Oily water treatment equipment control method and system based on Internet of Things

Abstract

The invention provides an oily water treatment equipment control method based on the Internet of Things, which comprises the following steps: feeding back data through an operation state, calling a multi-dimensional data analysis module, carrying out association calculation on energy consumption, treatment efficiency and adjustment parameters, predicting a subsequent operation trend by adopting a support vector machine algorithm, and obtaining a feasibility evaluation result of energy-saving mode switching; if the feasibility evaluation result of the energy-saving mode switching shows that the energy consumption optimization space is relatively large, starting an energy-saving mode switching module, reducing the power output of a non-key processing unit, maintaining the core separation efficiency, and determining new energy consumption balance point data; and through equipment state updating information, continuously monitoring an operation cost control module, comparing cleaned energy consumption data with historical records, calculating a cost saving range, and judging a potential improvement space for prolonging the service life of the equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment control, and in particular to an Internet of Things-based oily wastewater treatment equipment control method and system. Background Art

[0002] Oily wastewater treatment, a crucial area in environmental protection and industrial production, plays a crucial role in reducing pollution and ensuring water resource security. Its treatment efficiency and intelligence are directly linked to ecological balance and business operating costs. However, many current treatment devices still rely on traditional manual adjustments or simple automated systems, making them difficult to handle complex operating environments and emergencies. In particular, monitoring and responding to abnormalities often suffer from lags and inefficiencies, leading to poor treatment results and wasted resources.

[0003] Due to the limitations of existing methods, the control of oily wastewater treatment equipment faces significant technical challenges. The primary problem is the lack of real-time perception and dynamic response capabilities of the operating status. When the oil content of the influent suddenly increases, the system cannot adjust the treatment unit in time, which can easily cause equipment overload or incomplete treatment. As this problem deepens, another related challenge emerges, namely the lack of an intelligent trigger mechanism based on multi-dimensional data. Due to the inability to comprehensively analyze changes in key indicators such as energy consumption and efficiency, it is difficult for the equipment to automatically switch to energy-saving mode or start the self-cleaning program when an anomaly occurs, which in turn leads to increased operating costs and shortened equipment life. These two challenges are progressive and together constitute the bottleneck of intelligent control. Summary of the Invention

[0004] In order to solve the above-mentioned technical problems, the present invention provides an oily wastewater treatment equipment control method and system based on the Internet of Things.

[0005] The technical solution of the present invention is achieved as follows:

[0006] A method for controlling oily wastewater treatment equipment based on the Internet of Things, comprising:

[0007] The oil content of the inlet water is continuously collected through a sensor network. High-precision oil content sensors are deployed at key nodes in the inlet pipeline. Data streams are acquired every minute to generate a time series dataset containing oil content fluctuations, providing preliminary operational status information.

[0008] Based on the time series data set, the collected oil content fluctuation information is denoised to filter out outliers caused by environmental interference, and the short-term average and change trend are calculated to determine whether the oil content exceeds the preset threshold range;

[0009] If the oil content exceeds the preset threshold, the abnormality detection module is triggered. Combining historical operating data and current energy consumption indicators, the pre-established decision tree model is used to classify the degree of abnormality and determine whether the critical point of equipment overload risk has been reached.

[0010] Based on the abnormality degree classification results, the dynamic response mechanism module is activated to automatically adjust the operating parameters of the processing unit for scenarios with higher equipment overload risks, and obtain adjusted operating status feedback data.

[0011] Furthermore, obtaining preliminary operating status information includes:

[0012] High-precision sensors are deployed at key nodes through the sensor network to collect influent oil content data at a frequency of minutes, build real-time data streams, and obtain initial time series information;

[0013] Based on the initial time series information, the sliding window method is used to extract the characteristics of oil content fluctuation, obtain the fluctuation amplitude and periodic change characteristics, and determine the basic pattern of fluctuation;

[0014] If the basic pattern of fluctuations exceeds the preset threshold range, the anomaly detection mechanism is triggered, and the data stream is classified using the pre-established support vector machine model to determine whether there is an abnormal state;

[0015] According to the abnormal state judgment results, the local time series decomposition is performed on the abnormal data segment to obtain the trend component and noise component, and obtain the refined fluctuation composition information;

[0016] By using the refined fluctuation composition information, the autoregressive model is used to make short-term predictions on the trend component, obtain the oil content change trend in the future period, and determine the prediction sequence;

[0017] Based on the predicted sequence and combined with historical time series data, the operating status of key nodes is dynamically evaluated. If the deviation between the predicted sequence and historical data exceeds the preset range, the monitoring frequency is adjusted to obtain a more intensive data stream.

[0018] Through denser data streams, the oil content fluctuation characteristics of the real-time data acquisition module are updated, the time series data set is reconstructed, and the latest operating status information is obtained.

[0019] Furthermore, determining whether the oil content exceeds a preset threshold range includes:

[0020] The time series data set is preliminarily cleaned through the data preprocessing module, and the outliers caused by environmental interference are filtered out using denoising technology to obtain the cleaned oil content fluctuation data;

[0021] Based on the oil content fluctuation data after cleaning, the sliding window technology is used to analyze the local data segment, calculate the short-term average value, and obtain the stable characteristics within the local time period;

[0022] Focusing on the stable characteristics within a local time period, the sliding window technique is used to extract dynamic information on the changing trend and determine the short-term variation pattern of oil content fluctuations.

[0023] If the short-term change pattern shows that the oil content fluctuation exceeds the preset threshold range, the abnormal marking mechanism is triggered to distinguish the abnormal data segment from the normal data segment and obtain the marked data classification.

[0024] Furthermore, the determining whether the oil content exceeds a preset threshold range further includes:

[0025] According to the marked data classification, further data feature extraction is performed on the abnormal data segment to obtain the time point and amplitude information of the abnormal fluctuation and determine the specific distribution characteristics of the abnormality;

[0026] Based on the specific distribution characteristics of the anomaly, combined with the pre-established support vector machine model, the abnormal data segments are classified and processed to determine whether the anomaly has persistent characteristics, and the classified anomaly assessment results are obtained;

[0027] For the abnormal evaluation results after classification, if the persistence feature is confirmed, the frequency parameters of the data acquisition module are adjusted to obtain higher-density time series data and update the monitoring information of oil content fluctuations.

[0028] Furthermore, the determination of whether a critical point of equipment overload risk has been reached includes:

[0029] When the oil content exceeds the threshold, the anomaly detection process is activated, and historical records and current energy consumption data are integrated through data fusion technology to obtain preliminary evidence for anomaly determination.

[0030] Based on the preliminary abnormality judgment basis, a pre-built decision tree model is used to classify the abnormality level and determine the specific classification results of the abnormality level;

[0031] Regarding the abnormal level classification results, if the classification results show that the equipment is close to the critical state of overload, the risk assessment mechanism is triggered, and the confirmation information of the risk critical state is obtained by comparing the real-time monitoring data of the operating status.

[0032] Furthermore, the determining whether the critical point of equipment overload risk is reached also includes:

[0033] Based on the confirmed information of the critical risk state and the possibility of equipment overload, time series analysis tools are used to extract characteristics of historical fluctuations in operating status to obtain the distribution characteristics of potential overload hazards;

[0034] Based on the distribution characteristics of potential overload hazards, pattern matching is performed with similar scenarios in historical records to determine whether there are signs of persistent risks;

[0035] If signs of persistent risk are confirmed, adjust the data collection frequency parameters to obtain higher-density time series information and determine more accurate operating status monitoring data;

[0036] Based on more accurate operating status monitoring data and combined with the dynamic changes in energy consumption data, a logical comparison method is used to conduct a secondary verification of the possibility of equipment overload to obtain the final risk assessment conclusion.

