Real-time monitoring method and system for radioactive sewage treatment platform

Abstract

The invention relates to the technical field of sewage treatment, in particular to a real-time monitoring method and system for a radioactive sewage treatment platform. The method comprises the following steps: constructing an endpoint defining time period of each moment in a sewage data time sequence, and calculating a radionuclide concentration standard-exceeding index of each moment; in response to the fact that the radionuclide concentration standard-exceeding index at one moment is larger than an index threshold value, the ratio of the sewage flow standard deviation in the time period defined by the end point of the moment to the minimum value of the sewage flow standard deviation in the historical time period of the moment is obtained and recorded as a flow index, and the radionuclide concentration actual standard-exceeding index at the moment is obtained; calculating the k distance at each moment, wherein the k distance at each moment is in negative correlation with the average value of the actual standard-exceeding indexes of the radionuclide concentration within the default k distance at the moment; in the LOF algorithm, the k distance range is constructed by using the k distance at each moment to obtain the monitoring result at each moment, so that the accuracy of the real-time monitoring result of the radioactive sewage is effectively improved.

Description

Technical Field

[0001] The present invention 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 Art

[0002] In the fields of nuclear energy utilization, hospitals, and petrochemical energy, daily operations generate wastewater containing radioactive substances, known as nuclear wastewater. This wastewater contains multiple radioactive isotopes, is long-lasting and difficult to degrade. If discharged without treatment, it could cause serious harm to the ecological environment and human health.

[0003] Currently, nuclear wastewater treatment primarily relies on membrane technology to intercept and retain wastewater. This technology boasts high retention properties, achieving effective filtration by selecting a membrane with an appropriate pore size based on the characteristics of the radioactive molecules present. However, radioactive wastewater often contains other debris or contaminants, which can clog some of the membrane's pores. This forces wastewater to flow into unobstructed areas, increasing pressure in these areas. Under the impact of high flow rates, some radioactive molecules that should have been retained by the membrane may escape, resulting in poor membrane treatment effectiveness. Therefore, real-time monitoring of data collected by the radioactive wastewater treatment platform is necessary.

[0004] When membrane treatment equipment becomes clogged, radioactivity concentrations in the water area increase, deviating from normal levels, and this condition can worsen. While normal data changes are relatively stable and have high local density, the addition of abnormal data disrupts this stability, causing the local density in that area to decrease. Therefore, the LOF (Local Outlier Factor) algorithm can be used to capture these local density changes and identify outliers. However, if the K value in the LOF algorithm is set improperly, some outlier features within a local area may not be identified, or even minor normal changes may be identified as outliers.

[0005] Based on this, how to accurately adjust the K value according to the characteristics of nuclear wastewater data, so as to effectively improve the accuracy of real-time monitoring results of the radioactive wastewater treatment platform, is an urgent problem to be solved. Summary of the Invention

[0006] In order to solve the technical problem of how to accurately adjust the K value according to the characteristics of nuclear wastewater data, thereby effectively improving the accuracy of real-time monitoring results of the radioactive wastewater treatment platform, the present invention provides a real-time monitoring method and system for the radioactive wastewater treatment platform.

[0007] In a first aspect, the present invention provides a real-time monitoring method for a radioactive wastewater treatment platform, which adopts the following technical solution: A real-time monitoring method for a radioactive wastewater treatment platform comprises the following steps: Obtain the radioactive nuclide concentration and sewage flow corresponding to each moment in the sewage data time series; construct an endpoint definition period for each moment, and calculate the radioactive nuclide concentration exceedance index at each moment based on the radioactive nuclide concentration difference and time interval between each moment and the moment in its endpoint definition period; in response to the radioactive nuclide concentration exceedance index at a moment being greater than the index threshold, obtain the ratio of the sewage flow standard deviation in the endpoint definition period at that moment to the minimum sewage flow standard deviation in the historical period at that moment, and record it as the flow index; use the flow index as the index input of the radioactive nuclide concentration exceedance index at that moment to obtain the actual radioactive nuclide concentration exceedance index at that moment; calculate the k distance at each moment, and the k distance at each moment is negatively correlated with the average of the actual radioactive nuclide concentration exceedance index within the default k distance at that moment; use the k distance at each moment to construct the k distance range in the LOF algorithm to obtain the monitoring result at each moment.

