A real-time monitoring method and system for a radioactive wastewater treatment platform

Abstract

The present application relates to sewage treatment technical field, especially a kind of real-time monitoring method and system for radioactive sewage treatment platform.It comprises the following steps: constructing the endpoint defined period of each time in sewage data time sequence, calculating the exceeding standard index of radionuclide concentration at each time;In response to the exceeding standard index of radionuclide concentration at a time being greater than index threshold, obtain the ratio of the minimum value of sewage flow standard deviation in the endpoint defined period of the time and the minimum value of sewage flow standard deviation in the historical period of the time, denoted as flow index, to obtain the actual exceeding standard index of radionuclide concentration at the time;Calculate the k distance of each time, the k distance of each time is negatively correlated with the average of actual exceeding standard index of radionuclide concentration within the default k distance of the time;In LOF algorithm, the monitoring result of each time is obtained by using the k distance of each time to construct k distance range, effectively improve the accuracy of real-time monitoring result of radioactive sewage.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of sewage treatment, and in particular to a real-time monitoring method and system for a radioactive sewage treatment platform. BACKGROUND

[0002] In the fields of nuclear energy utilization, hospitals and petroleum and chemical energy, daily operation will produce sewage containing radioactive substances, i.e. nuclear sewage. These nuclear sewage contains various radioactive isotopes, has long-term harmfulness and is difficult to degrade. If not treated and directly discharged, it may cause serious harm to the ecological environment and human health.

[0003] At present, in the field of nuclear sewage treatment, membrane treatment technology is often used to intercept sewage. This technology has high interception characteristics. According to the current characteristics of radioactive molecules, a filter membrane with a suitable pore size is selected to achieve the filtering effect. However, the radioactive sewage often contains other impurities or pollutants, which may cause some pores on the membrane surface to be blocked. The sewage flows to the unblocked area of the pores, causing the pressure in this area to increase. Under the impact of high flow rate, some radioactive molecules that should be intercepted by the membrane may flow out, resulting in poor membrane treatment effect. Therefore, it is necessary to monitor the data collected by the radioactive sewage treatment platform in real time.

[0004] When the membrane treatment equipment is blocked, the radioactivity concentration of the water area will increase, deviating from the normal situation, and such conditions will further intensify. The normal data changes relatively stably, and the local density is relatively large. The addition of abnormal data will break this stability, causing the local density of the area to decrease. Therefore, the change in local density can be captured by the LOF (Local Outlier Factor) algorithm to identify abnormal points. However, if the K value in the LOF algorithm is not properly set, some outlier features in the local area may not be identified or normal minor changes may be identified as outlier points.

[0005] Therefore, how to accurately adjust the K value according to the characteristics of the nuclear sewage data to effectively improve the accuracy of the real-time monitoring result of the radioactive sewage treatment platform is a problem to be solved at present. SUMMARY

[0006] To solve the technical problem of how to accurately adjust the K value according to the characteristics of the nuclear sewage data to effectively improve the accuracy of the real-time monitoring result of the radioactive sewage treatment platform, the present application provides a real-time monitoring method and system for a radioactive sewage treatment platform.

[0007] In a first aspect, the present application provides a real-time monitoring method for a radioactive sewage treatment platform, which adopts the following technical solution:

[0008] A real-time monitoring method for a radioactive wastewater treatment platform, comprising the steps of:

[0009] acquiring the radionuclide concentration corresponding to each time in the wastewater data time sequence and the wastewater flow; constructing an endpoint defining period for each time, calculating the radionuclide concentration exceeding index for each time according to the radionuclide concentration difference between each time and the time in the endpoint defining period thereof and the time interval; in response to the radionuclide concentration exceeding index of a time being greater than an index threshold, acquiring the ratio of the wastewater flow standard deviation in the endpoint defining period of the time to the minimum value of the wastewater flow standard deviation in the historical period of the time, denoted as a flow indicator; taking the flow indicator as the index input of the radionuclide concentration exceeding index of the time to obtain the actual radionuclide concentration exceeding index of the time; calculating the k-distance of each time, and the k-distance of each time is negatively correlated with the average of the actual radionuclide concentration exceeding index within the default k-distance of the time; using the k-distance of each time to construct a k-distance range in the LOF algorithm to obtain the monitoring result of each time.

