to 1 or the operation signal changes from 1 to 0, the ultrafiltration device water pump start and stop monitoring status outputs 1;

[0093] Ultrafiltration device start-stop monitoring: monitor the operating status of the water production valve of the ultrafiltration device. When the operating status of the water production valve changes from 0 to 1 or the operating status of the water production valve changes from 1 to 0 or the operating status of the water production valve remains at 0, the ultrafiltration device start-stop monitoring status outputs 0; when the operating status of the water production valve is 1, the status outputs 1;

[0094] Monitoring of the number of consecutive abnormal states of the mathematical model: monitoring the number of abnormal states of the mathematical model. When the number of consecutive abnormal states of the mathematical model does not exceed 3 times, the monitoring output of the number of consecutive abnormal states of the mathematical model is 0; when the number of consecutive abnormal states of the mathematical model exceeds 3 times, the monitoring output of the number of consecutive abnormal states of the mathematical model is 1;

[0095] Figure 4 This is a flow chart of the mechanism model of the ultrafiltration device of this embodiment. After the state of the mathematical model of the ultrafiltration device is output, the ultrafiltration device needs to analyze the operating state with the help of the mechanism model, and combine the switch sequence output by the mathematical model and the mechanism model to determine whether it is necessary to issue an early warning message. The switch sequence determination method is as follows:

[0096] When the state output of the mathematical model of the ultrafiltration device is 0, the output result of the mathematical model of the ultrafiltration device is "normal state" and no warning information is issued;

[0097] When the state output of the UF device mathematical model is 1 and the start / stop monitoring state output of the UF device water pump is 1, no warning information is issued;

[0098] When the state output of the mathematical model of the ultrafiltration device is 1, the start-stop monitoring state output of the ultrafiltration device water pump is 0, and the start-stop monitoring state output of the ultrafiltration device is 0, no warning information is issued;

[0099] When the state output of the mathematical model of the ultrafiltration device is 1, the start and stop monitoring state output of the water pump of the ultrafiltration device is 0, the start and stop monitoring state output of the ultrafiltration device is 1, and the monitoring state output of the continuous abnormal state times of the mathematical model is 0, no warning information is issued;

[0100] When the status output of the mathematical model of the ultrafiltration device is 1, the start and stop monitoring status output of the ultrafiltration device water pump is 0, the start and stop monitoring status output of the ultrafiltration device is 1, and the monitoring status output of the mathematical model for the number of consecutive abnormal states is 1, the ultrafiltration device is in an overloaded operating state and issues an early warning message to remind personnel to adjust the operating parameters as soon as possible.

[0101] When the ultrafiltration device is seriously clogged or the flow exceeds the limit, based on the collected real-time operation data and the switch quantity sequence composed of the mathematical model and mechanism model of the ultrafiltration device, it is determined whether the ultrafiltration device needs to issue an early warning information, and the early warning information and adjustment measures are pushed to relevant personnel in a timely manner.

[0102] Embodiment 2

[0103] This embodiment discloses an ultrafiltration device operation fault warning system, including: a historical data storage module, a data acquisition and processing module, a mathematical model construction module, an operation status diagnosis module, and a warning information analysis module;

[0104] The historical data storage module is used to summarize cases where the ultrafiltration device has serious fouling and flow rate exceeds the limit, collect and store historical operation data, and propose corresponding solutions and treatment measures for different types of operation failures;

[0105] In this embodiment, a summary is made based on the relevant faults and causes that have occurred in the history of an ultrafiltration device for boiler feed water in a thermal power plant; by summarizing the cases of severe fouling and flow exceeding the limit in the ultrafiltration device, the most relevant parameter set is analyzed as the relevant parameters, and the historical operating data of the relevant parameters are collected and stored; corresponding solutions and optimal handling measures are proposed for different fault types in the ultrafiltration device operation failure cases.

[0106] The data acquisition and processing module is used to collect real-time operating data in the ultrafiltration device according to relevant parameters and perform relevant pre-processing measures;

[0107] In this embodiment, the relevant parameters include parameters such as the inlet flow rate of the ultrafiltration device, the inlet and outlet pressure difference of the ultrafiltration device, and the frequency feedback of the water pump of the ultrafiltration device. The real-time operating status data selected for the ultrafiltration device include the inlet flow rate of the A ultrafiltration device, the inlet and outlet pressure difference of the A ultrafiltration device, the inlet flow rate of the B ultrafiltration device, the inlet and outlet pressure difference of the B ultrafiltration device, the frequency feedback of the water pump of the A ultrafiltration device, and the frequency feedback of the water pump of the B ultrafiltration device;

[0108] Among them, the preprocessing methods mainly include working condition division and principal component analysis.

