Intelligent sewage treatment monitoring method and system

Abstract

The present application relates to the field of sewage treatment monitoring, in particular to an intelligent sewage treatment monitoring method and system. The method first uses the influent flow and influent temperature of the sewage treatment plant at each time of each day in the historical period to construct a working condition characteristic vector, clusters all days contained in the historical period based on the working condition characteristic vector, and statistically analyzes each water quality parameter of all times of all days in the working condition cluster, obtains the threshold interval of each water quality parameter in the working condition cluster, uses the real-time working condition characteristic vector constructed by the influent flow and influent temperature of each time of the current short period to match the working condition cluster, obtains the matching threshold interval of each water quality parameter of the current short period, compares each water quality parameter of each time of the current short period with the matching threshold interval, and realizes abnormal early warning. The present application can avoid that a single threshold value cannot adapt to the complex working conditions of sewage treatment, and reduce false alarms caused by normal changes in working conditions.

Description

Technical Field

[0001] This invention relates to the field of wastewater treatment monitoring, and specifically to an intelligent wastewater treatment monitoring method and system. Background Technology

[0002] Monitoring the wastewater treatment process is an indispensable part of modern environmental management and water treatment technology. Its core significance lies in transforming this complex biochemical treatment process from an uncontrollable "black box" into a transparent, scientific, and precise operation. It is the lifeline for ensuring water environment safety and the stable operation of treatment plants. Monitoring is the guarantee that wastewater treatment meets standards. By tracking key water quality indicators such as COD, ammonia nitrogen, and pH value in real time, it is possible to ensure that the effluent quality meets environmental protection standards, prevent pollution accidents, and protect the ecological balance of receiving water bodies.

[0003] Related technologies typically monitor core water quality parameters such as COD and dissolved oxygen and compare them with fixed thresholds to monitor and warn of potential abnormalities in the wastewater treatment process. However, the influent conditions, determined by the influent flow rate and temperature, are complex. When the influent flow rate changes drastically due to heavy rain, peak day and night water usage, or seasonal or weather-related fluctuations, the microbial activity and reaction rate of the entire treatment system change, leading to alterations in the normal fluctuation range of water quality parameters (such as COD and ammonia nitrogen). Furthermore, wastewater treatment plants do not operate in a single mode; they may operate under multiple modes, such as high influent load, low influent load, and seasonal operation. Under different modes, the normal levels of water quality parameters vary significantly. Consequently, existing wastewater treatment monitoring mechanisms using fixed thresholds cannot accurately capture abnormalities and cannot precisely identify and warn of abnormalities in the wastewater treatment process. Summary of the Invention

[0004] To address the technical problem of how to extend static thresholds to dynamic ranges adaptable to different operating conditions, thus avoiding the inability of a single threshold to adapt to complex operating conditions and resulting in incorrect identification of water quality anomalies, the present invention aims to provide an intelligent wastewater treatment monitoring method and system. The specific technical solution adopted is as follows:

[0005] A first aspect of this invention provides an intelligent wastewater treatment monitoring method, the method comprising:

[0006] Obtain the influent flow rate, influent temperature, and different types of water quality parameters of the wastewater treatment plant at each moment in a historical period;

[0007] Based on the distribution of the influent flow rate and the distribution of the influent temperature at each time of day, a daily operating condition feature vector is obtained; based on the operating condition feature vector, all days included in the historical period are clustered to obtain multiple operating condition clusters; statistical analysis is performed on each water quality parameter at all times of all days in each operating condition cluster to obtain the threshold range of each water quality parameter in each operating condition cluster.

[0008] The system acquires in real-time data on influent flow rate, influent temperature, and different types of water quality parameters at various times within the current short period. Based on the distribution of influent flow rate and influent temperature at each time point within the current short period, it obtains a real-time operating condition feature vector for the current short period. Based on the real-time operating condition feature vector for the current short period and the distance between each operating condition cluster, it obtains a matching threshold range for each water quality parameter within the current short period.

[0009] The system compares each water quality parameter at each moment of the current short period with the corresponding matching threshold range, and issues an anomaly warning for each water quality parameter.

[0010] Furthermore, obtaining the daily working condition feature vector includes:

[0011] The dispersion of the influent flow rate at all times of the day is analyzed to obtain the daily influent flow rate disorder.

[0012] The average of the inflow rate at all times of the day is taken as the total inflow rate for the day.

[0013] The influent flow disorder is used as the numerator, the overall influent flow is used as the denominator, and the ratio is used as the daily influent flow variation coefficient.