[0037] Furthermore, the obtaining of the adjusted operating status feedback data includes:

[0038] Based on the classification results, the dynamic response module is activated to automatically identify operating states with high equipment overload risks in scenarios with high abnormality levels, and obtain preliminary risk assessment information.

[0039] Based on the preliminary risk assessment information, the operating parameters of the processing unit are adjusted, the separator speed and dosage are dynamically configured, and the adjusted operating status data is obtained;

[0040] Use the adjusted operating status data in combination with the pre-established decision tree model to analyze the possibility of equipment overload and determine the distribution of potential operating hazards;

[0041] If the distribution of potential operational hazards exceeds the preset threshold range, the parameter optimization process is initiated to obtain the optimized parameter configuration plan through real-time comparison of the operational status data;

[0042] According to the optimized parameter configuration scheme, the operating parameters of the processing unit are readjusted, the separator speed and the dosage are calibrated again, and the updated operating status information is obtained.

[0043] Furthermore, the obtaining of the adjusted operating status feedback data further includes:

[0044] Continuously monitor scenarios with high risk of equipment overload using updated operating status information. If high-risk situations persist, save current operating status data using data logging tools to determine if there are signs of long-term potential hazards.

[0045] If signs of long-term hidden dangers are confirmed, the data collection frequency will be adjusted to obtain higher-density operating status data and determine more accurate equipment operation monitoring information.

[0046] Furthermore, the method further comprises:

[0047] By using the operating status feedback data, the multi-dimensional data analysis module is called to correlate energy consumption, processing efficiency and adjustment parameters. The support vector machine algorithm is used to predict the subsequent operating trend and obtain the feasibility assessment results of energy-saving mode switching.

[0048] If the feasibility assessment results of energy-saving mode switching show that there is a large room for energy consumption optimization, the energy-saving mode switching module is activated to reduce the power output of non-critical processing units while maintaining the core separation efficiency and determine the new energy consumption balance point data;

[0049] Based on the energy balance point data, the self-cleaning program control module is linked to automatically trigger the equipment self-cleaning process when it detects a decrease in treatment efficiency and a stable oil content. The equipment automatically flushes the sediment on the inner wall of the pipe at regular intervals and obtains equipment status updates after the cleaning is completed.

[0050] By updating the equipment status information, the operating cost control module is continuously monitored, and the energy consumption data after cleaning is compared with the historical records to calculate the cost savings and determine the potential room for improvement in extending the equipment life.

[0051] An oily wastewater treatment equipment control system based on the Internet of Things, the system comprising:

[0052] A real-time monitoring module is used to continuously collect the oil content of the inlet water through a sensor network. High-precision oil content sensors are deployed at key nodes of the inlet pipeline to obtain data streams every minute, generating a time series dataset containing oil content fluctuations and obtaining preliminary operational status information.

[0053] The data preprocessing module is used to perform denoising on the collected oil content fluctuation information based on the time series data set, filter out abnormal values ​​caused by environmental interference, calculate the short-term average value and change trend using the sliding window technology, and determine whether the oil content exceeds the preset threshold range;

[0054] The abnormality detection module is used to trigger the abnormality detection module if the oil content exceeds the preset threshold range. It combines historical operation data and current energy consumption indicators, classifies the degree of abnormality through a pre-established decision tree model, and determines whether the critical point of equipment overload risk has been reached;

[0055] The dynamic response mechanism module is used to activate the dynamic response mechanism module based on the abnormality classification results. In scenarios with high risk of equipment overload, the module automatically adjusts the operating parameters of the processing unit, such as increasing the separator speed or the dosage, and obtains feedback data on the adjusted operating status;

[0056] The multi-dimensional data analysis module is used to call the multi-dimensional data analysis module based on the operating status feedback data, perform correlation calculations on energy consumption, processing efficiency and adjustment parameters, and use the support vector machine algorithm to predict subsequent operating trends to obtain the feasibility assessment results of energy-saving mode switching;

[0057] An energy-saving mode switching module is used to activate the energy-saving mode switching module if the feasibility assessment result of the energy-saving mode switching shows that there is a large room for energy consumption optimization, reduce the power output of non-critical processing units, while maintaining the core separation efficiency, and determine the new energy consumption balance point data;

[0058] The self-cleaning program control module is used to link the self-cleaning program control module based on the energy consumption balance point data. When it detects that the processing efficiency has decreased and the oil content is stable, it automatically triggers the equipment self-cleaning process, flushes the sediment on the inner wall of the pipeline at a regular interval, and obtains the equipment status update information after the cleaning is completed;

[0059] The operating cost control module is used to continuously monitor the operating cost control module through equipment status update information, compare the energy consumption data after cleaning with historical records, calculate the cost savings, and determine the potential improvement space for extending the equipment life.

[0060] Compared with the prior art, the present invention has the following beneficial effects:

[0061] The present invention monitors the oil content of the incoming water in real time, automatically detects abnormal situations and dynamically adjusts the operating parameters to achieve efficient operation of the equipment and energy consumption optimization. The method first uses a sensor network to continuously collect data on the oil content of the incoming water, and determines whether it exceeds the threshold after preprocessing. When an abnormality is detected, the present invention will evaluate the risk of equipment overload based on historical data and current indicators, and automatically adjust the operating parameters to cope with it. Subsequently, the present invention predicts operating trends through multi-dimensional data analysis, evaluates the feasibility of switching to energy-saving mode, and reduces the power of non-critical units at the appropriate time to achieve energy consumption optimization. In addition, the present invention also includes a self-cleaning program control, which automatically triggers cleaning when a decrease in efficiency is detected, thereby extending the life of the equipment. Through this series of intelligent measures, the present invention significantly improves the operating efficiency and energy-saving level of the oil-water separation equipment, and realizes intelligent management of the entire life cycle of the equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 This is a flow chart of a method for controlling oily wastewater treatment equipment based on the Internet of Things in Example 1;

[0063] Figure 2 This is a flow chart of a method for controlling oily wastewater treatment equipment based on the Internet of Things according to Example 2;

[0064] Figure 3This is a framework diagram of an oily wastewater treatment equipment control system based on the Internet of Things in Example 3. DETAILED DESCRIPTION

[0065] In order to make the purposes, features, and advantages of the present invention more obvious and easy to understand, 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 embodiments described below are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0066] Example 1

[0067] like Figure 1 As shown, this embodiment provides an oily wastewater treatment equipment control method based on the Internet of Things, including:

[0068] The oil content of the inlet water is continuously collected through a sensor network. High-precision oil content sensors are deployed at key nodes in the inlet pipeline. Data streams are acquired every minute to generate a time series dataset containing oil content fluctuations, providing preliminary operational status information.

[0069] Based on the time series data set, the collected oil content fluctuation information is denoised to filter out outliers caused by environmental interference, and the short-term average and change trend are calculated to determine whether the oil content exceeds the preset threshold range;

[0070] If the oil content exceeds the preset threshold, the abnormality detection module is triggered. Combining historical operating data and current energy consumption indicators, the pre-established decision tree model is used to classify the degree of abnormality and determine whether the critical point of equipment overload risk has been reached.

[0071] Based on the abnormality degree classification results, the dynamic response mechanism module is activated to automatically adjust the operating parameters of the processing unit for scenarios with higher equipment overload risks, and obtain adjusted operating status feedback data.

