Multi-source data fusion and remote diagnosis method and system for sewage treatment equipment

Abstract

The present application relates to the technical field of operation and maintenance of sewage treatment equipment, and provides a multi-source data fusion and remote diagnosis method and system for sewage treatment equipment. The method includes: when it is monitored that there are risk components in each target component of the sewage treatment equipment, obtaining abnormal verification data corresponding to the risk component; verifying whether there is an abnormality in the risk component based on the abnormal verification data and the risk verification rules corresponding to the target component, and obtaining an abnormal verification result; when the abnormal verification result indicates that there is an abnormality in the risk component, determining the abnormal handling method for the risk equipment based on the abnormal operation data and abnormal verification data of the component operation data corresponding to the risk component; generating abnormal prompt information based on the abnormal handling method. By adopting the above technical solution, a multi-dimensional verification mechanism for abnormal status can be established, thereby reducing the risk of misjudgment of a single data source and ensuring the reliability of the fault diagnosis conclusion.

Description

Technical Field

[0001] The present application relates to the technical field of operation and maintenance of sewage treatment equipment, and in particular to a multi-source data fusion and remote diagnosis method and system for sewage treatment equipment. Background Art

[0002] The operation and maintenance of sewage treatment equipment is crucial for ensuring effective sewage treatment and environmental protection. With the development of the sewage treatment industry, equipment types are becoming increasingly diverse, and the scale of operations is expanding. The stable operation of these equipment is directly related to effluent water quality and environmental compliance. However, as typical electromechanical intelligent equipment, sewage treatment equipment involves multiple components, including sensors, controllers, and inverters, during actual operation. Its operating status is significantly affected by changes in the environment, water quality, and operating conditions. Therefore, achieving efficient monitoring and intelligent diagnosis of these equipment has become a key research direction in the industry.

[0003] In related technologies, the operation and management of sewage treatment equipment usually adopt the following means: manual inspection combined with on-site maintenance, completing inspections by regularly observing the equipment's operating status and reading instrument data; local control of the PLC / SCADA system to achieve automatic control and local data collection of some key equipment; a single-point alarm system based on local data collection to monitor and alarm indicators such as pump body temperature and current; and data assistance functions of management systems such as OA / ERP for archiving and statistical analysis of equipment operation records.

[0004] Although the technical means in related technologies can solve the problem of equipment management to a certain extent, the data sources are fragmented and centralized and unified management is not achieved. This results in monitoring only being able to be achieved based on simple rules during the monitoring process, which greatly affects the accuracy of equipment monitoring and therefore needs to be improved. Summary of the Invention

[0005] In order to help improve the accuracy of equipment monitoring, the present application provides a multi-source data fusion and remote diagnosis method and system for sewage treatment equipment.

[0006] In the first aspect, the present application provides a multi-source data fusion and remote diagnosis method for sewage treatment equipment, which adopts the following technical solutions:

[0007] A multi-source data fusion and remote diagnosis method for sewage treatment equipment, the method comprising:

[0008] For each target component in the sewage treatment equipment, risk monitoring is performed on the target component based on the component operation data of the target component and the risk monitoring rules corresponding to the target component;

[0009] When a risk component is detected, obtaining abnormality verification data corresponding to the risk component;

[0010] Verifying whether the risk component has an abnormality based on the abnormality verification data and the risk verification rule corresponding to the target component, and obtaining an abnormality verification result;

[0011] When the abnormality check result indicates that the risk component has an abnormality, determining an abnormality handling method for the risk device based on abnormal operation data with an abnormality in the component operation data corresponding to the risk component and the abnormality check data;

[0012] Generate exception prompt information based on the exception handling method.

[0013] By adopting the above technical solution, a multi-dimensional verification mechanism for abnormal conditions can be established, thereby reducing the risk of misjudgment from a single data source and ensuring the reliability of fault diagnosis conclusions.

[0014] Optionally, when a risk component is detected, obtaining abnormality verification data corresponding to the risk component includes:

[0015] In the case where the risk component is monitored, determining a risk type corresponding to the risk component based on the abnormal operation data;

[0016] Determining a verification data acquisition method based on the risk type;

[0017] In a case where the verification data acquisition method includes self-data, determining, among the component operation data corresponding to the risk component, associated operation data that has an associated relationship with the abnormal operation data as the abnormal verification data;

[0018] In a case where the verification data acquisition method includes associated component data, the abnormal verification data is determined from component operation data of associated components corresponding to the risk component.