[0014] The average of the inlet water temperature at all times of the day is taken as the overall inlet water temperature for the day.

[0015] The two-dimensional vector formed by the daily influent flow rate variation coefficient and the overall influent water temperature is used as the daily operating condition feature vector.

[0016] Furthermore, obtaining multiple working condition clusters includes:

[0017] The Euclidean distance between the feature vectors of the operating conditions of any two days is used as the distance metric between the two days.

[0018] Using the K-means clustering algorithm and based on the distance metric between any two days, clusters are formed for all days within a historical time period to obtain multiple working condition clusters.

[0019] Furthermore, obtaining the threshold range for each water quality parameter in each operating condition cluster includes:

[0020] Using any water quality parameter as the target water quality parameter, box plot analysis is performed on the target water quality parameter for all days and all times in each operating condition cluster to obtain the threshold range of the target water quality parameter in each operating condition cluster.

[0021] Furthermore, obtaining the real-time operating condition feature vector for the current short period includes:

[0022] Using the same calculation method as the daily operating condition feature vector, the real-time operating condition feature vector for the current short period is obtained based on the distribution of the influent flow rate and the distribution of the influent water temperature at each moment of the current short period.

[0023] Furthermore, obtaining the matching threshold range for each water quality parameter in the current short time period includes:

[0024] The average value of the feature vectors of all days in each working condition cluster is used as the cluster center of each working condition cluster.

[0025] Based on the real-time operating condition feature vector of the current short period and the distance between the cluster centers of each operating condition cluster, the matching operating condition cluster of the current short period is selected from all operating condition clusters.

[0026] The threshold range of each water quality parameter in the matching condition cluster is used as the matching threshold range for each water quality parameter in the current short period.

[0027] Furthermore, the step of selecting the matching working condition cluster for the current short period from all working condition clusters includes:

[0028] The Euclidean distance between the real-time operating condition feature vector of the current short period and the cluster center of each operating condition cluster is used as the operating condition difference degree between each operating condition cluster and the current short period.

[0029] The cluster of operating conditions corresponding to the minimum value of the operating condition difference between the current short period and the current short period is taken as the matching operating condition cluster of the current short period.

[0030] Furthermore, the abnormal early warning for each water quality parameter includes:

[0031] For any water quality parameter, if, within the current short period of time, the water quality parameter does not fall within the matching threshold range for at least a preset number of consecutive moments, the system will issue an abnormal warning for the water quality parameter.

[0032] A second aspect of the present invention provides an intelligent wastewater treatment monitoring system, the system comprising:

[0033] The historical data acquisition module is used to obtain the influent flow rate, influent temperature, and different types of water quality parameters of the wastewater treatment plant at each moment in a historical period.

[0034] The operating condition segmentation module is used to obtain a daily operating condition feature vector based on the distribution of the influent flow rate and the distribution of the influent water temperature at each time of day; based on the operating condition feature vector, clustering is performed on all days included in the historical period to obtain multiple operating condition clusters; statistical analysis is performed on each water quality parameter at all times of all days in each operating condition cluster to obtain the threshold range of each water quality parameter in each operating condition cluster.

[0035] The operating condition matching module is used to acquire in real time the influent flow rate, influent temperature, and different types of water quality parameters of the wastewater treatment plant at different times in the current short period. Based on the distribution of the influent flow rate and the distribution of the influent temperature at each time in the current short period, the module obtains the real-time operating condition feature vector for the current short period. Based on the real-time operating condition feature vector for the current short period and the distance between each operating condition cluster, the module obtains the matching threshold range for each water quality parameter in the current short period.

[0036] The water quality monitoring module is used to compare each water quality parameter at different times in the current short period with the corresponding matching threshold range, and to issue an abnormal warning for each water quality parameter.

[0037] A third aspect of the present invention provides an electronic device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of any of the intelligent wastewater treatment monitoring methods.

[0038] The present invention has the following beneficial effects:

[0039] 1. Solving the problem of false alarms under scenarios of drastic fluctuations in influent load and temperature: Traditional methods use fixed thresholds. When the influent flow rate changes drastically due to heavy rain, peak day and night water usage, or seasonal or weather changes, the microbial activity and reaction rate of the entire treatment system will change, resulting in changes in the normal fluctuation range of water quality parameters. This solution establishes independent threshold ranges for different operating conditions such as "high flow and high temperature" and "low flow and low temperature" through cluster analysis. This allows for automatic judgment based on reasonable standards when switching operating conditions, significantly reducing false alarms caused by normal changes in operating conditions.