[0010] The present application can accurately filter out the abnormal time of the radionuclide concentration by identifying the abnormal density in the wastewater data time sequence through the LOF algorithm. In the process of acquiring the k-distance of each time, the present application analyzes the change trend of the radionuclide concentration in the endpoint defining period by the pollution change characteristics of the filter membrane, constructs the current radionuclide concentration abnormal exceeding index, provides a basis for subsequent real outlier identification, and reduces the monitoring result error caused by improper k-distance setting. On this basis, the present application can effectively distinguish the short-term breakthrough effect caused by the sudden increase of the flow from the real membrane pollution phenomenon by introducing the wastewater flow change in the endpoint defining period of the current time as the flow indicator to correct the exceeding index, reduce the false positive rate, and effectively improve the accuracy of the real-time monitoring result of the radioactive wastewater treatment platform.

[0011] According to the real-time monitoring method for a radioactive wastewater treatment platform provided by the present application, the endpoint defining period of each time is constructed, including: presetting the endpoint defining period length of each time as ; taking a time as the end of the endpoint defining period of the time, and sequentially acquiring times from the historical times on the left side of the time to construct the endpoint defining period of the time.

[0012] According to the real-time monitoring method for a radioactive wastewater treatment platform provided by the present application, the radionuclide concentration exceeding index of each time is calculated, including:

[0013] ;

[0014] the radionuclide concentration exceeding index of the i th time, define the length of the period for the endpoint, , the radionuclide concentration difference between the i-th moment and the moment in the endpoint-defined period thereof, the time interval, , , the maximum radionuclide concentration difference between the i-th moment and each moment in the endpoint-defined period thereof, the maximum time interval, is a hyperbolic tangent function.

[0015] The present application provides a precise radionuclide concentration over-standard index calculation method, which analyzes whether the radionuclide concentration in the endpoint-defined period of each moment changes significantly with time, thereby accurately obtaining the possibility of abnormal over-standard.

[0016] According to the real-time monitoring method for the radioactive wastewater treatment platform provided by the present application, in response to the radionuclide concentration over-standard index of a moment being not greater than the index threshold value, the radionuclide concentration over-standard index of the moment is taken as the radionuclide concentration actual over-standard index of the moment.

[0017] The present application considers that in the case where the radionuclide concentration over-standard index of a moment is greater than the index threshold value, the abnormal over-standard source needs to be analyzed to obtain the radionuclide concentration actual over-standard index, but for the moment not greater than the index threshold value, the radionuclide concentration actual over-standard index can be directly obtained, thereby effectively reducing the data processing amount and improving the algorithm efficiency.

[0018] According to the real-time monitoring method for the radioactive wastewater treatment platform provided by the present application, the calculation of the k distance of each moment further comprises: performing maximum-minimum normalization processing on the radionuclide concentration of each moment in the endpoint-defined period of each moment to obtain the normalized concentration value of each moment; and taking each moment, the corresponding normalized concentration value and the radionuclide concentration actual over-standard index as a feature point.

[0019] According to the real-time monitoring method for the radioactive wastewater treatment platform provided by the present application, the calculation of the k distance of each moment comprises:

[0020] ;

[0021] is the k distance of the i-th moment, is a default k distance, is the radionuclide concentration actual over-standard index mean value in the default k distance of the i-th moment, is an exponential function with base e.

[0022] The application provides a precise k-distance calculation method at each moment, adjusts the default k-distance by analyzing the actual exceeding index of the radionuclide concentration in the default k-distance in the LOF algorithm, so that the obtained k-distance at each moment is more in line with the actual data distribution, thereby improving the accuracy of abnormal monitoring.

[0023] According to the real-time monitoring method for the radioactive sewage treatment platform provided by the application, the k-distance range is constructed by using the k-distance at each moment in the LOF algorithm, and the monitoring result at each moment is obtained, which comprises the following steps: obtaining the Euclidean distance between the feature points and other feature points in the endpoint defining period of a feature point, and regarding the feature points with the Euclidean distance less than the k-distance of the feature point at the corresponding moment as the feature points in the k-distance range of the feature point; calculating the local outlier factor in the k-distance range of the feature point, and obtaining the monitoring result at the corresponding moment of each feature point according to the comparison result of the local outlier factor and the outlier threshold.