[0019] By adopting the above technical solution, dynamic adaptation of the verification data acquisition method can be achieved, and the optimal data source can be automatically selected for different risk types, effectively reducing the processing volume of irrelevant data. At the same time, the accuracy of anomaly judgment is improved through multi-dimensional data cross-validation.

[0020] Optionally, determining the abnormality verification data from component operation data of associated components corresponding to the risk component includes:

[0021] In a case where the risk components include more than two, determining whether the associated component is a risk component;

[0022] In a case where the associated component is a risk component, determining abnormal device operation data in the operation data corresponding to the associated component as the abnormal verification data;

[0023] If the associated component is not the risk component, determining whether the associated component and the risk component are of the same type;

[0024] In a case where the associated component and the risk component are of the same type, determining data of the same type as the abnormal operation data in the component operation data corresponding to the associated component as the abnormality verification data;

[0025] In a case where the associated component and the risk component are of different types, abnormality verification data is determined from component operation data corresponding to the associated component based on the risk type.

[0026] By adopting the above technical solution, a three-level screening mechanism of risk status judgment, type matching, and risk type adaptation can be established, which will help solve the problem that traditional methods often adopt a unified data collection strategy when dealing with multiple risk components, resulting in redundant verification data or omission of key related information, and thus effectively improve the pertinence and reliability of abnormal verification data.

[0027] Optionally, the abnormality verification data includes numerical verification data represented by numerical values, and the risk verification rule includes a change matching rule, wherein the change matching rule is used to record the correspondence between the change of the numerical verification data determined when the risk component is abnormal and the change of the abnormality verification data. The abnormality verification data and the risk verification rule corresponding to the target component are used to verify whether the risk component is abnormal, and obtain an abnormality verification result, including:

[0028] Monitoring the changes in the numerical verification data and the abnormal verification data;

[0029] Determining whether a change in the numerical verification data matches a change in the abnormal verification data based on the change matching rule;

[0030] In a case where the change in the verification data matches the change in the verification data, an abnormality verification result is generated indicating that an abnormality exists in the risk component.

[0031] By adopting this technical solution, anomaly verification can be performed by matching changes in numerical verification data with changes in anomaly verification data, effectively identifying potential faults in risky components. Compared to traditional single-threshold judgment methods, this embodiment significantly improves the accuracy and reliability of anomaly verification results by introducing change matching rules and combining dynamic data features for comprehensive analysis, thus providing strong support for the stable operation of the equipment.

[0032] Optionally, the risk verification rule includes a verification range corresponding to the numerical verification data, where the verification range is determined when the risk component is abnormal. After determining whether a change in the numerical verification data matches a change in the abnormal verification data, the method further includes:

[0033] In the event that the change in the verification data does not match the change in the verification data, determining whether the numerical verification data falls within a corresponding verification range;

[0034] In a case where the numerical verification data includes more than two and some of the numerical verification data do not fall within the corresponding verification range, determining a verification adjustment parameter based on the number of pending verification data that do not fall within the corresponding verification range;

[0035] Adjusting the verification range corresponding to the pending verification data based on the verification adjustment parameter to obtain an adjusted range, wherein the adjusted range is larger than the verification range;

[0036] Determining whether the pending verification data falls within the corresponding adjusted range;

[0037] When each of the pending verification data falls within the corresponding adjusted range, an abnormality verification result indicating that the risk component is abnormal is generated.

[0038] By employing this technical solution, we can verify changes in numerical verification data at multiple levels, not only considering the matching of change trends but also introducing a dynamic adjustment mechanism for the verification range. In particular, when there are partial deviations in multiple sets of numerical verification data, adjusting the verification range ensures the comprehensiveness and rationality of abnormal verification results, thus providing a more scientific basis for the condition assessment of risky components.