[0008] The present invention uses the LOF algorithm to identify abnormal density in the sewage data time series, and can accurately screen out abnormal moments of radioactive nuclide concentration. In the process of obtaining the k distance at each moment, the present invention analyzes the changing trend of the radioactive nuclide concentration within the endpoint definition period through the pollution change characteristics of the filter membrane, constructs the abnormal exceeding index of the nuclide concentration at the current moment, provides a basis for the subsequent identification of true outliers, and reduces the monitoring result error caused by improper k distance setting. On this basis, the present invention introduces the sewage flow change within the endpoint definition period at the current moment as a flow index to correct the exceeding index, which can effectively distinguish between the short-term penetration effect caused by the sudden increase in flow and the real membrane pollution phenomenon, reduce the false alarm rate, and thus effectively improve the accuracy of the real-time monitoring results of the radioactive sewage treatment platform.

[0009] According to a real-time monitoring method for a radioactive wastewater treatment platform provided by the present invention, the construction of the endpoint definition period at each moment includes: presetting the length of the endpoint definition period at each moment to be ; Take a moment as the endpoint of the moment to define the end of the period, and obtain the historical moments to the left of the moment in sequence A moment is constructed to obtain the endpoint defining period of the moment.

[0010] According to a real-time monitoring method for a radioactive wastewater treatment platform provided by the present invention, the calculation of the radionuclide concentration excess index at each moment includes: ; is the radionuclide concentration exceeding standard index at the i-th moment, Defines the duration of the time period for the endpoint, 、 The i-th moment and its endpoint define the The difference in radionuclide concentration between moments, the time interval, 、 are the maximum difference in radionuclide concentration and the maximum time interval between the i-th moment and each moment in the period defined by its endpoint, respectively. is the hyperbolic tangent function.

[0011] The present invention provides an accurate method for calculating the index of radionuclide concentration exceeding the standard. By analyzing whether the radionuclide concentration within the time period defined by each endpoint at each moment has a significant increasing change over time, the possibility of abnormal exceeding the standard can be accurately determined.

[0012] According to a real-time monitoring method for a radioactive wastewater treatment platform provided by the present invention, in response to a radioactive nuclide concentration exceedance index at a moment being no greater than an index threshold, the radioactive nuclide concentration exceedance index at that moment is used as the actual radioactive nuclide concentration exceedance index at that moment.

[0013] The present invention takes into account that when the radionuclide concentration exceedance index at a certain moment is greater than the index threshold, it is necessary to analyze the source of the abnormal exceedance to obtain the actual radionuclide concentration exceedance index. However, for moments when the concentration is not greater than the index threshold, the actual radionuclide concentration exceedance index can be directly obtained, thereby effectively reducing the data processing amount and improving the algorithm efficiency.

[0014] According to a real-time monitoring method for a radioactive wastewater treatment platform provided by the present invention, the calculation of the k-distance at each moment also includes: performing maximum and minimum normalization processing on the radioactive nuclide concentration at each moment in the time period defined by the endpoint of each moment to obtain the normalized concentration value at each moment; and taking the normalized concentration value corresponding to each moment and the actual excess index of the radioactive nuclide concentration as a characteristic point.

[0015] According to a real-time monitoring method for a radioactive wastewater treatment platform provided by the present invention, the calculation of the k distance at each moment includes: ; is the k-distance at the i-th moment, is the default k distance, is the actual excess index of radionuclide concentration within the default k distance at the i-th moment, is an exponential function with base e.

[0016] The present invention provides an accurate k-distance calculation method for each moment. By analyzing the actual exceedance index of radionuclide concentration within the default k-distance in the LOF algorithm, the default k-distance is adjusted so that the obtained k-distance at each moment is more consistent with the actual data distribution, thereby improving the accuracy of anomaly monitoring.