[0072] Furthermore, obtaining preliminary operating status information includes:

[0073] High-precision sensors are deployed at key nodes through the sensor network to collect influent oil content data at a frequency of minutes, build real-time data streams, and obtain initial time series information;

[0074] Based on the initial time series information, the sliding window method is used to extract the characteristics of oil content fluctuation, obtain the fluctuation amplitude and periodic change characteristics, and determine the basic pattern of fluctuation;

[0075] If the basic pattern of fluctuations exceeds the preset threshold range, the anomaly detection mechanism is triggered, and the data stream is classified using the pre-established support vector machine model to determine whether there is an abnormal state;

[0076] According to the abnormal state judgment results, the local time series decomposition is performed on the abnormal data segment to obtain the trend component and noise component, and obtain the refined fluctuation composition information;

[0077] By using the refined fluctuation composition information, the autoregressive model is used to make short-term predictions on the trend component, obtain the oil content change trend in the future period, and determine the prediction sequence;

[0078] Based on the predicted sequence and combined with historical time series data, the operating status of key nodes is dynamically evaluated. If the deviation between the predicted sequence and historical data exceeds the preset range, the monitoring frequency is adjusted to obtain a more intensive data stream.

[0079] Through denser data streams, the oil content fluctuation characteristics of the real-time data acquisition module are updated, the time series data set is reconstructed, and the latest operating status information is obtained.

[0080] Specifically, a sensor network is used to continuously collect the oil content of the inlet water and build a real-time monitoring module. High-precision oil content sensors can be deployed at key nodes in the inlet pipeline. For example, a sensor device with a measurement accuracy of 0.01% can be installed at the main pipeline inlet and the branch confluence. The data stream is automatically collected once a minute to generate a time series data set containing oil content fluctuations. Specifically, the oil content value (such as 0.05% and 0.06%) is recorded every minute and transmitted to the cloud database via a wireless network to form a record table corresponding to the timestamp and oil content.

[0081] Next, a time series analysis algorithm, such as the moving average method, is used to calculate the average oil content over the past five minutes (for example, the data for the last five minutes is 0.05%, 0.06%, 0.04%, 0.05%, and 0.07%, with an average of 0.054%). This is used to smooth out data noise and preliminarily determine whether the operating status is stable. If the average fluctuation exceeds the set threshold of 0.02%, the system automatically marks it as an abnormal state and generates a warning signal.

[0082] At the same time, combined with historical data, time series forecasting models such as ARIMA are used to predict the oil content trend for the next 10 minutes. If the predicted value shows that the oil content may rise to 0.08%, the system will automatically compare it with historical abnormal cases and analyze the potential cause, which may be upstream equipment leakage. The analysis results are stored in the status log.

[0083] To further improve monitoring, the system can correlate water velocity data (assuming the current velocity is 2.5 m / s) and calculate the correlation coefficient between oil content and velocity (e.g., 0.75) through a correlation analysis algorithm. If the correlation is higher than 0.7, it is inferred that oil content fluctuations may be affected by the velocity, and the monitoring frequency is automatically adjusted to once every 30 seconds to improve data resolution.

[0084] Ultimately, all data and analysis results are updated to the visualization platform in real time, generating oil content fluctuation curves and status reports to ensure that operating status information is comprehensive and dynamically accessible.

[0085] Furthermore, determining whether the oil content exceeds a preset threshold range includes:

[0086] The time series data set is preliminarily cleaned through the data preprocessing module, and the outliers caused by environmental interference are filtered out using denoising technology to obtain the cleaned oil content fluctuation data;

[0087] Based on the oil content fluctuation data after cleaning, the sliding window technology is used to analyze the local data segment, calculate the short-term average value, and obtain the stable characteristics within the local time period;

[0088] Focusing on the stable characteristics within a local time period, the sliding window technique is used to extract dynamic information on the changing trend and determine the short-term variation pattern of oil content fluctuations.

[0089] If the short-term change pattern shows that the oil content fluctuation exceeds the preset threshold range, the abnormal marking mechanism is triggered to distinguish the abnormal data segment from the normal data segment and obtain the marked data classification.

[0090] Furthermore, the determining whether the oil content exceeds a preset threshold range further includes:

[0091] According to the marked data classification, further data feature extraction is performed on the abnormal data segment to obtain the time point and amplitude information of the abnormal fluctuation and determine the specific distribution characteristics of the abnormality;

[0092] Based on the specific distribution characteristics of the anomaly, combined with the pre-established support vector machine model, the abnormal data segments are classified and processed to determine whether the anomaly has persistent characteristics, and the classified anomaly assessment results are obtained;

[0093] For the abnormal evaluation results after classification, if the persistence feature is confirmed, the frequency parameters of the data acquisition module are adjusted to obtain higher-density time series data and update the monitoring information of oil content fluctuations.

[0094] Specifically, to process oil content fluctuations, the system first uses the data preprocessing module to denoise the collected time series data. It then uses a median filter algorithm to analyze each set of 10-minute consecutive data. For example, if a certain data segment has values ​​of 0.03%, 0.09%, 0.04%, 0.03%, and 0.05%, and 0.09% is identified as an outlier, the system automatically replaces it with the median of the adjacent data, 0.04%, thereby filtering out sudden changes caused by environmental interference and ensuring data smoothness.

[0095] Next, the system uses sliding window technology to set the window size to 3 minutes and calculate the short-term average. For example, if the data for the last 3 minutes is 0.04%, 0.03%, and 0.05%, the calculated average is 0.04%. The system further analyzes the changing trend of the average values ​​of adjacent windows. If the average value of the previous window is 0.045% and the current window drops to 0.04%, the system records the trend as a decrease of 0.005%.

[0096] The system then compares the short-term average value with a preset threshold range of 0.02% to 0.06%. If the calculated average value of 0.04% is within the range, it is marked as normal. If it is outside the range, for example, the average value of a certain window is 0.07%, the system automatically records it as an abnormality and associates it with the pipeline pressure data (assuming the current pressure is 1.2 MPa). The built-in algorithm analyzes the correlation between pressure and oil content fluctuations. If the correlation coefficient reaches 0.8, it is inferred that the abnormality may be related to pressure fluctuations. An analysis log is automatically generated and stored in the database to ensure that subsequent tracing can be based on evidence.

[0097] The entire process is completed automatically by the system, and data processing and analysis results are archived in real time to form a complete data chain and provide support for subsequent business optimization.

[0098] Furthermore, the determination of whether a critical point of equipment overload risk has been reached includes:

[0099] When the oil content exceeds the threshold, the anomaly detection process is activated, and historical records and current energy consumption data are integrated through data fusion technology to obtain preliminary evidence for anomaly determination.

[0100] Based on the preliminary abnormality judgment basis, a pre-built decision tree model is used to classify the abnormality level and determine the specific classification results of the abnormality level;

[0101] Regarding the abnormal level classification results, if the classification results show that the equipment is close to the critical state of overload, the risk assessment mechanism is triggered, and the confirmation information of the risk critical state is obtained by comparing the real-time monitoring data of the operating status.

[0102] Furthermore, the determining whether the critical point of equipment overload risk is reached also includes:

[0103] Based on the confirmed information of the critical risk state and the possibility of equipment overload, time series analysis tools are used to extract characteristics of historical fluctuations in operating status to obtain the distribution characteristics of potential overload hazards;

[0104] Based on the distribution characteristics of potential overload hazards, pattern matching is performed with similar scenarios in historical records to determine whether there are signs of persistent risks;

[0105] If signs of persistent risk are confirmed, adjust the data collection frequency parameters to obtain higher-density time series information and determine more accurate operating status monitoring data;

[0106] Based on more accurate operating status monitoring data and combined with the dynamic changes in energy consumption data, a logical comparison method is used to conduct a secondary verification of the possibility of equipment overload to obtain the final risk assessment conclusion.