[0109] The mathematical model training module is used to construct a mathematical model of an ultrafiltration device based on a One-Class-SVM classification algorithm;

[0110] Screen the historical operation data of the ultrafiltration device in the historical data storage module and perform relevant pre-processing measures;

[0111] Figure 2 This is a flow chart of the mathematical model of the ultrafiltration device of this embodiment. In this embodiment, by extracting the actual operation records of some measuring points in the feed water system of a thermal power plant for one year, the actual operation data of the ultrafiltration device with serious fouling and flow exceeding the limit are screened out as historical operation data.

[0112] Dividing the ultrafiltration device into a training data set and a test data set;

[0113] In this embodiment, the selected historical operation data is selected, and the average value of the data at every twenty seconds of operation is calculated as the modeling data set, and the training data set and the test data set of the mathematical model of the ultrafiltration device are divided, wherein the training data set is used to train the One-Class-SVM ultrafiltration device mathematical model, and the test data set is used to evaluate and adjust the One-Class-SVM ultrafiltration device mathematical model;

[0114] In this embodiment, a total of 85,457 ultrafiltration device samples (a set of data and labels is 1 sample) are generated, of which the number of training data set samples of the One-Class-SVM ultrafiltration device mathematical model is 64,093, and the number of test data set samples is 21,364.

[0115] Set the support vector machine classification model parameters and train the mathematical model;

[0116] By analyzing the divided ultrafiltration device training set data, one or more early warning mathematical models of ultrafiltration device operation failures are established; different algorithms or multiple algorithms can be used for data processing, such as a mixture of one or more algorithms such as support vector machine algorithm, DBSCN algorithm, random forest algorithm, etc.;

[0117] The mathematical logic of support vector machine is to map the vector to a higher dimensional space, establish a maximum margin decision hyperplane, build two parallel hyperplanes on both sides of the hyperplane that separates the data, and maximize the distance between the two parallel hyperplanes by establishing a separation hyperplane with a suitable direction. The larger the distance or gap between the parallel hyperplanes, the smaller the total error of the classifier.

[0118] In this embodiment, a one-class support vector machine (One-Class-SVM) classifier is used to construct a mathematical model of an ultrafiltration device; the logic of the One-Class-SVM classifier is to use a hypersphere, hoping to minimize the volume of the hypersphere, rather than using a hyperplane for division as in the traditional SVM algorithm; the algorithm obtains a spherical boundary around the data in the feature space, which can solve high-dimensional problems in large feature spaces; compared to the commonly used random forest model in the field and other traditional classification and regression support vector machine models based on supervised learning, the One-Class-SVM classifier uses an unsupervised learning method and does not need to mark the output labels of the training set;

[0119] The One-Class-SVM classifier can require that the distance from all training data points xi to the center is strictly less than r. After solving the Lagrangian duality, it determines whether the new data point z is inside. If the distance from z to the center is less than or equal to the radius r, it is not an outlier. If the distance from z to the center is outside the hypersphere, it is identified as an outlier.

[0120] In this embodiment, the One-Class-SVM classifier model uses a Gaussian kernel function, the outlier point ratio is set to 0.005, the residual convergence condition is 0.001, and the maximum number of iterations is not limited.

[0121] Verify the accuracy of the mathematical model of the ultrafiltration device;

[0122] The test data set is input into the trained mathematical model for verification and the accuracy of the model is calculated using the following formula:

[0123]

[0124] Among them, accuracy is the model prediction accuracy, N accuracy The number of data samples predicted correctly by the model, N test is the total number of samples in the validation dataset.