[0017] According to a real-time monitoring method for a radioactive wastewater treatment platform provided by the present invention, the k distance at each moment is used in the LOF algorithm to construct a k distance range to obtain the monitoring result at each moment, including: obtaining the Euclidean distance between a feature point and other feature points in the endpoint definition period of the feature point, and taking the feature point whose Euclidean distance is less than the k distance at the corresponding moment of the feature point as the feature point 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 corresponding to each feature point at the corresponding moment based on the comparison result of the local outlier factor and the outlier threshold.

[0018] The present invention takes into account that the change in sewage density at the latest moment is only related to the distribution of characteristic points in the time period defined by its endpoints. Therefore, when obtaining characteristic points within the k distance, they are only obtained in the time period defined by its endpoints, so that the distribution of each moment in the time period defined by its endpoints can be accurately measured to accurately extract abnormal discrete points.

[0019] According to a real-time monitoring method for a radioactive wastewater treatment platform provided by the present invention, the calculation of the local outlier factor in the k-distance range of the feature point includes: obtaining the average of the reachable distances between the feature point and each feature point in its k-distance range, and taking the inverse of the average of the reachable distances as the local outlier factor in the k-distance range of the feature point.

[0020] According to a real-time monitoring method for a radioactive wastewater treatment platform provided by the present invention, the monitoring result at the corresponding moment of each feature point is obtained based on 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 point, 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.

[0021] In a second aspect, the present invention provides a real-time monitoring system for a radioactive wastewater treatment platform, which adopts the following technical solutions: A real-time monitoring system for a radioactive wastewater treatment platform comprises a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the real-time monitoring method for a radioactive wastewater treatment platform is implemented.

[0022] By adopting the above technical solution, a computer program is generated using a real-time monitoring method for the above-mentioned radioactive wastewater treatment platform and stored in a memory to be loaded and executed by a processor, thereby making a terminal device based on the memory and the processor for easy use.

[0023] The present invention has the following technical effects: Based on the above technical solution, the present invention provides a real-time monitoring method and system for a radioactive wastewater treatment platform, which uses the LOF algorithm to identify the abnormal density in the sewage data time series, and can accurately screen out the abnormal moments of radioactive nuclide concentration. In the process of obtaining the k distance at each moment, the present invention analyzes the changing trend of the radioactive nuclide concentration within the endpoint definition period through the pollution change characteristics of the filter membrane, constructs the abnormal exceeding index of the nuclide concentration at the current moment, provides a basis for the subsequent identification of real outliers, and reduces the monitoring result error caused by improper k distance setting. On this basis, the present invention introduces the sewage flow change within the endpoint definition period at the current moment as a flow index to correct the exceeding index, which can effectively distinguish between the short-term penetration effect caused by the sudden increase in flow and the real membrane pollution phenomenon, reduce the false alarm rate, and thus effectively improve the accuracy of the real-time monitoring results of the radioactive wastewater treatment platform. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 A schematic flow chart of a real-time monitoring method for a radioactive wastewater treatment platform provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0025] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, but not all of the embodiments.

[0026] The embodiment of the present invention discloses a real-time monitoring method for a radioactive wastewater treatment platform. Figure 1 As shown, Figure 1 This is a flow chart of a real-time monitoring method for a radioactive wastewater treatment platform provided by an embodiment of the present invention. The method can accurately screen out outliers within normal and abnormal segments by adaptively calculating the k-distance at each moment in the LOF algorithm, effectively improving the accuracy of real-time monitoring of the radioactive wastewater treatment platform. The method specifically includes the following steps: S1: Obtain the radionuclide concentration corresponding to each moment in the sewage data time series.

[0027] 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, cobalt-60 and other molecules. The structure of this type of radioactive molecules is significantly larger than that of water molecules. Selecting suitable membrane equipment based on the structural characteristics of radioactive molecules that may exist in different scenarios can ensure the separation of radioactive molecules and water molecules in radioactive wastewater.

[0028] For example, in an embodiment of the present invention, the radioactive nuclide concentration and sewage flow corresponding to each moment in the sewage data time series sequence are obtained, including: collecting the radioactive nuclide concentration in the sewage at each moment, and arranging it from left to right according to the collection order after preprocessing to obtain the sewage data time series sequence.