[0024] The application considers that the sewage density change at the latest moment is only related to the distribution of the feature points in the endpoint defining period thereof, so that the feature points in the k-distance range are obtained only in the endpoint defining period, so that the distribution in the endpoint defining period at each moment can be accurately measured, and the abnormal discrete points can be accurately extracted.

[0025] According to the real-time monitoring method for the radioactive sewage treatment platform provided by the application, the calculation of the local outlier factor in the k-distance range of the feature point comprises the following steps: obtaining the reachable distance mean value between the feature point and each feature point in the k-distance range of the feature point, and taking the reciprocal of the reachable distance mean value as the local outlier factor in the k-distance range of the feature point.

[0026] According to the real-time monitoring method for the radioactive sewage treatment platform provided by the application, the monitoring result at the corresponding moment of each feature point is obtained according to the comparison result of the local outlier factor and the outlier threshold, which comprises the following steps: setting the outlier threshold; if the local outlier factor of a feature point is greater than the outlier threshold, the feature point is an abnormal outlier point, and the monitoring result at the corresponding moment of the feature point is abnormal; otherwise, the monitoring result at the corresponding moment of the feature point is normal.

[0027] In the second aspect, the application provides a real-time monitoring system for a radioactive sewage treatment platform, which adopts the following technical scheme:

[0028] A real-time monitoring system for a radioactive sewage treatment platform comprises a processor and a memory, and the memory stores computer program instructions, when the computer program instructions are executed by the processor, the above-mentioned real-time monitoring method for the radioactive sewage treatment platform is realized.

[0029] By adopting the technical scheme, the computer program of the real-time monitoring method for the radioactive wastewater treatment platform is generated and stored in the memory to be loaded and executed by the processor, so that the terminal equipment is manufactured according to the memory and the processor, and use is facilitated.

[0030] The present application has the following technical effects:

[0031] Based on the above technical scheme, the present application provides a real-time monitoring method and system for a radioactive wastewater treatment platform. The LOF algorithm is used to identify the abnormal density in the wastewater data time sequence, which can accurately filter out the abnormal moments of radionuclide concentration. In the process of obtaining the k-distance of each moment, the present application analyzes the change trend of radionuclide concentration in the endpoint defining period through the pollution change characteristics of the filter membrane, constructs the current radionuclide concentration anomaly exceeding index, provides a basis for subsequent real outlier identification, and reduces the monitoring result error caused by improper k-distance setting. On this basis, the present application introduces the wastewater flow change in the endpoint defining period of the current moment as a flow index to correct the exceeding index, which can effectively distinguish the short-term breakthrough effect caused by flow surge from the real membrane pollution phenomenon, reduce the false positive rate, and effectively improve the accuracy of the real-time monitoring result of the radioactive wastewater treatment platform. BRIEF DESCRIPTION OF DRAWINGS

[0032] Figure 1 A flowchart in a real-time monitoring method for a radioactive wastewater treatment platform is provided for the embodiments of the present application. DETAILED DESCRIPTION

[0033] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments.

[0034] The embodiments of the present application disclose a real-time monitoring method for a radioactive wastewater treatment platform, which is specifically described in the following Figure 1 , Figure 1 A flowchart in a real-time monitoring method for a radioactive wastewater treatment platform is provided for the embodiments of the present application. The method can accurately filter out outliers in normal and abnormal segments by adaptively calculating the k-distance of each moment in the LOF algorithm, effectively improve the accuracy of real-time monitoring of the radioactive wastewater treatment platform, and specifically include the following steps:

[0035] S1: Obtain the radionuclide concentration corresponding to each moment in the wastewater data time sequence.

[0036] It should be noted that the principle of radioactive wastewater treatment is to use membrane treatment technology to efficiently intercept radioactive molecules in wastewater. For example, hospital medical wastewater may contain different types of radioactive substances, such as iodine-131, cesium-137, strontium-90, and cobalt-60 molecules. The structure of these radioactive molecules is obviously larger than that of water molecules. By selecting suitable membrane equipment according to the structural characteristics of the radioactive molecules that may exist in different scenarios, the separation of radioactive molecules and water molecules in radioactive wastewater can be ensured.

[0037] For example, in the embodiment of the present application, the radionuclide concentration corresponding to each time in the wastewater data time sequence and the wastewater flow are obtained, including: collecting the radionuclide concentration in the wastewater at each time, and arranging from left to right after preprocessing according to the collection order to obtain the wastewater data time sequence.