[0125] Evaluate and adjust the mathematical model of ultrafiltration device;

[0126] When the accuracy of the mathematical model of the ultrafiltration device is verified to be lower than the set value, the parameters of the training model are modified in the mathematical model of the ultrafiltration device to repeat the training until the early warning accuracy required by the mathematical model of the ultrafiltration device is reached;

[0127] In this embodiment, the accuracy of the mathematical model of the ultrafiltration device is set to 90%; when the accuracy is less than 90%, the model parameters are modified and retrained;

[0128] Figure 3This is the confusion matrix diagram of the test set of this embodiment. The ultrafiltration device test set contains 2 types of labels and 21,364 samples;

[0129] In the confusion matrix, 0 represents no abnormality, and 1 represents abnormal conditions such as fouling or overflow. Figure 3 As shown, the number of samples predicted correctly is 21219, the number of samples predicted incorrectly is 145, and the test accuracy of the ultrafiltration device support vector machine classification model can reach 99.32%. Therefore, the mathematical model of the ultrafiltration device has a good predictive value.

[0130] The operation status diagnosis module is used to input the real-time operation data collected in the data collection and processing module into the trained ultrafiltration device mathematical model in the mathematical model training module, output the model calculation results, and judge whether the operation status of the ultrafiltration device is abnormal;

[0131] In this embodiment, if the analysis result of the mathematical model of the ultrafiltration device on the real-time operation data is "normal state", the mathematical model state output is 0; if the analysis result of the mathematical model of the ultrafiltration device on the real-time operation data is "abnormal state", the mathematical model state output is 1.

[0132] The early warning information analysis module is used to construct an ultrafiltration device mechanism model and obtain mechanism data, and output a switching quantity sequence in combination with the ultrafiltration device mathematical model to perform early warning information analysis.

[0133] In this embodiment, due to data transmission fluctuations, the real-time data collected has large noise, and there is a possibility of false alarm when it is transmitted to the mathematical model of the ultrafiltration device. Therefore, it is necessary to analyze the early warning information in combination with the operating status of the relevant equipment in the current ultrafiltration device. By constructing the mechanism model of the ultrafiltration device, obtaining the mechanism data and generating the sequence switch quantity, it is determined whether it is necessary to issue an early warning information;

[0134] In this embodiment, the mechanism data includes the ultrafiltration device water pump start-stop monitoring data, the ultrafiltration device start-stop monitoring data, and the number of abnormal state times of the ultrafiltration device mathematical model.

[0135] In this embodiment, the analysis logic of the ultrafiltration device mathematical model and mechanism model is as follows:

[0136] If the analysis result of the ultrafiltration device mathematical model on the real-time operation data is "normal state", the mathematical model state output is 0, and the ultrafiltration device mechanism model is not enabled; if the analysis result of the ultrafiltration device mathematical model on the real-time operation data is "abnormal state", the mathematical model state output is 1;

[0137] Ultrafiltration device water pump start and stop monitoring: monitor the ultrafiltration device water pump operation signal. When the operation signal does not change, the ultrafiltration device water pump start and stop monitoring status outputs 0; when the operation signal changes from 0 to 1 or the operation signal changes from 1 to 0, the ultrafiltration device water pump start and stop monitoring status outputs 1;

[0138] Ultrafiltration device start-stop monitoring: monitor the operating status of the water production valve of the ultrafiltration device. When the operating status of the water production valve changes from 0 to 1 or the operating status of the water production valve changes from 1 to 0 or the operating status of the water production valve remains at 0, the ultrafiltration device start-stop monitoring status outputs 0; when the operating status of the water production valve is 1, the status outputs 1;

[0139] Monitoring of the number of consecutive abnormal states of the mathematical model: monitoring the number of abnormal states of the mathematical model. When the number of consecutive abnormal states of the mathematical model does not exceed 3 times, the monitoring output of the number of consecutive abnormal states of the mathematical model is 0; when the number of consecutive abnormal states of the mathematical model exceeds 3 times, the monitoring output of the number of consecutive abnormal states of the mathematical model is 1;

[0140] Figure 4 This is a flow chart of the mechanism model of the ultrafiltration device of this embodiment. After the state of the mathematical model of the ultrafiltration device is output, the ultrafiltration device needs to analyze the operating state with the help of the mechanism model, and combine the switch sequence output by the mathematical model and the mechanism model to determine whether it is necessary to issue an early warning message. The switch sequence determination method is as follows:

[0141] When the state output of the mathematical model of the ultrafiltration device is 0, the output result of the mathematical model of the ultrafiltration device is "normal state" and no warning information is issued;

[0142] When the state output of the UF device mathematical model is 1 and the start / stop monitoring state output of the UF device water pump is 1, no warning information is issued;