[0029] Among them, the preprocessing method can be missing data interpolation, etc., which can be set according to actual needs.

[0030] Specifically, a radioactive in-situ monitoring sensor is placed in the water area after being treated by the membrane device, and the concentration of radioactive nuclides after membrane treatment is collected according to a fixed collection frequency and collection time.

[0031] The collection frequency can be set to collect data once per second, and the collection duration can be set to 1 day. The collection frequency and collection duration can be set according to actual needs, and the embodiment of the present invention does not impose too many restrictions on this.

[0032] It is understandable that if the collection time is 1 day, then the time interval between the start and end of the sewage data time series obtained after each collection is completed is 1 day. The historical data at the current moment is all the historical moments in the sewage data time series at that moment.

[0033] It should be noted that when the membrane equipment treats nuclear wastewater normally, radioactive molecules can be effectively separated, so the concentration of radioactive nuclides in the water area after membrane treatment will be stable. 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 concentration of radioactive nuclides in the nuclear wastewater in other areas to show a significant incremental change.

[0034] Therefore, the embodiment of the present invention can analyze the change in radionuclide concentration at each moment within a recent period of time to determine the possibility of abnormal exceeding of the standard, that is, perform the following steps.

[0035] S2: Construct an endpoint definition period for each moment, and calculate the radionuclide concentration exceedance index at each moment based on the difference in radionuclide concentration between each moment and the moment in its endpoint definition period and the time interval.

[0036] For example, in an embodiment of the present invention, constructing the endpoint definition period at each moment includes: presetting the endpoint definition period length at each moment to be ; Take a moment as the endpoint of the moment to define the end of the period, and obtain the historical moments to the left of the moment in sequence A moment is constructed to obtain the endpoint defining period of the moment.

[0037] The length of the endpoint definition period can be set to 60 seconds. The length of the endpoint definition period can be set according to actual needs, and the embodiment of the present invention does not impose too many restrictions on this.

[0038] Take an example to illustrate how to obtain the endpoint definition period of a moment: if the current moment is the 75th moment, then starting from the 74th moment, obtain 59 moments in sequence, and finally obtain the period between the 15th moment and the 75th moment as the endpoint definition period of the 75th moment.

[0039] It is understandable that since the sewage data time series is collected in real time, when analyzing the changes in radionuclide concentrations at each moment, we can only rely on the changes in its historical data. Therefore, when constructing the endpoint definition period of each moment, we can use each moment as the end of its endpoint definition period for construction, and start from the adjacent moment on the left side of each moment to obtain them one by one, and finally obtain the endpoint definition period of each moment, which includes each moment itself.

[0040] The endpoint definition period of each moment is of equal length. For moments that are partially at the endpoint, the number of historical moments in the current sewage data time series may not be sufficient to construct the endpoint definition period. For such moments, they can be supplemented at the terminal collection moment in the previous complete sewage data time series. This can be achieved using existing technologies and will not be described in detail in the embodiments of the present invention.

[0041] It should be noted that the abnormal change in radioactive nuclide concentration is gradual. During the stage of abnormal increase in radioactive nuclide concentration, the concentration will gradually increase with time. The closer the time interval, the more significant the cumulative increase in concentration.

[0042] Therefore, embodiments of the present invention can measure this incremental change characteristic by measuring the difference in radionuclide concentration between the current moment and each moment in the period defined by its endpoints. If the difference is greater than 0, it indicates that the radionuclide concentration is continuously increasing. Furthermore, embodiments of the present invention further time-weight the incremental change characteristic by measuring the time interval between the current moment and each moment in the period defined by its endpoints. Longer time intervals indicate a more significant cumulative increase in concentration. Therefore, a greater difference in radionuclide concentration indicates a stronger concentration enhancement characteristic and a higher radionuclide concentration exceedance index. This characteristic does not occur at normal concentrations, allowing us to identify moments where concentrations may exceed the standard.