[0038] The preprocessing method can be interpolation of missing data, and can be set according to actual needs.

[0039] Specifically, the in-situ monitoring sensor is placed in the water area after the membrane equipment treatment, and the radionuclide concentration after the membrane treatment is collected according to the fixed collection frequency and collection time.

[0040] The collection frequency can be set to collect data once every second, and the collection time can be set to 1 day. The collection frequency and collection time can be set according to actual needs, and the embodiment of the present application does not make too many limitations here.

[0041] It can be understood that if the collection time is 1 day, the time interval between the start collection time and the end collection time of the wastewater data time sequence obtained after each collection is 1 day. The historical data at the current time is all the historical times in the wastewater data time sequence at this time.

[0042] It should be noted that when the membrane equipment normally processes nuclear wastewater, the radioactive molecules can be effectively separated, so the radionuclide concentration in the water area after membrane treatment is in a stable state. However, during the treatment process, the nuclear elements in the wastewater will slowly accumulate, and some areas on the filter membrane will be blocked during this period. The nuclear wastewater that should have flowed to this area will be diverted to other areas, causing the radionuclide concentration in the nuclear wastewater in other areas to increase significantly.

[0043] Therefore, the embodiment of the present application can analyze the radionuclide concentration change in a recent period of time at each time to obtain the possibility of abnormal exceeding, that is, the following steps are performed.

[0044] S2: Construct the endpoint definition period for each time point, and calculate the radionuclide concentration exceedance index for each time point based on the difference in radionuclide concentration between each time point and the time points in its endpoint definition period, as well as the time interval.

[0045] For example, in an embodiment of the present invention, constructing the endpoint definition period for each moment includes: presetting the length of the endpoint definition period for each moment to be... ;Use a given moment as the endpoint to define the end of the time period, and sequentially obtain the historical moments to the left of that moment. At each moment, the endpoint-defined time period is constructed.

[0046] The endpoint definition time period can be set to 60 seconds; the specific endpoint definition time period can be set according to actual needs, and this embodiment of the invention does not impose too many restrictions here.

[0047] Here's an example of how to obtain the endpoint definition period for a given time: If the current time is the 75th time, then starting from the 74th time, obtain 59 times sequentially backwards, and finally obtain the period between the 15th time and the 75th time as the endpoint definition period for the 75th time.

[0048] Understandably, since the wastewater data time series is collected in real time, when analyzing the changes in radionuclide concentration at each moment, we can only rely on the changes in its historical data. Therefore, when constructing the endpoint definition period for each moment, we can use each moment as the end of its endpoint definition period and start from the adjacent moment to the left of each moment to obtain the endpoint definition period for each moment. The endpoint definition period includes each moment itself.

[0049] In this embodiment, the endpoint definition time period is of equal length for each moment. For some moments that are at the endpoint, the number of historical moments in the current wastewater data time series may not be sufficient to construct the endpoint definition time period. For such moments, they can be supplemented in the last collection moment of the previous complete wastewater data time series. This can be achieved through existing technologies, which will not be elaborated here in this embodiment of the invention.

[0050] It should be noted that the abnormal changes in the concentration of radionuclides are gradual. During the stage of abnormal increase in the concentration of radionuclides, the concentration will gradually increase over time. The closer the time interval, the more significant the cumulative increase in concentration.

[0051] Therefore, embodiments of the present invention can measure this incremental change characteristic by obtaining the difference in radionuclide concentration between the current time and each time point within the time period defined by its endpoint. If the difference is greater than 0, it indicates that the radionuclide concentration is continuously increasing. Based on this, embodiments of the present invention further time-weight the incremental change characteristic by obtaining the time interval between the current time and each time point within the time period defined by its endpoint. The longer the time interval, the more significant the cumulative increase in concentration. Therefore, if the difference in radionuclide concentration is larger, it indicates a stronger concentration enhancement characteristic and a higher radionuclide concentration exceedance index. This characteristic does not appear under normal concentration conditions, and based on this, moments that may indicate concentration exceedances can be screened out.