[0039] Optionally, the abnormality verification data includes operation and maintenance data of the risk component, and before monitoring the change of the numerical verification data and the change of the abnormality verification data, the method further includes:

[0040] Analyzing the operation and maintenance data to obtain risk prediction data corresponding to the risk components;

[0041] determining whether there is the risk prediction data matching the abnormal operation data;

[0042] In the absence of the risk prediction data matching the abnormal operation data, the step of determining whether the numerical verification data falls within a verification range corresponding to the risk data is performed.

[0043] By adopting the above technical solution, before performing abnormal verification on risk components in combination with abnormal verification data, the risk components can be verified for abnormalities in combination with historical operation and maintenance data. This not only improves the accuracy of fault diagnosis, but also reduces the false alarm rate, thereby improving the reliability and efficiency of the entire sewage treatment equipment remote diagnosis system.

[0044] Optionally, the risk component includes a water pump, the numerical verification data includes vibration data of the water pump, and the abnormality verification data includes current data and outlet pressure data of the water pump. Determining whether a change in the numerical verification data matches a change in the abnormality verification data based on the change matching rule includes:

[0045] In the case where the vibration data continuously rises within a preset time period, determining whether the fluctuation amplitude of the current data reaches a preset fluctuation amplitude threshold, and whether the outlet pressure data decreases;

[0046] When the fluctuation amplitude reaches the fluctuation amplitude threshold and the outlet pressure data decreases, it is determined that the change of the numerical verification data matches the change of the abnormal verification data.

[0047] By adopting the above technical solution, when the vibration data of the water pump continues to rise within a certain period of time, a comprehensive analysis can be further performed in combination with the fluctuation amplitude of the current data and the change trend of the outlet pressure data, thereby improving the accuracy of water pump abnormality judgment.

[0048] Optionally, the determining of an abnormality handling method for the risky device based on abnormal operation data and the abnormality verification data in the component operation data corresponding to the risky component includes:

[0049] Inputting the abnormal operation data and the abnormal verification data into a pre-trained expert model to obtain the abnormality handling method;

[0050] Wherein, the expert model is constructed based on training using an expert rule base;

[0051] After generating the exception prompt information based on the exception handling method, the method further includes:

[0052] Obtaining maintenance records for the sewage treatment equipment;

[0053] Operation and maintenance data are extracted from the maintenance records, and an expert rule base is updated based on the operation and maintenance data.

[0054] By adopting the above technical solution, it is possible to automatically determine the precise treatment method for abnormal situations of sewage treatment equipment, and optimize the treatment strategy through continuous learning, thereby providing accurate reference for the work of operation and maintenance personnel.

[0055] Optionally, when the abnormality check result indicates that the risk component is abnormal, the method further includes:

[0056] Performing abnormality prediction on the sewage treatment equipment based on the abnormal operation data to obtain an abnormality prediction result;

[0057] updating the device status of associated equipment based on the abnormality prediction result and the position of the sewage treatment equipment in the sewage treatment process;

[0058] The sewage treatment effect is predicted based on the equipment status of each preset reference equipment in the sewage treatment process to obtain a predicted treatment result.

[0059] By adopting the above technical solution, when an anomaly is detected, the impact of the anomaly on the sewage treatment effect can be accurately judged in combination with the sewage treatment process, which can help assist operation and maintenance personnel in making operation and maintenance decisions.

[0060] In a second aspect, the present application provides a multi-source data fusion and remote diagnosis system for sewage treatment equipment, which adopts the following technical solutions:

[0061] A multi-source data fusion and remote diagnosis system for sewage treatment equipment, the system comprising a data acquisition component and a cloud server in communication with the data acquisition component;

[0062] The data acquisition component is used to acquire component operation data of each component in the sewage treatment equipment and send it to the cloud server;

[0063] A cloud operation and maintenance platform runs on the cloud server, and the cloud operation and maintenance platform is used to implement the multi-source data fusion and remote diagnosis method of any sewage treatment equipment provided in the first aspect.

[0064] In summary, this application includes at least one of the following beneficial technical effects:

[0065] 1. Establish a multi-dimensional verification mechanism for abnormal conditions, reducing the risk of misjudgment from a single data source and ensuring the reliability of fault diagnosis conclusions.