[0043] For example, in an embodiment of the present invention, calculating the radionuclide concentration excess index at each moment includes: ; is the radionuclide concentration exceeding standard index at the i-th moment, Defines the duration of the time period for the endpoint, The first moment in the period defined by its endpoints The difference in radionuclide concentration between the The first moment in the period defined by its endpoints The time interval between moments, is the maximum difference in radionuclide concentration between the i-th moment and each moment in the period defined by its endpoint, The maximum time interval between the i-th moment and each moment in the period defined by its endpoint, is the hyperbolic tangent function.

[0044] In this calculation method, Used to measure the incremental change characteristics. If the value is positive, it means that the radionuclide concentration at the i-th moment is greater than the i-th moment in the endpoint definition period. The radionuclide concentration at the moment The concentration of radioactive nuclides increases from the moment to the moment i. If the value is 0, it means that The radionuclide concentration does not change significantly from moment i to moment i, and remains at a low concentration. If the value is negative, it means that the membrane cleaning of the equipment may have caused the radionuclide concentration to gradually decrease.

[0045] and Used for and Perform normalization processing. Function is used to The value of is quantized between 0 and 1.

[0046] It reflects the performance of the corresponding period when the membrane equipment is abnormal. If the value is greater than 0, A larger value indicates a longer time interval and a higher concentration, indicating that radioactive molecules may be penetrating the membrane at the current moment, and therefore the radionuclide concentration exceedance index at the current moment will be higher. Conversely, it indicates that the membrane may have been cleaned, in which case the corresponding radionuclide concentration exceedance index will be relatively lower. However, during the reduction process, such changes will still be weighed according to the time weight, and the radionuclide concentration exceedance index will gradually decrease.

[0047] According to the above steps, the radionuclide concentration exceedance index at each moment can be obtained. The larger the value, the greater the possibility that radioactive molecules penetrate the membrane at this time.

[0048] S3: In response to the radionuclide concentration exceeding standard index at a moment being greater than the index threshold, the sewage flow corresponding to each moment in the sewage data time series is obtained, and the actual radionuclide concentration exceeding standard index at the moment is calculated.

[0049] The index threshold may be set to 0.5; the index threshold may be specifically set according to actual needs.

[0050] For example, in an embodiment of the present invention, the sewage flow rate corresponding to each moment in the sewage data time series is obtained, and the radioactive nuclide concentration at each moment in the sewage data time series corresponds to the sewage flow rate one by one.

[0051] Specifically, a laser displacement sensor can be placed above the membrane equipment treatment area to monitor the flow of radioactive wastewater in real time.

[0052] It should be noted that the amount of radioactive wastewater generated during the operation of equipment that produces nuclear wastewater is related to the intensity of the equipment's operation. During high-intensity use of the equipment, wastewater volumes can surge, increasing the impact pressure on the membrane treatment area and potentially preventing radioactive molecules from being filtered out for a short period of time. However, this situation clearly does not indicate contamination or blockage of the membrane equipment. Directly reporting abnormalities based on the radionuclide concentration exceedance index obtained in the above steps could result in reduced wastewater treatment effectiveness.

[0053] Therefore, the embodiment of the present invention can obtain the sewage flow rate within the same time period, and correct the radionuclide concentration excess index obtained in the above steps by analyzing the change of the sewage flow rate.

[0054] For example, in response to the radionuclide concentration exceedance index at a moment being greater than the index threshold, the ratio of the sewage flow standard deviation in the endpoint defined period at that moment to the minimum sewage flow standard deviation in the historical period at that moment can be recorded as the flow index; the flow index is used as the index input of the radionuclide concentration exceedance index at that moment to obtain the actual radionuclide concentration exceedance index at that moment.

[0055] Specifically, if the current radionuclide concentration exceedance index exceeds the index threshold, it is necessary to rule out the possibility that the radionuclide concentration exceedance index is abnormally amplified due to sewage flow. The flow index reflects the stability of flow fluctuations within the defined time period at the current endpoint relative to historical flow fluctuations.