[0052] For example, in an embodiment of the present invention, calculating the exceedance index of radionuclide concentration at each time point includes:

[0053] ;

[0054] Let be the index indicating the excess concentration of radionuclides at time i. Define the time period length for the endpoints. Let the i-th time point and its endpoint define the time interval of the i-th time point. The difference in radionuclide concentration between time points Let the i-th time point and its endpoint define the time interval of the i-th time point. The time interval between moments This represents the maximum difference in radionuclide concentration between all times within the time interval defined by the i-th time point and its endpoint. Let $i$ be the maximum time interval between all times in the time period defined by the $i$-th time point and its endpoint. It is the hyperbolic tangent function.

[0055] In this calculation method, Used to measure the incremental change characteristics, a positive value indicates that the concentration of the radionuclide at time i is greater than that at time i in the interval defined by its endpoint. The concentration of radionuclides at time t, from the t... The concentration of radionuclides increases from time i to time i. If the value is 0, it means that from time i... The concentration of the radionuclide did not change significantly from time i to time i, remaining at a low concentration. If the value is negative, it indicates that the concentration of the radionuclide may have gradually decreased due to membrane cleaning of the equipment.

[0056] and Used for respectively and Normalization is performed. The function is used to... The value of the concentration of the radionuclide is quantified between 0 and 1.

[0057] Reflects the corresponding period performance when the membrane device is abnormal, if the value is greater than 0, The greater the value, the longer the time interval and the greater the concentration, indicating that the membrane device at the current time may have a radionuclide penetration membrane, so the radionuclide concentration exceeds the index at the current time. Higher. On the contrary, it means that the device may be used for membrane cleaning, at which time the corresponding radionuclide concentration exceeds the index will be relatively reduced, but the radionuclide concentration exceeds the index will gradually decrease during the reduction process. The change is still measured according to the time weight.

[0058] According to the above steps, the radionuclide concentration exceeds the index at each time can be obtained. The greater the value, the greater the possibility of radionuclide penetration membrane at this time.

[0059] S3: In response to the radionuclide concentration exceeding the index at a time being greater than the index threshold, obtaining the sewage flow corresponding to each time in the sewage data time sequence, and calculating the radionuclide concentration actual exceeding index at the time.

[0060] The index threshold can be set to 0.5. The index threshold can be set according to actual needs.

[0061] For example, in the embodiment of the application, the sewage flow corresponding to each time in the sewage data time sequence is obtained, and the radionuclide concentration at each time in the sewage data time sequence corresponds to the sewage flow one by one.

[0062] Specifically, the laser displacement sensor can be placed above the membrane device processing area to monitor the radioactive sewage flow in real time.

[0063] It should be noted that the sewage containing radioactive substances generated during the operation of the related equipment producing nuclear sewage is related to the working intensity of the equipment. In the process of high-intensity use of the related equipment, the sewage amount will suddenly increase, and the impact pressure on the membrane processing area will also increase, resulting in the situation that the radionuclide is not filtered out in a short period of time. But it is obvious that such a situation is not a membrane device pollution and blockage. If the radionuclide concentration exceeds the index obtained by the above steps is directly reported as an exception, it may lead to ineffective sewage treatment.

[0064] Therefore, the embodiment of the application can obtain the sewage flow in the same time period, and correct the radionuclide concentration exceeding index obtained by the above steps by analyzing the change of the sewage flow.

[0065] For example, in response to the radionuclide concentration exceeding index of a time point being greater than the index threshold, a ratio of a standard deviation of the wastewater flow in the endpoint defining period at the time point to a minimum value of the standard deviation of the wastewater flow in the historical period of the time point can be recorded as a flow index; the flow index is taken as the index input of the radionuclide concentration exceeding index of the time point, and the actual radionuclide concentration exceeding index of the time point is obtained.

[0066] Specifically, in the case where the radionuclide concentration exceeding index of the current time point is greater than the index threshold, it is necessary to exclude the possibility that the radionuclide concentration exceeding index caused by the wastewater flow is abnormally amplified. The flow index reflects the stability of the flow fluctuation in the endpoint defining period at the current time point relative to the historical flow fluctuation.

[0067] If the flow index is not greater than 1, it indicates that the flow fluctuation in the endpoint defining period at the current time point is smaller than the historical flow fluctuation, and thus the possibility that the radionuclide concentration rise at the current time point is caused by the membrane pollution blockage is greater, and thus it is necessary to relatively increase or maintain the radionuclide concentration exceeding index of the current time point.