[0107] Specifically, during the oil content monitoring process, if the system detects that the oil content exceeds the preset threshold range, for example, the set range is 0.01% to 0.05%, and the current value is 0.08%, the abnormal situation detection module is automatically triggered for in-depth analysis;

[0108] First, the system calls up historical operating data, such as the oil content records from the past 24 hours. Assuming an average of 0.03% and a standard deviation of 0.01%, and combining this with current energy consumption indicators, for example, if the equipment's operating power is 1500 watts, which exceeds the normal range of 1200 to 1400 watts, the system initially determines that there is an abnormal fluctuation.

[0109] Next, the system classifies the degree of abnormality using a pre-established decision tree model. This model, trained based on historical data, contains multiple decision nodes. For example, if the oil content exceeds a threshold of 0.02% and energy consumption increases by more than 10%, it is classified as a moderate abnormality; if the oil content exceeds 0.03% and energy consumption increases by more than 20%, it is classified as a severe abnormality.

[0110] Taking the current data as an example, if the oil content exceeds 0.03%, the energy consumption increases by 25%, and the system determines it as a severe abnormality;

[0111] The system then further assesses whether the critical point of equipment overload risk has been reached. Suppose the critical point is defined as an oil content of 0.09% and an energy consumption of 1600 watts. The current data has not yet reached this point, but the system will analyze historical trends. If the oil content has continued to rise from 0.04% to 0.08% over the past 6 hours, it is predicted that the critical point may be approached within the next 2 hours.

[0112] To complete the logical chain, the system also correlates temperature data. For example, if the current device temperature is 75 degrees Celsius, which is higher than the normal range of 60 to 70 degrees Celsius, the system infers that the anomaly may be related to overheating. A risk warning report is automatically generated and stored in the database for subsequent analysis and reference.

[0113] The entire process is automatically executed by the system, ensuring seamless integration of real-time monitoring and risk assessment.

[0114] Furthermore, the obtaining of the adjusted operating status feedback data includes:

[0115] Based on the classification results, the dynamic response module is activated to automatically identify operating states with high equipment overload risks in scenarios with high abnormality levels, and obtain preliminary risk assessment information.

[0116] Based on the preliminary risk assessment information, the operating parameters of the processing unit are adjusted, the separator speed and dosage are dynamically configured, and the adjusted operating status data is obtained;

[0117] Use the adjusted operating status data in combination with the pre-established decision tree model to analyze the possibility of equipment overload and determine the distribution of potential operating hazards;

[0118] If the distribution of potential operational hazards exceeds the preset threshold range, the parameter optimization process is initiated to obtain the optimized parameter configuration plan through real-time comparison of the operational status data;

[0119] According to the optimized parameter configuration scheme, the operating parameters of the processing unit are readjusted, the separator speed and the dosage are calibrated again, and the updated operating status information is obtained.

[0120] Furthermore, the obtaining of the adjusted operating status feedback data further includes:

[0121] Continuously monitor scenarios with high risk of equipment overload using updated operating status information. If high-risk situations persist, save current operating status data using data logging tools to determine if there are signs of long-term potential hazards.

[0122] If signs of long-term hidden dangers are confirmed, the data collection frequency will be adjusted to obtain higher-density operating status data and determine more accurate equipment operation monitoring information.

[0123] Specifically, during equipment operation, when the system determines a high-risk scenario based on the abnormality classification results, it will automatically activate the dynamic response mechanism module to adjust the operating parameters of the processing unit to reduce the risk of equipment overload;

[0124] For example, if the system detects that the oil-water separation efficiency of the current separator has decreased, with a separation residual rate of 0.07%, which is higher than the normal range of 0.02% to 0.04%, and the abnormal classification result shows a high risk level, the system will immediately initiate a parameter optimization algorithm;

[0125] The algorithm calculates the optimal speed adjustment value based on historical data analysis and current flow data (for example, the processing capacity is 5000 liters per hour, which is higher than the normal 4000 liters). Assume that the linear regression model predicts that the speed needs to be increased from 800 rpm to 950 rpm to improve separation efficiency;

[0126] At the same time, based on the correlation model between the residual rate and the dosage, the system calculates that the dosage needs to be increased from 2.5 mg per liter to 3.2 mg per liter to enhance the oil-water separation effect;

[0127] After the adjustment, the system collected feedback data in real time. For example, the separation residual rate dropped to 0.03% within 30 minutes, and the flow rate was maintained at 4900 liters / hour, indicating that the parameter adjustment was effective;

[0128] If the feedback data still does not meet the standard, the system will further call the backup algorithm to analyze the flow fluctuation trend (for example, the standard deviation of the flow rate in the past two hours is 300 liters). Combined with the equipment vibration frequency data (currently 5.2 Hz, the normal range is 3.0 to 4.5 Hz), it will infer that there may be a partial blockage in the pipeline. The auxiliary cleaning program will be automatically triggered, and the cleaning cycle will be set to once an hour for 10 minutes until the vibration frequency returns to below 4.0 Hz.

[0129] The entire process is automatically executed by the system's built-in intelligent control unit, ensuring that parameter adjustment and state feedback form a closed-loop logic. At the same time, all adjustment records and feedback data are uploaded to the cloud database to facilitate subsequent optimization of the algorithm model.

[0130] Example 2

[0131] like Figure 2 As shown, this embodiment provides an oily wastewater treatment equipment control method based on the Internet of Things, including:

[0132] The oil content of the inlet water is continuously collected through a sensor network. High-precision oil content sensors are deployed at key nodes in the inlet pipeline. Data streams are acquired every minute to generate a time series dataset containing oil content fluctuations, providing preliminary operational status information.

[0133] Based on the time series data set, the collected oil content fluctuation information is denoised to filter out outliers caused by environmental interference, and the short-term average and change trend are calculated to determine whether the oil content exceeds the preset threshold range;

[0134] If the oil content exceeds the preset threshold, the abnormality detection module is triggered. Combining historical operating data and current energy consumption indicators, the pre-established decision tree model is used to classify the degree of abnormality and determine whether the critical point of equipment overload risk has been reached.

[0135] Based on the abnormality degree classification results, the dynamic response mechanism module is activated to automatically adjust the operating parameters of the processing unit for scenarios with higher equipment overload risks, and obtain adjusted operating status feedback data.

[0136] Furthermore, obtaining preliminary operating status information includes:

[0137] High-precision sensors are deployed at key nodes through the sensor network to collect influent oil content data at a frequency of minutes, build real-time data streams, and obtain initial time series information;

[0138] Based on the initial time series information, the sliding window method is used to extract the characteristics of oil content fluctuation, obtain the fluctuation amplitude and periodic change characteristics, and determine the basic pattern of fluctuation;

[0139] If the basic pattern of fluctuations exceeds the preset threshold range, the anomaly detection mechanism is triggered, and the data stream is classified using the pre-established support vector machine model to determine whether there is an abnormal state;

[0140] According to the abnormal state judgment results, the local time series decomposition is performed on the abnormal data segment to obtain the trend component and noise component, and obtain the refined fluctuation composition information;

[0141] By using the refined fluctuation composition information, the autoregressive model is used to make short-term predictions on the trend component, obtain the oil content change trend in the future period, and determine the prediction sequence;

[0142] Based on the predicted sequence and combined with historical time series data, the operating status of key nodes is dynamically evaluated. If the deviation between the predicted sequence and historical data exceeds the preset range, the monitoring frequency is adjusted to obtain a more intensive data stream.

[0143] Through denser data streams, the oil content fluctuation characteristics of the real-time data acquisition module are updated, the time series data set is reconstructed, and the latest operating status information is obtained.