[0056] If the flow index is not greater than 1, it means that the flow fluctuation within the endpoint definition period at the current moment is smaller than the historical flow fluctuation. Therefore, the increase in radionuclide concentration at the current moment is more likely to be caused by membrane contamination and blockage. Therefore, it is necessary to relatively increase or maintain the radionuclide concentration exceedance index at the current moment.

[0057] On the contrary, if the flow index is greater than 1, it means that the flow fluctuation within the endpoint definition period at the current moment is large, and the sewage flow has increased suddenly. Therefore, the increase in radionuclide concentration at the current moment may be partly caused by sewage flow. Such changes are normal changes, so the radionuclide concentration exceedance index at the current moment needs to be reduced accordingly.

[0058] According to the above steps, the larger radionuclide concentration exceeding index can be adjusted, so as to accurately obtain the actual radionuclide concentration exceeding index at the current moment.

[0059] It is understandable that in order to reduce the amount of data processing, only the larger radionuclide concentration excess index that may have a blockage anomaly can be corrected, thereby reducing the possibility of a larger radionuclide concentration excess index caused by an increase in flow rate; for smaller radionuclide concentrations, there is no blockage anomaly, and there is no need to further determine the source of the radionuclide concentration excess index, so no additional correction is required.

[0060] For example, in response to the radionuclide concentration exceedance index at a moment being not greater than the index threshold, the radionuclide concentration exceedance index at the moment is used as the actual radionuclide concentration exceedance index at the moment.

[0061] Specifically, if the current radionuclide concentration exceedance index is not greater than the index threshold, it indicates that the processing platform may have initiated a membrane cleaning process or the filter membrane is functioning normally, causing the radionuclide concentration to return from increasing to normal, or the radionuclide concentration to remain normal. Therefore, no adjustment is required, and the current radionuclide concentration exceedance index is the actual radionuclide concentration exceedance index at that moment.

[0062] Based on the above steps, the actual excursion index of radionuclide concentration at each moment can be obtained. In order to improve the accuracy of the LOF algorithm's nuclear wastewater monitoring results, it is necessary to adjust its k distance according to the actual excursion index of radionuclide concentration at each moment, thereby increasing the possibility of detecting abnormal data, that is, continue to perform the following steps.

[0063] S4: Calculate the k distance at each moment, use the k distance at each moment to construct the k distance range in the LOF algorithm, and obtain the monitoring result at each moment.

[0064] It should be noted that the conventional LOF algorithm records points with local density significantly lower than that of the neighborhood as outliers by setting a fixed k distance. However, for multidimensional data points in abnormal segments, they often show outlier characteristics. If the k value is too large, it may mask the local outlier characteristics; if the k value is too small, normal small changes in the normal segment may be misjudged as outliers.

[0065] Therefore, the embodiment of the present invention uses the actual radionuclide concentration exceedance index obtained through the above steps to adaptively adjust the default k distance value in the conventional LOF algorithm, thereby improving the accuracy and reliability of outlier identification.

[0066] For example, in an embodiment of the present invention, the k distance at each moment is calculated, and the process also includes: performing maximum and minimum normalization processing on the radioactive nuclide concentration at each moment in the time period defined by the endpoints of each moment to obtain a normalized concentration value; and taking the normalized concentration value corresponding to each moment and the actual exceedance index of the radioactive nuclide concentration as a characteristic point.

[0067] For example, in an embodiment of the present invention, calculating the k distance at each moment includes: ; is the k-distance at the i-th moment, is the default k distance, is the actual excess index of radionuclide concentration within the default k distance at the i-th moment, It is an exponential function with base e, where e is a natural constant.

[0068] The default k distance may be set to 0.5; the default k distance may be set according to actual needs.

[0069] In this calculation method, It reflects the working status of the membrane equipment within the default k-distance at the i-th moment. The larger the value, the higher the possibility of abnormality in the membrane equipment processing within the default k-distance at the i-th moment. At this time, it is necessary to reduce the default k-distance at the current moment to focus more accurately on the local situation around the current feature point, avoid including too many points with high abnormality possibilities, and thus more accurately identify whether the current feature point is an outlier.