[0068] On the contrary, if the flow index is greater than 1, it indicates that the flow fluctuation in the endpoint defining period at the current time point is greater, and the wastewater flow appears a sudden increase, and thus the radionuclide concentration rise at the current time point may be caused by the wastewater flow, and such a change belongs to normal change, and thus it is necessary to correspondingly reduce the radionuclide concentration exceeding index of the current time point.

[0069] According to the above steps, the radionuclide concentration exceeding index that is relatively large can be adjusted, so as to accurately obtain the actual radionuclide concentration exceeding index of the current time point.

[0070] It can be understood that, in order to reduce the data processing amount, only the radionuclide concentration exceeding index that is relatively large and may exist blockage anomaly can be corrected, so as to reduce the possibility that the radionuclide concentration exceeding index is relatively large due to the flow increase; for the radionuclide concentration that is relatively small and does not exist blockage anomaly, the source of the radionuclide concentration exceeding index does not need to be further judged, and thus no additional correction is needed.

[0071] For example, in response to the radionuclide concentration exceeding index of a time point being not greater than the index threshold, the radionuclide concentration exceeding index of the time point is taken as the actual radionuclide concentration exceeding index of the time point.

[0072] Specifically, if the radionuclide concentration exceeding index of the current time is not greater than the index threshold, it is indicated that the treatment platform may have started a membrane cleaning program or the filter membrane is working normally, so that the radionuclide concentration is changed from increasing to normal, or the radionuclide concentration is continuously in a normal state. Therefore, no adjustment is needed, and the radionuclide concentration exceeding index of the current time is the actual radionuclide concentration exceeding index of the current time.

[0073] Based on the above steps, the actual radionuclide concentration exceeding index of each time can be obtained. In order to improve the accuracy of the LOF algorithm for monitoring results of nuclear contaminated water, it is necessary to adjust the k distance according to the actual radionuclide concentration exceeding index of each time, so as to improve the possibility that abnormal data can be detected, that is, the following steps are continued to be executed.

[0074] S4: Calculate the k distance of each time, and use the k distance of each time in the LOF algorithm to build a k distance range to obtain the monitoring result of each time.

[0075] It should be noted that the conventional LOF algorithm records a point with a local density significantly lower than the neighborhood as an outlier by setting a fixed k distance. For multi-dimensional data points in an abnormal segment, they are mostly characterized by outliers. If the k value is too large, the local outlier characteristics may be covered. If the k value is too small, the normal small changes in the normal segment may be misjudged as outliers.

[0076] Therefore, the actual radionuclide concentration exceeding index obtained by the above steps of the embodiment of the present application is used to adaptively adjust the default k distance value in the conventional LOF algorithm, so as to improve the accuracy and reliability of outlier identification.

[0077] For example, in the embodiment of the present application, calculating the k distance of each time further includes: performing maximum and minimum normalization processing on the radionuclide concentration of each time in the endpoint defined period of each time to obtain a normalized concentration value; and taking each time, the corresponding normalized concentration value and the actual radionuclide concentration exceeding index as a feature point.

[0078] For example, in the embodiment of the present application, calculating the k distance of each time includes:

[0079] ;

[0080] is the k distance of the i th time, is the default k distance, is the average of the actual radionuclide concentration exceeding index within the default k distance of the i th time, is an exponential function with e as the base, and e is a natural constant.

[0081] Wherein, the default k-distance can be set as 0.5; the default k-distance can be set according to actual needs.

[0082] In the calculation method, reflects the film equipment working state within the default k-distance at the i th moment, the greater the value, the higher the possibility of abnormality of the film equipment processing at the i th moment within the default k-distance, at this time, the default k-distance at the current moment needs to be reduced, in order to focus more accurately on the local situation around the current feature point, avoid including too many points with high abnormality possibility, and more accurately identify whether the current feature point is an outlier.

[0083] On the contrary, if the value is smaller, the lower the abnormality possibility of the film equipment processing at all moments within the default k-distance at the current moment, at this time, the default k-distance at the current moment needs to be increased to include more relatively normal moments, and reduce the misjudgment of normal changes.

[0084] Based on the above steps, the k-distance at each moment can be obtained.

[0085] For example, in the embodiment of the application, the k-distance at each moment is used in the LOF algorithm to construct a k-distance range, and the monitoring result at each moment is obtained, including: obtaining the Euclidean distance between the feature point and other feature points in the endpoint period of the feature point, and taking the feature points with the Euclidean distance less than the k-distance of the feature point corresponding to the moment as the feature points within the k-distance range of the feature point; calculating the local outlier factor in the k-distance range of the feature point, and obtaining the monitoring result at each moment corresponding to each feature point according to the comparison result of the local outlier factor and the outlier threshold.