[0144] Furthermore, determining whether the oil content exceeds a preset threshold range includes:

[0145] The time series data set is preliminarily cleaned through the data preprocessing module, and the outliers caused by environmental interference are filtered out using denoising technology to obtain the cleaned oil content fluctuation data;

[0146] Based on the oil content fluctuation data after cleaning, the sliding window technology is used to analyze the local data segment, calculate the short-term average value, and obtain the stable characteristics within the local time period;

[0147] Focusing on the stable characteristics within a local time period, the sliding window technique is used to extract dynamic information on the changing trend and determine the short-term variation pattern of oil content fluctuations.

[0148] If the short-term change pattern shows that the oil content fluctuation exceeds the preset threshold range, the abnormal marking mechanism is triggered to distinguish the abnormal data segment from the normal data segment and obtain the marked data classification.

[0149] Furthermore, the determining whether the oil content exceeds a preset threshold range further includes:

[0150] According to the marked data classification, further data feature extraction is performed on the abnormal data segment to obtain the time point and amplitude information of the abnormal fluctuation and determine the specific distribution characteristics of the abnormality;

[0151] Based on the specific distribution characteristics of the anomaly, combined with the pre-established support vector machine model, the abnormal data segments are classified and processed to determine whether the anomaly has persistent characteristics, and the classified anomaly assessment results are obtained;

[0152] For the abnormal evaluation results after classification, if the persistence feature is confirmed, the frequency parameters of the data acquisition module are adjusted to obtain higher-density time series data and update the monitoring information of oil content fluctuations.

[0153] Furthermore, the determination of whether a critical point of equipment overload risk has been reached includes:

[0154] When the oil content exceeds the threshold, the anomaly detection process is activated, and historical records and current energy consumption data are integrated through data fusion technology to obtain preliminary evidence for anomaly determination.

[0155] Based on the preliminary abnormality judgment basis, a pre-built decision tree model is used to classify the abnormality level and determine the specific classification results of the abnormality level;

[0156] Regarding the abnormal level classification results, if the classification results show that the equipment is close to the critical state of overload, the risk assessment mechanism is triggered, and the confirmation information of the risk critical state is obtained by comparing the real-time monitoring data of the operating status.

[0157] Furthermore, the determining whether the critical point of equipment overload risk is reached also includes:

[0158] Based on the confirmed information of the critical risk state and the possibility of equipment overload, time series analysis tools are used to extract characteristics of historical fluctuations in operating status to obtain the distribution characteristics of potential overload hazards;

[0159] Based on the distribution characteristics of potential overload hazards, pattern matching is performed with similar scenarios in historical records to determine whether there are signs of persistent risks;

[0160] If signs of persistent risk are confirmed, adjust the data collection frequency parameters to obtain higher-density time series information and determine more accurate operating status monitoring data;

[0161] Based on more accurate operating status monitoring data and combined with the dynamic changes in energy consumption data, a logical comparison method is used to conduct a secondary verification of the possibility of equipment overload to obtain the final risk assessment conclusion.

[0162] Furthermore, the obtaining of the adjusted operating status feedback data includes:

[0163] Based on the classification results, the dynamic response module is activated to automatically identify operating states with high equipment overload risks in scenarios with high abnormality levels, and obtain preliminary risk assessment information.

[0164] Based on the preliminary risk assessment information, the operating parameters of the processing unit are adjusted, the separator speed and dosage are dynamically configured, and the adjusted operating status data is obtained;

[0165] Use the adjusted operating status data in combination with the pre-established decision tree model to analyze the possibility of equipment overload and determine the distribution of potential operating hazards;

[0166] If the distribution of potential operational hazards exceeds the preset threshold range, the parameter optimization process is initiated to obtain the optimized parameter configuration plan through real-time comparison of the operational status data;

[0167] According to the optimized parameter configuration scheme, the operating parameters of the processing unit are readjusted, the separator speed and the dosage are calibrated again, and the updated operating status information is obtained.

[0168] Furthermore, the obtaining of the adjusted operating status feedback data further includes:

[0169] Continuously monitor scenarios with high risk of equipment overload using updated operating status information. If high-risk situations persist, save current operating status data using data logging tools to determine if there are signs of long-term potential hazards.

[0170] If signs of long-term hidden dangers are confirmed, the data collection frequency will be adjusted to obtain higher-density operating status data and determine more accurate equipment operation monitoring information.

[0171] Furthermore, the method further comprises:

[0172] By using the operating status feedback data, the multi-dimensional data analysis module is called to correlate energy consumption, processing efficiency and adjustment parameters. The support vector machine algorithm is used to predict the subsequent operating trend and obtain the feasibility assessment results of energy-saving mode switching.

[0173] If the feasibility assessment results of energy-saving mode switching show that there is a large room for energy consumption optimization, the energy-saving mode switching module is activated to reduce the power output of non-critical processing units while maintaining the core separation efficiency and determine the new energy consumption balance point data;

[0174] Based on the energy balance point data, the self-cleaning program control module is linked to automatically trigger the equipment self-cleaning process when it detects a decrease in treatment efficiency and a stable oil content. The equipment automatically flushes the sediment on the inner wall of the pipe at regular intervals and obtains equipment status updates after the cleaning is completed.

[0175] By updating the equipment status information, the operating cost control module is continuously monitored, and the energy consumption data after cleaning is compared with the historical records to calculate the cost savings and determine the potential room for improvement in extending the equipment life.

[0176] Furthermore, obtaining a feasibility evaluation result of energy-saving mode switching includes:

[0177] Through the operation status feedback data, the multi-dimensional analysis module is called to preliminarily organize the energy consumption data and processing efficiency data, and match and analyze the adjustment parameters to obtain the basic data set after correlation processing;

[0178] Based on the basic data set after correlation processing, the support vector machine algorithm is used to predict and analyze the subsequent trends, obtain the changing characteristics of the operating status under different parameter configurations, and determine the prediction results of the subsequent trends;

[0179] Based on the subsequent trend prediction results and the relevant requirements of the energy-saving mode, a data comparison is performed on the switching evaluation. If the prediction results show that the energy consumption data exceeds the preset threshold range, a preliminary judgment on the energy-saving mode switching is initiated, and the initial conclusion of the switching evaluation is obtained;

[0180] Based on the initial conclusions of the switching evaluation, real-time feedback data from the operating status is obtained, and a secondary analysis is conducted on the balance between processing efficiency and energy consumption data to determine whether there is room for optimization and determine the specific conditions for switching to energy-saving mode;

[0181] Based on the specific conditions for switching to energy-saving mode, a pre-established decision support tool is called to dynamically calibrate the adjustment parameters, obtain the calibrated parameter configuration plan, and obtain the optimization result suitable for the current operating status;

[0182] Based on the optimization results, the system continuously monitors the changing characteristics of the operating status and records the feedback data in real time. If the changing characteristics indicate a decrease in processing efficiency, the parameter fine-tuning process is triggered to obtain the updated operating configuration information.

[0183] Through the updated operating configuration information, combined with the multi-dimensional analysis module, subsequent trends are continuously tracked, and recording tools are used to save key data to determine the long-term feasibility assessment results of energy-saving mode switching.

[0184] Specifically, during the equipment operation status monitoring process, the system first collects real-time operating status feedback data through sensors, such as the current energy consumption of 120 kWh per hour and the treatment efficiency of 4,500 liters of wastewater per hour, and then transmits this data to the multi-dimensional data analysis module for processing;

[0185] The system automatically calls the built-in correlation calculation program to compare and analyze energy consumption, processing efficiency and historical adjustment parameters (for example, the average energy consumption in the past week was 110 kWh, and the processing efficiency fluctuated between 4300 and 4600 liters) to generate a multidimensional data matrix;

[0186] The system then uses a support vector machine algorithm to predict future operating trends. This involves mapping current data points against historical data points and calculating the nonlinear relationship between energy consumption and processing efficiency. The system predicts that energy consumption could rise to 125 kWh over the next two hours, while processing efficiency could drop to 4,400 liters per hour.