[0040] 2. Solving the challenge of accurate monitoring under multiple operating modes in wastewater treatment plants: Wastewater treatment plants do not operate in a single mode in reality; they may operate under multiple modes such as high load, low load, and seasonal operation. The normal levels of water quality parameters vary greatly under different modes. This solution automatically extracts different representative operating condition clusters from historical data through unsupervised learning (clustering). Without the need for manual pre-definition of modes, it can achieve multi-mode self-identification and adaptive monitoring, enabling the monitoring system to flexibly adapt to the complex actual operating conditions of the plant.

[0041] 3. Solving the problem of slow or unknown changes in influent water quality characteristics: When the population structure, industrial enterprise emission ratio, etc. in the service area change, the inherent characteristics of influent water quality (such as biodegradability) may drift slowly. Fixed thresholds cannot adapt to such changes. This solution updates the historical data pool on a "daily" basis and recalculates the operating condition clusters and thresholds, so that the threshold range can be adaptively updated as the water quality baseline drifts slowly, maintaining the long-term effectiveness of the early warning system. Attached Figure Description

[0042] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 This is a flowchart of an intelligent wastewater treatment monitoring method provided in one embodiment of the present invention;

[0044] Figure 2 A framework diagram of an intelligent wastewater treatment monitoring system provided in one embodiment of the present invention;

[0045] Figure 3 This is a schematic diagram of the structure of an electronic device provided in one embodiment of the present invention. Detailed Implementation

[0046] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of an intelligent wastewater treatment monitoring method and system proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0047] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0048] The following description, in conjunction with the accompanying drawings, details the specific scheme of the intelligent wastewater treatment monitoring method and system provided by the present invention.

[0049] Please see Figure 1 The diagram illustrates a flowchart of an intelligent wastewater treatment monitoring method according to an embodiment of the present invention, the method comprising:

[0050] Step S1: Obtain the influent flow rate, influent temperature, and different types of water quality parameters of the wastewater treatment plant at each moment in the historical period.

[0051] Sewage treatment typically requires strict control to prevent substandard wastewater from flowing into the environment and causing pollution. Existing methods usually monitor core water quality parameters such as Chemical Oxygen Demand (COD) and dissolved oxygen and compare them with fixed thresholds to monitor and warn of abnormal states that may occur during the sewage treatment process (such as abnormally high COD). However, the influent conditions in the sewage treatment process are quite complex, and the abnormal manifestations of water quality parameters under different influent conditions are also different. This makes the existing method of comparing fixed thresholds unable to accurately identify and warn of abnormal states in the sewage treatment process.

[0052] In the process of purifying wastewater at a wastewater treatment plant, the influent flow rate and influent temperature are key factors affecting the influent operating conditions. The influent flow rate directly determines the hydraulic load and pollutant load of the wastewater treatment plant. For example, when the influent flow rate is unstable, the residence time of wastewater in structures such as sedimentation tanks and biological treatment tanks will also be unstable, and the concentration of pollutants (COD, ammonia nitrogen) degraded by microorganisms in the biological treatment tank will also fluctuate. Furthermore, the influent temperature is a key physical parameter affecting microbial activity, especially in biological treatment processes (especially denitrification). In the case of phosphorus, water temperature plays a decisive role. For example, fluctuations in influent water temperature lead to unstable activity of nitrifying bacteria, which in turn causes fluctuations in the ammonia nitrogen content in the wastewater. Therefore, in this embodiment of the invention, a flow sensor and a temperature sensor are first installed at the inlet to monitor the influent flow rate and influent water temperature. The influent flow rate and influent water temperature of the wastewater treatment plant are also collected at each moment in a historical period. In one embodiment of the invention, the historical period is set to the most recent year, and the data is collected on a daily basis.

[0053] In the process of wastewater treatment, it is usually necessary to monitor water quality parameters such as COD, ammonia nitrogen, total phosphorus, and pH, which reflect the water quality status, in order to monitor any abnormal water quality conditions that may occur during the wastewater treatment process. Therefore, this embodiment of the invention also requires the installation of various sensors for analyzing water quality parameters (such as COD analyzers, ammonia nitrogen analyzers, pH sensors, etc.) at the outlet to assess whether there are any abnormalities in the effluent water quality, and to use the corresponding analyzers or sensors to monitor the water quality parameters at the outlet, and to collect different types of water quality parameters of the wastewater treatment plant at each moment in historical periods.