[0143] When the state output of the mathematical model of the ultrafiltration device is 1, the start-stop monitoring state output of the ultrafiltration device water pump is 0, and the start-stop monitoring state output of the ultrafiltration device is 0, no warning information is issued;

[0144] When the state output of the mathematical model of the ultrafiltration device is 1, the start and stop monitoring state output of the water pump of the ultrafiltration device is 0, the start and stop monitoring state output of the ultrafiltration device is 1, and the monitoring state output of the continuous abnormal state times of the mathematical model is 0, no warning information is issued;

[0145] When the status output of the mathematical model of the ultrafiltration device is 1, the start and stop monitoring status output of the ultrafiltration device water pump is 0, the start and stop monitoring status output of the ultrafiltration device is 1, and the monitoring status output of the mathematical model for the number of consecutive abnormal states is 1, the ultrafiltration device is in an overloaded operating state and issues an early warning message to remind personnel to adjust the operating parameters as soon as possible.

[0146] When the ultrafiltration device is seriously clogged or the flow exceeds the limit, based on the collected real-time operation data and the switch quantity sequence composed of the mathematical model and mechanism model of the ultrafiltration device, it is determined whether the ultrafiltration device needs to issue an early warning information, and the early warning information and adjustment measures are pushed to relevant personnel in a timely manner.

[0147] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for early warning of ultrafiltration device operation failure, characterized in that: The method comprises: Step 1: Summarize the cases of serious fouling and flow exceeding the limit of the ultrafiltration device, collect and store historical operation data, and propose corresponding solutions and treatment measures for different types of operation failures; Step 2: Collect real-time operating data in the ultrafiltration device according to relevant parameters and perform relevant pretreatment measures; Step 3: construct a mathematical model of the ultrafiltration device based on the One-Class-SVM classification algorithm, including: (1) screening the historical operation data of the ultrafiltration device in step 1 and performing relevant pretreatment measures; (2) dividing the training data set and the test data set of the ultrafiltration device; (3) Set the support vector machine classification model parameters and train the mathematical model; (4) Verify the accuracy of the mathematical model of the ultrafiltration device; (5) Evaluate and adjust the mathematical model of the ultrafiltration device; Step 4, input the real-time operation data collected in step 2 into the mathematical model of the ultrafiltration device trained in step 3, output the model calculation results, and determine whether the operation status of the ultrafiltration device is abnormal; Step 5: Construct an ultrafiltration device mechanism model and obtain mechanism data, combine the ultrafiltration device mathematical model to output the switch quantity sequence, and perform early warning information analysis.

2. The method for early warning of an ultrafiltration device operation failure according to claim 1, characterized in that: The relevant parameters in step 2 include the inlet flow rate of the ultrafiltration device, the inlet and outlet pressure difference of the ultrafiltration device, and the frequency feedback of the water pump of the ultrafiltration device.

3. The method for early warning of an ultrafiltration device operation failure according to claim 1, characterized in that: The preprocessing measures described in step 2 include operating condition division and principal component analysis.

4. The method for early warning of an ultrafiltration device operation failure according to claim 3, characterized in that: The method of dividing the training data set and the test data set of the ultrafiltration device in step (2) in step 3 is as follows: According to the historical operation data selected and screened in (1), the average value of the data at a certain time interval during operation is calculated as the modeling data set, and the training data set and the test data set of the mathematical model of the ultrafiltration device are divided, wherein the training data set is used to train the One-Class-SVM ultrafiltration device mathematical model, and the test data set is used to evaluate and adjust the One-Class-SVM ultrafiltration device mathematical model.

5. The method for early warning of an ultrafiltration device operation failure according to claim 4, characterized in that: The calculation formula for verifying the accuracy of the mathematical model of the ultrafiltration device described in step 3 (4) is as follows: Among them, accuracy is the model prediction accuracy, N accuracy The number of data samples predicted correctly by the model, N test is the total number of samples in the validation dataset.

6. The method for early warning of an ultrafiltration device operation failure according to claim 5, characterized in that: The method for evaluating and adjusting the mathematical model of the ultrafiltration device (5) in step 4 includes: when the accuracy of the mathematical model of the ultrafiltration device verified in (4) is lower than the set value, returning to (3) to modify the parameters of the training model and repeat the training until the warning accuracy required by the mathematical model of the ultrafiltration device is reached.