[0187] Based on this prediction, the system automatically assesses the feasibility of switching to energy-saving mode. Using a built-in evaluation model, it calculates a 78% probability that energy consumption can be reduced to 105 kWh in energy-saving mode. This is then combined with current ambient temperature data (e.g., 25.5 degrees Celsius, suitable for energy-saving mode operation) to make a comprehensive assessment.

[0188] If the evaluation results show that switching is feasible, the system will automatically generate an energy-saving mode parameter adjustment plan, such as reducing the cooling pump power from 100% to 85%, and store the data to the local server for subsequent analysis, forming a complete closed-loop logic from data collection to predictive evaluation, ensuring the stability of system operation.

[0189] Furthermore, determining the new energy consumption balance point data includes:

[0190] By starting the energy-saving mode switching module, the power output of non-critical units is preliminarily adjusted, the adjusted energy consumption data is obtained, and the initial energy consumption balance state is determined;

[0191] Based on the initial energy balance state, a pre-established monitoring tool is called to collect real-time data on the core separation efficiency to determine whether the efficiency is maintained within the preset threshold range. If the collected data is lower than the threshold, the efficiency compensation process is triggered to obtain the adjusted efficiency data.

[0192] The adjusted efficiency data is combined with the energy consumption balance status to analyze the stability of the mode switching using the support vector machine algorithm to obtain the stability evaluation results and determine whether the switching module should continue to operate;

[0193] Based on the stability assessment results, the operating status of the processing units is dynamically tracked to obtain the power output change characteristics of each unit to determine whether there is room for further optimization. If the change characteristics show imbalance, a secondary adjustment process is initiated to obtain an updated power configuration.

[0194] By combining the updated power configuration with the feasibility assessment data, the long-term adaptability of the energy-saving mode is analyzed, the adaptability analysis results are obtained, and the final conditions for mode switching are determined;

[0195] Based on the final conditions, the preset decision-making tool is called to calibrate the energy consumption balance point data, obtain the calibrated balance point information, and determine whether the requirements of the energy-saving mode are met;

[0196] Through the calibrated balance point information, the changes in the optimization space are continuously monitored, and recording tools are used to save key data to determine the operating effect of the energy-saving mode.

[0197] Specifically, when the feasibility assessment results of the energy-saving mode switch indicate significant room for energy consumption optimization, the system automatically activates the energy-saving mode switch module and uses the built-in power allocation algorithm to adjust non-critical treatment units. For example, the power output of the auxiliary pump is reduced from 80% to 60%. At the same time, the operating parameters of the core separation unit are fine-tuned to maintain its efficiency above the baseline value of 3,800 liters of wastewater per hour.

[0198] The system first extracted energy consumption distribution data from similar operating conditions over the past 30 days from a historical operation database. It calculated that reducing the auxiliary pump power by 10% would reduce energy consumption by approximately 8.5 kWh / h. A linear regression algorithm was then used to analyze the impact of power adjustment on overall system stability, concluding that a core separation efficiency fluctuation range of 3,750 to 3,850 liters / h was acceptable.

[0199] Next, the system performs a secondary check based on the current ambient humidity data (e.g., relative humidity of 65%). Using a correlation model between humidity and device heat dissipation efficiency, the system calculates that the device temperature rise after power reduction will not exceed 2.3 degrees Celsius, ensuring hardware safety.

[0200] On this basis, the system automatically determines the new energy balance point data, for example, setting the total energy consumption target at 95 kWh per hour, and uses the dynamic monitoring module to compare the deviation between actual energy consumption and the target value in real time. If the deviation exceeds 5%, the backup adjustment mechanism is triggered, automatically adjusting the auxiliary pump power to 65% to balance the system load;

[0201] All adjustment data is uploaded to the cloud for backup, forming a complete logical chain from evaluation to adjustment to monitoring, ensuring operational stability in energy-saving mode.

[0202] Furthermore, obtaining the device status update information after the cleaning is completed includes:

[0203] By combining energy balance data with the monitoring results of treatment efficiency and oil content, pre-established detection tools are used to analyze the changing characteristics of the equipment's operating status and determine whether the triggering conditions for the self-cleaning program are met;

[0204] If the detected change characteristics show that the processing efficiency has decreased and the oil content is stable, a command is sent through the control module to automatically trigger the equipment cleaning process and obtain the start status of the cleaning task;

[0205] According to the start status of the cleaning task, a timed flushing mechanism is used to treat the sediment on the inner wall of the pipeline and obtain the pipeline cleaning data after flushing;

[0206] By combining the pipeline cleaning data after flushing with the real-time feedback of equipment cleaning, the status update tool is called to analyze the impact of cleaning on equipment operation and determine the updated information of equipment status;

[0207] Based on updated information about device status, a support vector machine algorithm is used to perform pattern analysis on the associated data of energy balance and processing efficiency to obtain stability assessment results of the operating status.

[0208] If the stability assessment results show deviations, the cleaning frequency is adjusted through the control module, and combined with the information acquisition tool, the equipment operating parameters are continuously tracked to obtain the adjusted operating data;

[0209] Based on the adjusted operating data, the feedback mechanism of the self-cleaning program is linked to analyze the impact of long-term operation on the inner wall of the pipeline and determine the optimization plan for subsequent cleaning cycles;

[0210] Specifically, based on the energy balance point data, the system automatically links the self-cleaning program control module. By monitoring the equipment's operating status in real time, it triggers the equipment's self-cleaning process when it detects that the treatment efficiency has dropped below 3,200 liters of wastewater per hour and the oil content is stable at less than 5.2 mg / L.

[0211] First, the system uses sensors to collect data on the thickness of sediment on the inner wall of the pipe. If the current thickness reaches 3.5 mm, which exceeds the preset threshold of 2.8 mm, it determines that cleaning is needed;

[0212] Next, the system calls the cleaning scheduling algorithm, combines historical cleaning records to analyze the accumulation rate of pipeline sediment in the past seven days, and calculates the optimal cleaning time to be 25 minutes. The flushing water pressure is set to 4.5 bar to ensure a removal efficiency of more than 90%.

[0213] The system then automatically starts the high-pressure water flushing process, regularly cleaning the inner wall of the pipe in sections. At the same time, the flow sensor monitors the amount of water used in real time, recording a water consumption of approximately 180 liters per minute to avoid wasting resources.

[0214] During the cleaning process, the system uses a neural network prediction model to analyze the impact of cleaning on the equipment's vibration frequency and concludes that vibration frequency fluctuations within ±0.3 Hz are within a safe range.

[0215] After cleaning is complete, the system automatically updates the equipment status, detecting, for example, that the thickness of the sediment on the inner wall of the pipe has been reduced to 0.7 mm, restoring the treatment efficiency to over 3,500 liters per hour, and updating the oil content data to 4.8 mg / L.

[0216] All data is uploaded to the central database via the IoT interface, generating a comparison report of the status before and after cleaning. If the efficiency recovery value does not meet the standard, the system automatically triggers the secondary cleaning logic and adjusts the water pressure to 5.0 bar for additional cleaning to ensure the stability of equipment operation.

[0217] To form a complete logical chain, the system also links cleaning data with subsequent maintenance plans, automatically generates a forecast for the next cleaning in 15 days, and sends a reminder to the management platform 48 hours in advance to ensure long-term and efficient operation of the equipment.

[0218] Furthermore, potential improvements in determining the extension of equipment life include:

[0219] By using the updated information of the equipment status, the pre-established monitoring tool is called to continuously obtain the data of the operating cost control module and obtain the real-time change characteristics of the equipment status;

[0220] Based on the real-time changing characteristics of equipment status and the correlation between cleaning data and energy consumption data, comparative analysis tools are used to process the differences between energy consumption data after cleaning and historical records to determine preliminary cost savings results;

[0221] Based on the preliminary results of cost savings and combined with long-term data from historical records, the support vector machine algorithm is used to analyze the fluctuation trend of the savings rate and obtain quantitative indicators of the savings rate.