[0054] It should be noted that different types of data have different dimensions. Therefore, the embodiments of the present invention also need to standardize the collected different types of data to eliminate the influence of dimensions. Data standardization is a technical means well known to those skilled in the art and will not be elaborated here.

[0055] Step S2: Based on the distribution of influent flow rate and influent temperature at each time of day, obtain the daily operating condition feature vector; based on the operating condition feature vector, cluster all days included in the historical period to obtain multiple operating condition clusters; perform statistical analysis on each water quality parameter at all times of all days in each operating condition cluster to obtain the threshold range of each water quality parameter in each operating condition cluster.

[0056] Since the inlet flow rate and inlet water temperature are key factors affecting the inlet operating conditions, as the above analysis shows, when the inlet flow rate and inlet water temperature are unstable or fluctuate, the water quality parameters at the outlet will also be unstable or fluctuate accordingly. This makes it impossible to accurately identify abnormal water quality states in wastewater treatment using existing fixed thresholds. Therefore, this embodiment of the invention first analyzes the distribution of inlet flow rate and inlet water temperature at each time of day, using days as the unit, to construct an operating condition feature vector representing the characteristics of inlet flow rate and inlet water temperature on a daily basis. Subsequently, based on the operating condition feature vector, days with similar inlet operating conditions can be classified into the same category. Thus, based on the type of inlet operating condition, the threshold range for judging water quality parameters can be accurately calculated, expanding the static threshold to a dynamic range that adapts to the operating conditions, avoiding the inability of a single threshold to adapt to complex operating conditions.

[0057] Preferably, in one embodiment of the present invention, the method for obtaining the daily working condition feature vector specifically includes:

[0058] First, the dispersion of the inflow rate at all times of the day is analyzed to obtain the daily inflow rate disorder. The greater the inflow rate disorder on a certain day, the more unstable the inflow rate at each time of that day.

[0059] In embodiments of the present invention, the standard deviation or variance of the influent flow rate at all times of the day can be used as the daily influent flow rate disorder, thereby enabling the analysis of the dispersion of the influent flow rate at all times of the day, which is not limited here.

[0060] The average inflow rate at all times of the day is taken as the overall inflow rate for that day. The higher the overall inflow rate on a given day, the higher the overall level of inflow rate at all times of that day.

[0061] Using the influent flow disorder as the numerator and the overall influent flow as the denominator, the ratio is used as the daily influent flow variation coefficient. The influent flow variation coefficient is a relative indicator that reflects the degree of dispersion of influent flow at different times within a unit of day. It avoids the interference of the absolute value of influent flow on the judgment of the degree of fluctuation. It can accurately reflect the stability of influent load and is one of the core indicators for operating condition classification. The larger the value, the more drastic the fluctuation of influent flow relative to its average level, and the worse the stability of influent load.

[0062] Then, a straight line is fitted to the inflow rate at all times of the day, and the slope of the fitted straight line is used as the daily inflow rate trend value. The inflow rate trend value reflects the daily inflow rate change trend. In the embodiments of the present invention, the least squares method can be used for straight line fitting, which is not limited or elaborated here.

[0063] Meanwhile, considering that the influent water temperature can directly affect the activity of microorganisms, when the water temperature is low in winter, the degradation efficiency of microorganisms decreases and the normal fluctuation range of water quality parameters will expand. When the water temperature is high in summer, the degradation efficiency increases and the fluctuation range shrinks. Influent water temperature is also an important environmental factor for operating condition classification. Therefore, in this embodiment of the invention, the average influent water temperature at all times of the day is taken as the overall influent water temperature for the day. The overall influent water temperature reflects the overall level of influent water temperature at each time of the day.

[0064] Furthermore, the three-dimensional vector composed of the daily influent flow variation coefficient, influent flow trend value, and overall influent water temperature is used as the daily operating condition feature vector.

[0065] After obtaining the daily operating condition feature vector, which reflects the influent operating condition characteristics of a unit day through the influent flow rate and influent water temperature, this embodiment of the invention further clusters all days included in the historical period based on the operating condition feature vector, thereby grouping unit days with similar influent operating conditions in the historical period into the same cluster, so that each operating condition cluster represents a type of influent operating condition. Subsequently, for each operating condition cluster, the dynamic range of water quality parameters with appropriate operating conditions can be accurately analyzed.