[0086] For example, in the embodiment of the application, the local outlier factor in the k-distance range of the feature point is calculated, including: obtaining the reachable distance mean between the feature point and each feature point in the k-distance range of the feature point, and taking the reciprocal of the reachable distance mean as the local outlier factor in the k-distance range of the feature point.

[0087] For example, in the embodiment of the application, the monitoring result at each moment corresponding to each feature point is obtained according to the comparison result of the local outlier factor and the outlier threshold, including: setting the outlier threshold; if the local outlier factor of a feature point is greater than the outlier threshold, the feature point is an abnormal outlier, and the monitoring result at the moment corresponding to the feature point is abnormal; otherwise, the monitoring result at the moment corresponding to the feature point is normal.

[0088] Wherein, the outlier threshold can be set as 1; the outlier threshold can be set according to actual needs.

[0089] It can be understood that if the local outlier factor of a feature point is greater than the outlier threshold 1, it indicates that the local density of the current feature point is lower than the local density of other feature points in the k-distance range, the current feature point is an outlier, and the possibility of device pollution blockage corresponding to the time point of the current feature point is also higher. Therefore, for such time points, the platform can timely issue a warning.

[0090] For example, after obtaining the real-time monitoring result of the nuclear wastewater, the method further includes: in response to the monitoring result of the time point corresponding to the feature point being abnormal, issuing a warning to the outside.

[0091] The warning mode can be set according to actual needs, and the embodiments of the present application do not make too many limitations here.

[0092] As can be seen, in the embodiments of the present application, when the real-time monitoring result of the radioactive wastewater treatment platform is obtained, the radionuclide concentration and the wastewater flow rate corresponding to each time point in the wastewater data time sequence can be obtained; the endpoint defining period of each time point is constructed, the radionuclide concentration exceeding index of each time point is calculated according to the radionuclide concentration difference and the time interval between each time point and the time points in the endpoint defining period thereof; in response to the radionuclide concentration exceeding index of a time point being greater than the index threshold, the ratio of the wastewater flow rate standard deviation in the endpoint defining period of the time point to the minimum value of the wastewater flow rate standard deviation in the historical period of the time point is obtained, which is recorded as the flow index; the flow index is taken as the index input of the radionuclide concentration exceeding index of the time point, and the actual radionuclide concentration exceeding index of the time point is obtained; the k-distance of each time point is calculated, and the k-distance of each time point is negatively correlated with the average value of the actual radionuclide concentration exceeding index within the default k-distance of the time point; the k-distance range of each time point is constructed using the k-distance of each time point in the LOF algorithm, and the monitoring result of each time point is obtained, thereby effectively improving the accuracy of the real-time monitoring result of the radioactive wastewater.

[0093] The embodiments of the present application also disclose a real-time monitoring system for a radioactive wastewater treatment platform, which comprises a processor and a memory, and the memory stores computer program instructions, which realize the real-time monitoring method for the radioactive wastewater treatment platform provided by the present application when executed by the processor.

[0094] The above system also includes a communication bus and a communication interface and other components well known to those skilled in the art, the settings and functions of which are known in the art, and therefore will not be described here.

[0095] In the present application, the aforementioned memory can be any tangible medium containing or storing a program, which can be used or combined with an instruction execution system, device or apparatus.

[0096] The above are preferred embodiments of the present application, and do not limit the protection scope of the present application, so: any equivalent changes made according to the structure, shape, principle of the present application should be covered within the protection scope of the present application.