[0222] Based on the quantitative indicators of savings, information processing tools are used to extract relevant features of improvement space and determine the possibility of extending equipment life.

[0223] If the possibility of extending the equipment life is lower than the preset threshold, the operating parameters are adjusted through the control module, and the operating cost optimization plan is processed in combination with the feedback information of the cleaning data to obtain the adjusted operating parameters;

[0224] By adjusting the operating parameters, continuously track the changing trend of potential extension, use data collection tools to analyze the dynamic characteristics of the improvement space, and determine the final evaluation results of equipment life extension;

[0225] Based on the final evaluation results, the operating cost control module is linked to obtain the optimized operating strategy based on the long-term impact of cost savings and determine the continued stability of the equipment status.

[0226] Specifically, the system continuously monitors the operating cost control module through equipment status update information, automatically obtains energy consumption data after cleaning and compares and analyzes it with historical records to calculate cost savings and evaluate potential improvements to extend equipment life;

[0227] First, the system collects real-time energy consumption data after cleaning from the IoT interface. Assuming the current equipment consumes 12.5 kWh per hour, compared to the historical average of 14.8 kWh, the system calculates a 15.5% reduction in energy consumption. This data is then stored in the cost analysis database.

[0228] Next, the system uses a cost-savings assessment algorithm, combined with the energy consumption trend data from the past 30 days, to analyze that every kilowatt-hour of energy consumption reduction after cleaning can save approximately 3.2 yuan in operating costs. This extrapolates to a daily savings of approximately 17.3 yuan, and further predicts that the total annual savings could reach more than 6,300 yuan.

[0229] At the same time, the system uses an equipment life prediction model to analyze the impact of energy consumption reduction on equipment wear. Assuming the average life of core equipment components is 5.7 years, combined with current energy consumption optimization data, the system calculates the life extension potential to be 0.4 years and generates a life improvement report.

[0230] To form a complete logical chain, the system links cost-saving data with the equipment spare parts replacement cycle, automatically adjusts spare parts procurement plans, and predicts that the next replacement time can be postponed to 8 months. The relevant information is synchronized with the asset management system to ensure optimized resource allocation.

[0231] Through the above analysis, the system automatically generates a comprehensive cost and life assessment report and uploads it to the cloud platform to provide data support for subsequent equipment management strategy adjustments.

[0232] Example 3

[0233] like Figure 3 As shown, this embodiment provides an oily wastewater treatment equipment control system based on the Internet of Things, which mainly includes:

[0234] A real-time monitoring module is used to continuously collect the oil content of the inlet water through a sensor network. High-precision oil content sensors are deployed at key nodes of the inlet pipeline to obtain data streams every minute, generating a time series dataset containing oil content fluctuations and obtaining preliminary operational status information.

[0235] The data preprocessing module is used to perform denoising on the collected oil content fluctuation information based on the time series data set, filter out abnormal values ​​caused by environmental interference, calculate the short-term average value and change trend using the sliding window technology, and determine whether the oil content exceeds the preset threshold range;

[0236] The abnormality detection module is used to trigger the abnormality detection module if the oil content exceeds the preset threshold range. It combines historical operation data and current energy consumption indicators, classifies the degree of abnormality through a pre-established decision tree model, and determines whether the critical point of equipment overload risk has been reached;

[0237] The dynamic response mechanism module is used to activate the dynamic response mechanism module based on the abnormality classification results. In scenarios with high risk of equipment overload, the module automatically adjusts the operating parameters of the processing unit, such as increasing the separator speed or the dosage, and obtains feedback data on the adjusted operating status;

[0238] The multi-dimensional data analysis module is used to call the multi-dimensional data analysis module based on the operating status feedback data, perform correlation calculations on energy consumption, processing efficiency and adjustment parameters, and use the support vector machine algorithm to predict subsequent operating trends to obtain the feasibility assessment results of energy-saving mode switching;

[0239] An energy-saving mode switching module is used to activate the energy-saving mode switching module if the feasibility assessment result of the energy-saving mode switching shows that there is a large room for energy consumption optimization, reduce the power output of non-critical processing units, while maintaining the core separation efficiency, and determine the new energy consumption balance point data;

[0240] The self-cleaning program control module is used to link the self-cleaning program control module based on the energy consumption balance point data. When it detects that the processing efficiency has decreased and the oil content is stable, it automatically triggers the equipment self-cleaning process, flushes the sediment on the inner wall of the pipeline at a regular interval, and obtains the equipment status update information after the cleaning is completed;

[0241] The operating cost control module is used to continuously monitor the operating cost control module through equipment status update information, compare the energy consumption data after cleaning with historical records, calculate the cost savings, and determine the potential improvement space for extending the equipment life.

[0242] The above disclosure is only a preferred embodiment of the present invention, and it is certainly not intended to limit the scope of the present invention. A person skilled in the art can understand that all or part of the processes of the above embodiment and equivalent changes made in accordance with the claims of the present invention are still within the scope of the invention.

Claims

1. A method for controlling oily wastewater treatment equipment based on the Internet of Things, characterized in that: The method comprises: The oil content of the inlet water is continuously collected through a sensor network. High-precision oil content sensors are deployed at key nodes in the inlet pipeline. Data streams are acquired every minute to generate a time series dataset containing oil content fluctuations, providing preliminary operational status information. Based on the time series data set, the collected oil content fluctuation information is denoised to filter out outliers caused by environmental interference, and the short-term average and change trend are calculated to determine whether the oil content exceeds the preset threshold range; If the oil content exceeds the preset threshold, the abnormality detection module is triggered. Combining historical operating data and current energy consumption indicators, the pre-established decision tree model is used to classify the degree of abnormality and determine whether the critical point of equipment overload risk has been reached. Based on the abnormality degree classification results, the dynamic response mechanism module is activated to automatically adjust the operating parameters of the processing unit for scenarios with higher equipment overload risks, and obtain adjusted operating status feedback data.

2. The oily wastewater treatment equipment control method based on the Internet of Things according to claim 1 is characterized in that: The obtaining of preliminary operating status information includes: High-precision sensors are deployed at key nodes through the sensor network to collect influent oil content data at a frequency of minutes, build real-time data streams, and obtain initial time series information; Based on the initial time series information, the sliding window method is used to extract the characteristics of oil content fluctuation, obtain the fluctuation amplitude and periodic change characteristics, and determine the basic pattern of fluctuation; If the basic pattern of fluctuations exceeds the preset threshold range, the anomaly detection mechanism is triggered, and the data stream is classified using the pre-established support vector machine model to determine whether there is an abnormal state; According to the abnormal state judgment results, the local time series decomposition is performed on the abnormal data segment to obtain the trend component and noise component, and obtain the refined fluctuation composition information; By using the refined fluctuation composition information, the autoregressive model is used to make short-term predictions on the trend component, obtain the oil content change trend in the future period, and determine the prediction sequence; Based on the predicted sequence and combined with historical time series data, the operating status of key nodes is dynamically evaluated. If the deviation between the predicted sequence and historical data exceeds the preset range, the monitoring frequency is adjusted to obtain a more intensive data stream. Through denser data streams, the oil content fluctuation characteristics of the real-time data acquisition module are updated, the time series data set is reconstructed, and the latest operating status information is obtained.