[0070] On the contrary, if the value is smaller, it means that the possibility of abnormality in the membrane equipment processing at all times within the current default k distance is lower. In this case, it is necessary to increase the current default k distance to include more relatively normal moments and reduce the misjudgment of normal changes.

[0071] Based on the above steps, the k distance at each moment can be obtained.

[0072] For example, in an embodiment of the present invention, the k distance at each moment is used in the LOF algorithm to construct a k distance range to obtain the monitoring result at each moment, including: obtaining the Euclidean distance between the feature point and other feature points in the endpoint definition period of a feature point, and taking the feature point whose Euclidean distance is less than the k distance at the corresponding moment of the feature point as the feature point 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 the corresponding moment of each feature point based on the comparison result of the local outlier factor and the outlier threshold.

[0073] For example, in an embodiment of the present invention, calculating the local outlier factor in the k-distance range of a feature point includes: obtaining the mean of the reachable distances between the feature point and each feature point in its k-distance range, and taking the inverse of the mean of the reachable distances as the local outlier factor in the k-distance range of the feature point.

[0074] For example, in an embodiment of the present invention, the monitoring result of each feature point at the corresponding moment is obtained based on 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, then the feature point is an abnormal outlier point, 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.

[0075] The outlier threshold may be set to 1; the outlier threshold may be set according to actual needs.

[0076] It can be understood that if the local outlier factor of a feature point is greater than the outlier threshold 1, it means that the local density of the current feature point is lower than the local density of other feature points in its k-distance range. The current feature point is an outlier, and the possibility of equipment contamination and blockage at the moment corresponding to the current feature point is also higher. Therefore, for such moments, timely warnings can be issued through the platform.

[0077] For example, after obtaining the real-time monitoring result of nuclear wastewater, the method further includes: in response to the monitoring result at the moment corresponding to the characteristic point being abnormal, issuing an early warning.

[0078] The warning mode can be specifically set according to actual needs, and the embodiment of the present invention does not impose too many restrictions here.

[0079] It can be seen that in an embodiment of the present invention, when obtaining the real-time monitoring results of the radioactive wastewater treatment platform, the radioactive nuclide concentration and sewage flow corresponding to each moment in the sewage data time series can be obtained; an endpoint definition period is constructed for each moment, and the radioactive nuclide concentration exceedance index at each moment is calculated based on the radioactive nuclide concentration difference and time interval between each moment and the moment in its endpoint definition period; in response to the radioactive nuclide concentration exceedance index at a moment being greater than the index threshold, the ratio of the sewage flow standard deviation in the endpoint definition period at that moment to the minimum sewage flow standard deviation in the historical period at that moment is obtained, and recorded as the flow index; the flow index is used as the index input of the radioactive nuclide concentration exceedance index at that moment to obtain the actual radioactive nuclide concentration exceedance index at that moment; the k distance at each moment is calculated, and the k distance at each moment is negatively correlated with the average of the actual radioactive nuclide concentration exceedance index within the default k distance at that moment; the k distance at each moment is used in the LOF algorithm to construct the k distance range to obtain the monitoring result at each moment, thereby effectively improving the accuracy of the real-time monitoring results of radioactive wastewater.

[0080] An embodiment of the present invention also discloses a real-time monitoring system for a radioactive wastewater treatment platform, comprising a processor and a memory, wherein the memory stores computer program instructions. When the computer program instructions are executed by the processor, a real-time monitoring method for a radioactive wastewater treatment platform provided by the present invention is implemented.

[0081] The above system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface. The configuration and functions of these components are known in the art and will not be described in detail here.

[0082] In the present invention, the aforementioned memory may be any tangible medium that contains or stores a program, which may be used by or in combination with an instruction execution system, apparatus, or device.