[0066] Preferably, in one embodiment of the present invention, the method for obtaining multiple working condition clusters specifically includes:

[0067] Before clustering, a distance metric between any two days needs to be defined. Therefore, the distance metric between any two days can be obtained first based on the difference between the working condition feature vectors of any two days. Subsequently, clustering operations can be performed on the unit day based on the distance metric.

[0068] Preferably, in one embodiment of the present invention, the Euclidean distance between the feature vectors of the operating conditions of any two days can be used as the distance metric between any two days. In other embodiments of the present invention, for example, the Manhattan distance can also be used, and there is no limitation here.

[0069] Then, the K-means clustering algorithm is used, and based on the distance metric between any two days, all days in the historical period are clustered to obtain multiple working condition clusters. The number of clusters can be determined based on the existing elbow method. In other embodiments of the present invention, other distance-based clustering algorithms such as K-Medoids can also be used, which are not limited here.

[0070] The distribution characteristics of influent flow rate and influent water temperature are similar across units of day within the same operating condition cluster. In other words, each operating condition cluster represents a type of influent operating condition. Therefore, this embodiment of the invention further analyzes the water quality parameters under each type of influent operating condition for each operating condition cluster to obtain the threshold range of each water quality parameter in each operating condition cluster. This expands the static threshold into a dynamic range that adapts to the operating condition, avoiding the problem that a single threshold cannot adapt to complex operating conditions, which could lead to subsequent errors in water quality anomaly identification. At the same time, using a threshold form with an interval range is more flexible than using a single value threshold form, can accommodate normal fluctuations in data, and has a higher tolerance for anomaly identification.

[0071] Preferably, in one embodiment of the present invention, the method for obtaining the threshold range of each water quality parameter for each operating condition cluster specifically includes:

[0072] Using any water quality parameter as the target water quality parameter, box plot analysis is performed on the target water quality parameter for all days and all times in each operating condition cluster to obtain the threshold range of the target water quality parameter for each operating condition cluster. Box plots are a well-known technique in the field. By analyzing the data, a box plot can obtain a range; data within this range is considered normal data, and data outside this range is considered abnormal data. In other embodiments of the present invention, for example... The principle is to obtain the threshold range of the target water quality parameters for each operating condition cluster, without making any restrictions here.

[0073] Using the same method described above, the threshold range for each water quality parameter in each operating condition cluster can be obtained.

[0074] Step S3: Real-time acquisition of influent flow rate, influent temperature, and different types of water quality parameters of the wastewater treatment plant at different times in the current short period. Based on the distribution of influent flow rate and influent temperature at each time in the current short period, obtain the real-time operating condition feature vector for the current short period. Based on the real-time operating condition feature vector for the current short period and the distance between each operating condition cluster, obtain the matching threshold range for each water quality parameter in the current short period.

[0075] After obtaining the threshold range of each water quality parameter for each operating condition cluster through the above steps, the abnormal water quality status of the sewage treatment plant during the sewage treatment process can be identified and warned in real time. In this embodiment of the invention, the same method as in step S1 is used to collect the influent flow rate, influent water temperature and different types of water quality parameters of the sewage treatment plant at different times in the current short period. The value range of the current short period is usually 1 to 2 hours to ensure the real-time nature of water quality monitoring. In one embodiment of the invention, the current short period is set to 1 hour.

[0076] It should be noted that after collecting the influent flow rate, influent temperature, and different types of water quality parameters at different times in the current short period, the collected data needs to be standardized using the same method as in step S1 to ensure the consistency of data processing.

[0077] Then, based on the distribution of influent flow rate and influent temperature at various moments within the current short period, it is necessary to analyze the characteristics of the wastewater treatment plant's influent operating conditions during the current short period. This will yield a real-time operating condition feature vector for the current short period. Subsequently, based on this real-time short-term operating condition feature vector and combined with various operating condition clusters, the operating condition type of the wastewater treatment plant in the current short period can be accurately matched. This will allow for accurate analysis of the influent operating condition type during the current short period, facilitating the dynamic adjustment of the threshold range of water quality parameters based on the current influent operating conditions. This avoids the problem of a single threshold being unable to adapt to complex operating conditions, which could lead to incorrect identification of water quality anomalies.