3. The oily wastewater treatment equipment control method based on the Internet of Things according to claim 1 is characterized in that: Determining whether the oil content exceeds a preset threshold range includes: The time series data set is preliminarily cleaned through the data preprocessing module, and the outliers caused by environmental interference are filtered out using denoising technology to obtain the cleaned oil content fluctuation data; Based on the oil content fluctuation data after cleaning, the sliding window technology is used to analyze the local data segment, calculate the short-term average value, and obtain the stable characteristics within the local time period; Focusing on the stable characteristics within a local time period, the sliding window technique is used to extract dynamic information on the changing trend and determine the short-term variation pattern of oil content fluctuations. If the short-term change pattern shows that the oil content fluctuation exceeds the preset threshold range, the abnormal marking mechanism is triggered to distinguish the abnormal data segment from the normal data segment and obtain the marked data classification.

4. The method for controlling oily wastewater treatment equipment based on the Internet of Things according to claim 3 is characterized in that: Determining whether the oil content exceeds a preset threshold range further includes: According to the marked data classification, further data feature extraction is performed on the abnormal data segment to obtain the time point and amplitude information of the abnormal fluctuation and determine the specific distribution characteristics of the abnormality; Based on the specific distribution characteristics of the anomaly, combined with the pre-established support vector machine model, the abnormal data segments are classified and processed to determine whether the anomaly has persistent characteristics, and the classified anomaly assessment results are obtained; For the abnormal evaluation results after classification, if the persistence feature is confirmed, the frequency parameters of the data acquisition module are adjusted to obtain higher-density time series data and update the monitoring information of oil content fluctuations.

5. The oily wastewater treatment equipment control method based on the Internet of Things according to claim 1 is characterized in that: The determination of whether the critical point of equipment overload risk has been reached includes: When the oil content exceeds the threshold, the anomaly detection process is activated, and historical records and current energy consumption data are integrated through data fusion technology to obtain preliminary evidence for anomaly determination. Based on the preliminary abnormality judgment basis, a pre-built decision tree model is used to classify the abnormality level and determine the specific classification results of the abnormality level; Regarding the abnormal level classification results, if the classification results show that the equipment is close to the critical state of overload, the risk assessment mechanism is triggered, and the confirmation information of the risk critical state is obtained by comparing the real-time monitoring data of the operating status.

6. The method for controlling oily wastewater treatment equipment based on the Internet of Things according to claim 5, characterized in that: The determination of whether a critical point of equipment overload risk has been reached further includes: Based on the confirmed information of the critical risk state and the possibility of equipment overload, time series analysis tools are used to extract characteristics of historical fluctuations in operating status to obtain the distribution characteristics of potential overload hazards; Based on the distribution characteristics of potential overload hazards, pattern matching is performed with similar scenarios in historical records to determine whether there are signs of persistent risks; If signs of persistent risk are confirmed, adjust the data collection frequency parameters to obtain higher-density time series information and determine more accurate operating status monitoring data; Based on more accurate operating status monitoring data and combined with the dynamic changes in energy consumption data, a logical comparison method is used to conduct a secondary verification of the possibility of equipment overload to obtain the final risk assessment conclusion.

7. The method for controlling oily wastewater treatment equipment based on the Internet of Things according to claim 1, characterized in that: The obtaining of the adjusted operating status feedback data includes: Based on the classification results, the dynamic response module is activated to automatically identify operating states with high equipment overload risks in scenarios with high abnormality levels, and obtain preliminary risk assessment information. Based on the preliminary risk assessment information, the operating parameters of the processing unit are adjusted, the separator speed and dosage are dynamically configured, and the adjusted operating status data is obtained; Use the adjusted operating status data in combination with the pre-established decision tree model to analyze the possibility of equipment overload and determine the distribution of potential operating hazards; If the distribution of potential operational hazards exceeds the preset threshold range, the parameter optimization process is initiated to obtain the optimized parameter configuration plan through real-time comparison of the operational status data; According to the optimized parameter configuration scheme, the operating parameters of the processing unit are readjusted, the separator speed and the dosage are calibrated again, and the updated operating status information is obtained.

8. The method for controlling oily wastewater treatment equipment based on the Internet of Things according to claim 1, characterized in that: The obtaining of the adjusted operating status feedback data further includes: Continuously monitor scenarios with high risk of equipment overload using updated operating status information. If high-risk situations persist, save current operating status data using data logging tools to determine whether there are signs of long-term potential hazards. If signs of long-term hidden dangers are confirmed, the data collection frequency will be adjusted to obtain higher-density operating status data and determine more accurate equipment operation monitoring information.

9. The method for controlling oily wastewater treatment equipment based on the Internet of Things according to claim 1, characterized in that: Also includes: By using the operating status feedback data, the multi-dimensional data analysis module is called to correlate energy consumption, processing efficiency and adjustment parameters. The support vector machine algorithm is used to predict the subsequent operating trend and obtain the feasibility assessment results of energy-saving mode switching. If the feasibility assessment results of energy-saving mode switching show that there is a large room for energy consumption optimization, the energy-saving mode switching module is activated to reduce the power output of non-critical processing units while maintaining the core separation efficiency and determine the new energy consumption balance point data; Based on the energy balance point data, the self-cleaning program control module is linked to automatically trigger the equipment self-cleaning process when it detects a decrease in treatment efficiency and a stable oil content. The equipment automatically flushes the sediment on the inner wall of the pipe at regular intervals and obtains equipment status updates after the cleaning is completed. By updating the equipment status information, the operating cost control module is continuously monitored, and the energy consumption data after cleaning is compared with the historical records to calculate the cost savings and determine the potential room for improvement in extending the equipment life.

10. An oily wastewater treatment equipment control system based on the Internet of Things, characterized in that: The system comprises: A real-time monitoring module is used to continuously collect the oil content of the inlet water through a sensor network. High-precision oil content sensors are deployed at key nodes of the inlet pipeline to obtain data streams every minute, generating a time series dataset containing oil content fluctuations and obtaining preliminary operational status information. The data preprocessing module is used to perform denoising on the collected oil content fluctuation information based on the time series data set, filter out abnormal values ​​caused by environmental interference, calculate the short-term average value and change trend using the sliding window technology, and determine whether the oil content exceeds the preset threshold range; The abnormality detection module is used to trigger the abnormality detection module if the oil content exceeds the preset threshold range. It combines historical operation data and current energy consumption indicators, classifies the degree of abnormality through a pre-established decision tree model, and determines whether the critical point of equipment overload risk has been reached; The dynamic response mechanism module is used to activate the dynamic response mechanism module based on the abnormality classification results. In scenarios with high risk of equipment overload, the module automatically adjusts the operating parameters of the processing unit, such as increasing the separator speed or the dosage, and obtains feedback data on the adjusted operating status; The multi-dimensional data analysis module is used to call the multi-dimensional data analysis module based on the operating status feedback data, perform correlation calculations on energy consumption, processing efficiency and adjustment parameters, and use the support vector machine algorithm to predict subsequent operating trends to obtain the feasibility assessment results of energy-saving mode switching; An energy-saving mode switching module is used to activate the energy-saving mode switching module if the feasibility assessment result of the energy-saving mode switching shows that there is a large room for energy consumption optimization, reduce the power output of non-critical processing units, while maintaining the core separation efficiency, and determine the new energy consumption balance point data; The self-cleaning program control module is used to link the self-cleaning program control module based on the energy consumption balance point data. When it detects that the processing efficiency has decreased and the oil content is stable, it automatically triggers the equipment self-cleaning process, flushes the sediment on the inner wall of the pipeline at a regular interval, and obtains the equipment status update information after the cleaning is completed; The operating cost control module is used to continuously monitor the operating cost control module through equipment status update information, compare the energy consumption data after cleaning with historical records, calculate the cost savings, and determine the potential improvement space for extending the equipment life.

Images