Claims

1. A method for real-time monitoring of a radioactive wastewater treatment platform, characterized in that, The method comprises the following steps: The radionuclide concentration corresponding to each time in the sewage data time sequence and the sewage flow are acquired; an endpoint defining period of each time is constructed; the radionuclide concentration exceeding index of each time is calculated according to the radionuclide concentration difference and the time interval between each time and the time in the endpoint defining period thereof, including: ; is the radionuclide concentration exceeding index of the i th time, is the endpoint defining period length, , are respectively the radionuclide concentration difference and the time interval between the i th time and the j th time in the endpoint defining period thereof, , , are respectively the radionuclide concentration difference maximum value and the time interval maximum value between the i th time and each time in the endpoint defining period thereof, is the hyperbolic tangent function; In response to the exceeding index of the radionuclide concentration at a time point being greater than an index threshold, a ratio of a standard deviation of sewage flow in an endpoint defining period of the time point to a minimum value of a standard deviation of sewage flow in a historical period of the time point is obtained, and the ratio is recorded as a flow index; the flow index is taken as an index input of the exceeding index of the radionuclide concentration at the time point, and an actual exceeding index of the radionuclide concentration at the time point is obtained; A k-distance at each time point is calculated, and the k-distance at each time point is negatively correlated with a mean value of the actual exceeding index of the radionuclide concentration within a default k-distance at the time point; A k-distance range is constructed in the LOF algorithm by using the k-distance at each time point, and a monitoring result at each time point is obtained.

2. A real-time monitoring method for a radioactive wastewater treatment platform according to claim 1, characterized in that, The endpoint defining period at each time point is constructed, and the construction comprises the following steps: The length of the end point defining period of each time is preset as ; A time point is taken as the end of the time period defined by the end point of the time point, and a historical time point to the left of the time point is sequentially obtained A time point is taken as the end of the time period defined by the end point of the time point, and a historical time point to the left of the time point is sequentially obtained 3. A method for real-time monitoring of a platform for treatment of radioactive contaminated water as claimed in claim 1, wherein, In response to the exceeding index of the radionuclide concentration at a time point being not greater than an index threshold, the exceeding index of the radionuclide concentration at the time point is taken as an actual exceeding index of the radionuclide concentration at the time point.

4. The method for real-time monitoring of a platform for the treatment of wastewater contaminated with radioactivity according to claim 1, characterized in that, The k-distance at each time point is calculated, and the calculation further comprises the following steps: Maximum and minimum normalization processing is performed on the radionuclide concentration at each time point in the endpoint defining period at each time point, and a normalized concentration value at each time point is obtained; each time point, the corresponding normalized concentration value and the actual exceeding index of the radionuclide concentration are taken as a feature point.

5. The method for real-time monitoring of a platform for the treatment of wastewater contaminated with radioactivity according to claim 1, characterized in that, The k-distance at each time point is calculated, and the calculation comprises the following steps: ； is the k-distance for the i-th time instant, is the default k-distance, is the actual overage index mean of the radionuclide concentration within the default k-distance for the i-th time instant, is the exponential function with base e.

6. A method for real-time monitoring of a platform for treatment of radioactive contaminated water according to claim 4, characterized in that, The k-distance range is constructed in the LOF algorithm by using the k-distance at each time point, and the monitoring result at each time point is obtained, and the construction comprises the following steps: In the endpoint defining period of a feature point, a Euclidean distance between the feature point and other feature points is obtained, and a feature point with a Euclidean distance less than the k-distance of the feature point corresponding to the time point is taken as a feature point within the k-distance range of the feature point; a local outlier factor in the k-distance range of the feature point is calculated, and a monitoring result at a time point corresponding to each feature point is obtained according to a comparison result of the local outlier factor and an outlier threshold.

7. A method for real-time monitoring of a platform for treatment of radioactive contaminated water according to claim 6, characterized in that, The local outlier factor in the k-distance range of the feature point is calculated, and the calculation comprises the following steps: A mean value of reachable distances between the feature point and feature points in the k-distance range of the feature point is obtained, and a reciprocal of the mean value of the reachable distances is taken as the local outlier factor in the k-distance range of the feature point.

8. The method for real-time monitoring of a platform for the treatment of wastewater contaminated with radioactivity according to claim 6, characterized in that, The monitoring result at the time point corresponding to each feature point is obtained according to the comparison result of the local outlier factor and the outlier threshold, and the obtaining comprises the following steps: The outlier threshold is set; if the local outlier factor of a feature point is greater than the outlier threshold, the feature point is an abnormal outlier, and a monitoring result at a time point corresponding to the feature point is abnormal; otherwise, the monitoring result at the time point corresponding to the feature point is normal.

9. A real time monitoring system for a radioactive wastewater treatment platform, characterized in that, The method comprises the following steps: A processor and a memory are provided, and the memory stores computer program instructions; when the computer program instructions are executed by the processor, a real-time monitoring method for a radioactive sewage treatment platform according to any one of claims 1-8 is realized.

Images