[0078] Preferably, in one embodiment of the present invention, the method for obtaining the real-time operating condition feature vector for the current short period specifically includes:

[0079] Using the same calculation method as the daily operating condition feature vector, based on the distribution of influent flow rate and influent temperature at each moment of the current short period, the real-time operating condition feature vector for the current short period is obtained. Specifically, the process is as follows: First, the dispersion of influent flow rate at all moments within the current short period is analyzed to obtain the real-time influent flow rate disorder. The average influent flow rate at all moments within the current short period is taken as the real-time overall influent flow rate for the current short period. The influent flow rate disorder is used as the numerator, the overall influent flow rate as the denominator, and the ratio is used as the... The coefficient of variation of the real-time influent flow rate for the current short period is calculated. Then, using the same linear fitting method, a linear fit is performed on the influent flow rate at all times within the current short period. The slope of the fitted line is taken as the real-time influent flow rate trend value for the current short period. The average influent water temperature at all times within the current short period is taken as the real-time overall influent water temperature for the current short period. Finally, the three-dimensional vector composed of the real-time influent flow rate coefficient of variation, the real-time influent flow rate trend value, and the real-time overall influent water temperature for the current short period is taken as the real-time operating condition feature vector for the current short period.

[0080] After obtaining the real-time operating condition feature vector for the current short period, the operating condition type of the wastewater treatment plant can be accurately matched according to the real-time operating condition feature vector for the current short period and the distance between each operating condition cluster. This yields the matching threshold range for each water quality parameter in the current short period. The matching threshold range for each water quality parameter in the current short period is adapted to the current influent operating condition of the wastewater treatment plant. Subsequently, based on the matching threshold range, the abnormal water quality status at the outlet of the wastewater treatment plant under the current influent operating condition can be accurately identified and warned.

[0081] Preferably, in one embodiment of the present invention, the method for obtaining the matching threshold range of each water quality parameter in the current short time period specifically includes:

[0082] First, it is necessary to analyze the type of influent operating conditions in the current short period. The average value of the operating condition feature vectors for all days in each operating condition cluster is used as the cluster center of each operating condition cluster. The closer the real-time operating condition feature vector of the current short period is to the cluster center of a certain operating condition cluster, the more likely that the influent operating condition type represented by that operating condition cluster is the influent operating condition type of the wastewater treatment plant in the current short period. Therefore, the matching operating condition clusters for the current short period can be selected from all operating condition clusters based on the distance between the real-time operating condition feature vector of the current short period and the cluster center of each operating condition cluster.

[0083] Preferably, in one embodiment of the present invention, the method for obtaining the current short-term matching condition cluster specifically includes:

[0084] The Euclidean distance between the real-time operating condition feature vector of the current short period and the cluster center of each operating condition cluster is used as the operating condition difference degree between each operating condition cluster and the current short period. The smaller the operating condition difference degree between a certain operating condition cluster and the current short period, the more the water intake operating condition type represented by the operating condition cluster matches the water intake operating condition characteristics of the current short period.

[0085] Therefore, the cluster of operating conditions corresponding to the minimum difference between the current short-term operating conditions and the current short-term operating conditions can be used as the matching cluster of operating conditions for the current short-term operating conditions.

[0086] Then, the threshold range of each water quality parameter in the matching working condition cluster is used as the matching threshold range of each water quality parameter in the current short period.

[0087] Thus, for each water quality parameter, a matching threshold range that conforms to the current short-term influent conditions of the wastewater treatment plant has been obtained. Subsequently, the matching threshold range can be used to accurately identify and warn of water quality anomalies during the wastewater treatment process.

[0088] Step S4: Compare each water quality parameter and its corresponding matching threshold range at each moment in the current short period, and issue an abnormal warning for each water quality parameter.

[0089] The matching threshold range for each water quality parameter in the current short period of time is adapted to the current influent operating conditions of the sewage treatment plant. Therefore, each water quality parameter and its corresponding matching threshold range at each moment in the current short period of time can be compared to provide anomaly warnings for each water quality parameter. This avoids the problem of incorrect water quality anomaly identification caused by a single threshold not being able to adapt to complex operating conditions.

[0090] Preferably, in one embodiment of the present invention, the method for issuing anomaly warnings for each water quality parameter specifically includes:

[0091] For any water quality parameter, if, within a short period of time, the water quality parameter does not fall within the matching threshold range for at least a preset number of consecutive moments, it indicates a high probability that the water quality parameter is abnormal. The system then issues an abnormality warning for this water quality parameter. This requirement for continuous warnings filters out instantaneous measurement noise or accidental interference, improving sensitivity while ensuring alarm reliability. The preset number of moments typically ranges from [value missing]. The integer within, where N represents the number of sampling moments contained in the current short period. In one embodiment of the present invention, the preset number is set to ,in, This indicates the rounding up symbol. The preset number can be set by the implementer according to the specific implementation scenario, and is not limited here. The abnormal warning method can use, for example, traffic lights or buzzers, and is not limited here.