[0083] The above are all preferred embodiments of the present invention, and are not intended to limit the scope of protection of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. A real-time monitoring method for a radioactive wastewater treatment platform, characterized in that: include: Obtain the radionuclide concentration and sewage flow rate corresponding to each moment in the sewage data time series; construct an endpoint definition period for each moment, and calculate the radionuclide concentration exceedance index at each moment based on the radionuclide concentration difference and time interval between each moment and the moment in its endpoint definition period; In response to the radionuclide concentration exceedance index at a moment being greater than an index threshold, obtaining a ratio of a sewage flow rate standard deviation in an endpoint defined period at that moment to a minimum sewage flow rate standard deviation in a historical period at that moment, and recording the ratio as a flow rate index; using the flow rate index as an index input of the radionuclide concentration exceedance index at that moment, and obtaining an actual radionuclide concentration exceedance index at that moment; Calculate the k distance at each moment. The k distance at each moment is negatively correlated with the mean of the actual exceedance index of the radionuclide concentration within the default k distance at that moment. In the LOF algorithm, the k distance at each moment is used to construct the k distance range to obtain the monitoring result at each moment.

2. The real-time monitoring method for a radioactive wastewater treatment platform according to claim 1, characterized in that: The construction of the endpoint definition period at each moment includes: The endpoint of each moment is preset to define the time period length as ; Use a moment as the endpoint to define the end of the period, and obtain the historical moments to the left of the moment in sequence A moment is constructed to obtain the endpoint defining period of the moment.

3. The real-time monitoring method for a radioactive wastewater treatment platform according to claim 1, characterized in that: The calculation of the radionuclide concentration excess index at each moment includes: ； is the radionuclide concentration exceeding standard index at the i-th moment, Defines the duration of the time period for the endpoint, 、 The i-th moment and its endpoint define the The difference in radionuclide concentration between moments, the time interval, 、 are the maximum difference in radionuclide concentration and the maximum time interval between the i-th moment and each moment in the period defined by its endpoint, respectively. is the hyperbolic tangent function.

4. The real-time monitoring method for a radioactive wastewater treatment platform according to claim 1, characterized in that: In response to the radionuclide concentration exceedance index at a moment being not greater than the index threshold, the radionuclide concentration exceedance index at the moment is used as the actual radionuclide concentration exceedance index at the moment.

5. The real-time monitoring method for a radioactive wastewater treatment platform according to claim 1, characterized in that: The calculation of the k distance at each moment also includes: The radioactive nuclide concentrations at each moment in the endpoint definition period are normalized to the maximum and minimum values ​​to obtain the normalized concentration value at each moment; each moment and its corresponding normalized concentration value and the actual excess index of the radioactive nuclide concentration are taken as a characteristic point.

6. The real-time monitoring method for a radioactive wastewater treatment platform according to claim 1, characterized in that: The calculation of the k distance at each moment includes: ； is the k-distance at the i-th moment, is the default k distance, is the average value of the actual exceeding standard index of radionuclide concentration within the default k distance at the i-th moment, is an exponential function with base e.

7. The real-time monitoring method for a radioactive wastewater treatment platform according to claim 5, characterized in that: The LOF algorithm uses the k distance at each moment to construct a k distance range to obtain the monitoring result at each moment, including: The Euclidean distance between a feature point and other feature points is obtained during the endpoint definition period of a feature point, and the feature point whose Euclidean distance is less than the k distance at the corresponding moment of the feature point is regarded as the feature point within the k distance range of the feature point; the local outlier factor in the k distance range of the feature point is calculated, and the monitoring result at the corresponding moment of each feature point is obtained based on the comparison result of the local outlier factor and the outlier threshold.

8. The real-time monitoring method for a radioactive wastewater treatment platform according to claim 7, characterized in that: The calculating of the local outlier factor within the k-distance range of the feature point includes: The average of the reachable distances between the feature point and each feature point in its k-distance range is obtained, and the inverse of the average of the reachable distances is used as the local outlier factor in the k-distance range of the feature point.

9. The real-time monitoring method for a radioactive wastewater treatment platform according to claim 7, characterized in that: The monitoring result of each feature point at the corresponding moment is obtained based on the comparison result of the local outlier factor and the outlier threshold, including: Set 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 time corresponding to the feature point is abnormal; otherwise, the monitoring result at the time corresponding to the feature point is normal.

10. A real-time monitoring system for a radioactive wastewater treatment platform, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a real-time monitoring method for a radioactive wastewater treatment platform according to any one of claims 1 to 9 is implemented.

Images