7. The method for early warning of an ultrafiltration device operation failure according to claim 6, characterized in that: The mechanism data in step 5 include the start and stop monitoring data of the ultrafiltration device water pump, the start and stop monitoring data of the ultrafiltration device, and the number of abnormal states of the mathematical model.

8. The method for early warning of an ultrafiltration device operation failure according to claim 7, characterized in that: The method for outputting the switching sequence described in step 5 is as follows: a. If the analysis result of the mathematical model of the ultrafiltration device on the real-time operation data is "normal state", the mathematical model state output is 0, and the ultrafiltration device mechanism model is not enabled; if the analysis result of the mathematical model of the ultrafiltration device on the real-time operation data is "abnormal state", the mathematical model state output is 1; b. If the ultrafiltration device water supply pump start-stop monitoring operation signal does not change, the ultrafiltration device water supply pump start-stop monitoring state outputs 0; if the ultrafiltration device water supply pump start-stop monitoring operation signal changes from 0 to 1 or the operation signal changes from 1 to 0, the ultrafiltration device water supply pump start-stop monitoring state outputs 1; c. If the operating state of the water production valve of the ultrafiltration device changes from 0 to 1 or the operating state of the water production valve changes from 1 to 0 or the operating state of the water production valve remains at 0, the start-stop monitoring state of the ultrafiltration device outputs 0; if the operating state of the water production valve of the ultrafiltration device is 1, the start-stop monitoring state of the ultrafiltration device outputs 1; d. If the number of consecutive abnormal states of the mathematical model does not exceed 3 times, the monitoring output of the number of consecutive abnormal states of the mathematical model is 0; if the number of consecutive abnormal states of the mathematical model exceeds 3 times, the monitoring output of the number of consecutive abnormal states of the mathematical model is 1.

9. The method for early warning of an ultrafiltration device operation failure according to claim 8, characterized in that: The method for analyzing the warning information described in step 5 is as follows: a. When the state output of the mathematical model of the ultrafiltration device is 0, no warning information is issued; b. When the state output of the UF device mathematical model is 1 and the start / stop monitoring state output of the UF device water pump is 1, no warning information is issued; c. When the state output of the UF device mathematical model is 1, the start / stop monitoring state output of the UF device water pump is 0, and the start / stop monitoring state output of the UF device is 0, no warning information is issued; d. When the state output of the UF device mathematical model is 1, the start and stop monitoring state output of the UF device water pump is 0, the start and stop monitoring state output of the UF device is 1, and the number of continuous abnormal state monitoring state output of the mathematical model is 0, no warning information is issued; e. When the status output of the mathematical model of the ultrafiltration device is 1, the start and stop monitoring status output of the ultrafiltration device water pump is 0, the start and stop monitoring status output of the ultrafiltration device is 1, and the monitoring status output of the mathematical model for the number of consecutive abnormal states is 1, the ultrafiltration device is in an overloaded operation state and issues a warning message.

10. An ultrafiltration device operation failure warning system, characterized in that: include: Historical data storage module, mathematical model training module, data acquisition and processing module, operation status diagnosis module, and early warning information analysis module; The historical data storage module is used to summarize cases where the ultrafiltration device has serious fouling and flow rate exceeds the limit, collect and store historical operation data, and propose corresponding solutions and treatment measures for different types of operation failures; The data acquisition and processing module is used to collect real-time operating data in the ultrafiltration device according to relevant parameters and perform relevant pre-processing measures; The mathematical model training module is used to construct a mathematical model of an ultrafiltration device based on a One-Class-SVM classification algorithm, including: Screen the historical operation data of the ultrafiltration device in the historical data storage module and perform relevant pre-processing measures; Dividing the ultrafiltration device into a training data set and a test data set; Set the support vector machine classification model parameters and train the mathematical model; Verify the accuracy of the mathematical model of the ultrafiltration device; Evaluate and adjust the mathematical model of ultrafiltration device; The operation status diagnosis module is used to input the real-time operation data collected in the data collection and processing module into the trained ultrafiltration device mathematical model in the mathematical model training module, output the model calculation results, and judge whether the operation status of the ultrafiltration device is abnormal; The early warning information analysis module is used to construct an ultrafiltration device mechanism model and obtain mechanism data, and output a switching quantity sequence in combination with the ultrafiltration device mathematical model to perform early warning information analysis.

Images