[0092] It should be noted that the embodiments of the present invention also need to update the historically collected data (including influent flow rate, influent water temperature and different types of water quality parameters). For example, data from the most recent historical period (e.g., within the most recent year) can be re-collected every month or quarter, and then clustering and threshold range analysis can be performed again to avoid problems such as data drift and model degradation.

[0093] This invention also provides an intelligent wastewater treatment monitoring system; please refer to [link / reference]. Figure 2 The diagram illustrates a framework of an intelligent wastewater treatment monitoring system according to an embodiment of the present invention. The system includes:

[0094] The historical data acquisition module 101 is used to acquire the influent flow rate, influent temperature, and different types of water quality parameters of the sewage treatment plant at each moment in the historical period.

[0095] The operating condition segmentation module 102 is used to obtain the daily operating condition feature vector based on the distribution of influent flow rate and influent temperature at each time of day; based on the operating condition feature vector, it clusters all days included in the historical period to obtain multiple operating condition clusters; and it performs statistical analysis on each water quality parameter at all times of all days in each operating condition cluster to obtain the threshold range of each water quality parameter in each operating condition cluster.

[0096] The operating condition matching module 103 is used to acquire in real time the influent flow rate, influent temperature and different types of water quality parameters of the sewage treatment plant at different times in the current short period. Based on the distribution of influent flow rate and influent temperature at each time in the current short period, the module obtains the real-time operating condition feature vector of the current short period. Based on the real-time operating condition feature vector of the current short period and the distance between each operating condition cluster, the module obtains the matching threshold range of each water quality parameter in the current short period.

[0097] The water quality monitoring module 104 is used to compare each water quality parameter at each moment in the current short period with the corresponding matching threshold range, and to issue an abnormal warning for each water quality parameter.

[0098] It should be noted that the above explanation of an embodiment of an intelligent wastewater treatment monitoring method also applies to an intelligent wastewater treatment monitoring system of the same embodiment, and will not be repeated here.

[0099] This invention also provides an electronic device, please refer to [link / reference]. Figure 3The diagram illustrates the structure of an electronic device provided in an embodiment of the present invention. The electronic device may include: a memory 201, a processor 202, and a computer program stored in the memory 201 and executable on the processor 202. When the processor 202 executes the program, it implements an intelligent wastewater treatment monitoring method provided in the above embodiment.

[0100] Furthermore, the electronic device also includes a communication interface 203 for communication between the memory 201 and the processor 202.

[0101] The memory 201 may include high-speed RAM memory, and may also include nonvolatile memory, such as at least one disk storage.

[0102] If the memory 201, processor 202, and communication interface 203 are implemented independently, then the communication interface 203, memory 201, and processor 202 can be interconnected via a bus to complete communication between them. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 3 A bus is represented by a single line, but this does not mean that there is only one bus or one type of bus.

[0103] Optionally, in a specific implementation, if the memory 201, processor 202, and communication interface 203 are integrated on a single chip, then the memory 201, processor 202, and communication interface 203 can communicate with each other through an internal interface.

[0104] The processor 202 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention.

[0105] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0106] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A smart wastewater treatment monitoring method, characterized in that, The method includes: Obtain the influent flow rate, influent temperature, and different types of water quality parameters of the wastewater treatment plant at each moment in a historical period; Based on the distribution of the influent flow rate and the distribution of the influent temperature at each time of day, a daily operating condition feature vector is obtained; based on the operating condition feature vector, all days included in the historical period are clustered to obtain multiple operating condition clusters; statistical analysis is performed on each water quality parameter at all times of all days in each operating condition cluster to obtain the threshold range of each water quality parameter in each operating condition cluster. The system acquires in real-time data on influent flow rate, influent temperature, and different types of water quality parameters at various times within the current short period. Based on the distribution of influent flow rate and influent temperature at each time point within the current short period, it obtains a real-time operating condition feature vector for the current short period. Based on the real-time operating condition feature vector for the current short period and the distance between each operating condition cluster, it obtains a matching threshold range for each water quality parameter within the current short period. The system compares each water quality parameter at each moment of the current short period with the corresponding matching threshold range, and issues an anomaly warning for each water quality parameter.

2. The intelligent wastewater treatment monitoring method according to claim 1, characterized in that, The process of obtaining the daily work condition feature vector includes: The dispersion of the influent flow rate at all times of the day is analyzed to obtain the daily influent flow rate disorder. The average of the inflow rate at all times of the day is taken as the total inflow rate for the day. The influent flow disorder is used as the numerator, the overall influent flow is used as the denominator, and the ratio is used as the daily influent flow variation coefficient. A linear fit is performed on the influent flow rate at all times of the day, and the slope of the fitted line is used as the daily influent flow rate trend value. The average of the inlet water temperature at all times of the day is taken as the overall inlet water temperature for the day. The three-dimensional vector consisting of the daily influent flow rate variation coefficient, the influent flow rate trend value, and the overall influent water temperature is used as the daily operating condition feature vector.

3. The intelligent wastewater treatment monitoring method according to claim 1, characterized in that, The process of obtaining multiple working condition clusters includes: The Euclidean distance between the feature vectors of the operating conditions of any two days is used as the distance metric between any two days. Using the K-means clustering algorithm and based on the distance metric between any two days, clusters are formed for all days within a historical time period to obtain multiple working condition clusters.

4. The intelligent wastewater treatment monitoring method according to claim 1, characterized in that, The threshold range for obtaining each water quality parameter in each operating condition cluster includes: Using any water quality parameter as the target water quality parameter, box plot analysis is performed on the target water quality parameter for all days and all times in each operating condition cluster to obtain the threshold range of the target water quality parameter in each operating condition cluster.

5. The intelligent wastewater treatment monitoring method according to claim 2, characterized in that, The process of obtaining the real-time operating condition feature vector for the current short period includes: Using the same calculation method as the daily operating condition feature vector, the real-time operating condition feature vector for the current short period is obtained based on the distribution of the influent flow rate and the distribution of the influent water temperature at each moment of the current short period.

6. The intelligent wastewater treatment monitoring method according to claim 1, characterized in that, The matching threshold range for each water quality parameter in the current short time period includes: The average value of the feature vectors of all days in each working condition cluster is used as the cluster center of each working condition cluster. Based on the real-time operating condition feature vector of the current short period and the distance between the cluster centers of each operating condition cluster, the matching operating condition cluster of the current short period is selected from all operating condition clusters. The threshold range of each water quality parameter in the matching condition cluster is used as the matching threshold range for each water quality parameter in the current short period.

7. The intelligent wastewater treatment monitoring method according to claim 6, characterized in that, The step of selecting the matching working condition cluster for the current short period from all working condition clusters includes: The Euclidean distance between the real-time operating condition feature vector of the current short period and the cluster center of each operating condition cluster is used as the operating condition difference degree between each operating condition cluster and the current short period. The cluster of operating conditions corresponding to the minimum value of the operating condition difference between the current short period and the current short period is taken as the matching operating condition cluster of the current short period.

8. The intelligent wastewater treatment monitoring method according to claim 1, characterized in that, The abnormal warning for each water quality parameter includes: For any water quality parameter, if, within the current short period of time, the water quality parameter does not fall within the matching threshold range for at least a preset number of consecutive moments, the system will issue an abnormal warning for the water quality parameter.

9. An intelligent wastewater treatment monitoring system, the system comprising: The historical data acquisition module is used to obtain the influent flow rate, influent temperature, and different types of water quality parameters of the wastewater treatment plant at each moment in a historical period. The operating condition segmentation module is used to obtain a daily operating condition feature vector based on the distribution of the influent flow rate and the distribution of the influent water temperature at each time of day; based on the operating condition feature vector, clustering is performed on all days included in the historical period to obtain multiple operating condition clusters; statistical analysis is performed on each water quality parameter at all times of all days in each operating condition cluster to obtain the threshold range of each water quality parameter in each operating condition cluster. The operating condition matching module is used to acquire in real time the influent flow rate, influent temperature, and different types of water quality parameters of the wastewater treatment plant at different times in the current short period. Based on the distribution of the influent flow rate and the distribution of the influent temperature at each time in the current short period, the module obtains the real-time operating condition feature vector for the current short period. Based on the real-time operating condition feature vector for the current short period and the distance between each operating condition cluster, the module obtains the matching threshold range for each water quality parameter in the current short period. The water quality monitoring module is used to compare each water quality parameter at different times in the current short period with the corresponding matching threshold range, and to issue an abnormal warning for each water quality parameter.

10. An electronic device, characterized in that, The electronic device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement an intelligent wastewater treatment monitoring method as described in any one of claims 1 to 8.

Images