[0066] 2. Determine reasonable exception handling methods based on abnormal operation data and verification data and generate prompt information, which helps operation and maintenance personnel take quick measures to reduce the impact of equipment failures and ensure the stable operation and treatment efficiency of the sewage treatment system. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 This is a flow chart of a multi-source data fusion and remote diagnosis method for sewage treatment equipment provided in an embodiment of the present application;

[0068] Figure 2 This is a flow chart of a method for obtaining verification data provided in an embodiment of the present application;

[0069] Figure 3 This is a flow chart of a method for determining abnormal verification data provided by an embodiment of the present application;

[0070] Figure 4 This is a flow chart of a method for determining abnormal verification results provided by an embodiment of the present application;

[0071] Figure 5 This is a flow chart of a method for determining an exception handling method provided in an embodiment of the present application;

[0072] Figure 6 This is a flow chart of a sewage treatment effect update method provided in an embodiment of the present application;

[0073] Figure 7 This is a structural diagram of a multi-source data fusion and remote diagnosis system for sewage treatment equipment provided in an embodiment of the present application. DETAILED DESCRIPTION

[0074] In order to make the purpose, technical solutions and advantages of this application more clear, the following Figure 1-7 It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application.

[0075] In related technologies, the operation and maintenance of sewage treatment equipment primarily relies on manual inspections and distributed data acquisition systems, with equipment operating data dispersed across multiple isolated systems such as PLCs, SCADA, and independent sensors. A sewage treatment plant is equipped with over ten types of equipment, including blowers, water pumps, and dosing devices. The operating data for each type of equipment is stored in local databases in different control cabinets. When the dissolved oxygen concentration in the aeration tank is abnormal, operators must sequentially retrieve data from multiple platforms, including fan frequency, pipeline pressure, and current fluctuations. They then manually compare historical operating curves to determine the cause of the abnormality, causing fault diagnosis to take several hours and the risk of misjudgment due to data update delays.

[0076] To address these issues, analysis revealed that data silos lead to inefficient anomaly diagnosis, and that single sensor data cannot accurately reflect the true status of the equipment. Research into equipment failure mechanisms revealed that electromechanical equipment anomalies are often accompanied by linked changes in multiple parameters. For example, bearing wear in a water pump can simultaneously cause increased vibration amplitude and current harmonic distortion. Based on this, a multi-source data fusion and remote diagnosis method and system for sewage treatment equipment were proposed to address these issues.

[0077] The embodiment of the present application discloses a multi-source data fusion and remote diagnosis method for sewage treatment equipment. Figure 1 ,The multi-source data fusion and remote diagnosis method of sewage treatment equipment includes the following steps:

[0078] Step 101 : For each target component in the sewage treatment equipment, risk monitoring is performed on the target component based on the component operation data of the target component and the risk monitoring rules corresponding to the target component.

[0079] The target component refers to an electromechanical unit with independent functions in the sewage treatment process. In actual implementation, the target component includes at least one of the core components such as the blower impeller, water pump bearing, and dosing pump plunger.

[0080] Component operating data reflects the real-time characteristics of the equipment's operating status. Specifically, component operating data can be collected by sensors installed on the component, or it can be component operating parameters. The type of component operating data and the method of obtaining it may vary depending on the component. In actual implementation, component operating data may include at least one of the following parameters: vibration acceleration, winding temperature, operating current, etc.

[0081] Risk monitoring rules refer to pre-set logic for determining component health status. These can be implemented using algorithms such as threshold comparison, trend slope calculation, and waveform spectrum analysis to automatically identify components that may pose risks.

[0082] Step 102: When a risk component is detected, abnormality verification data corresponding to the risk component is obtained.

[0083] Abnormality verification data refers to auxiliary monitoring data that is physically or logically associated with the target component and is used to cross-verify whether the risk component has an abnormality. In actual implementation, abnormality verification data can be correlated parameters of different sensors on the same device, or it can be operating data of components of equipment that has a relationship with the target device (for example, upstream and downstream process equipment).

[0084] Step 103 : Verify whether the risk component has any abnormality based on the abnormality verification data and the risk verification rule corresponding to the target component, and obtain an abnormality verification result.

[0085] The risk verification rules are used to verify whether the risky components have abnormalities. The risk verification rules are pre-set. In one example, the risk verification rules are set in the same way as the risk monitoring rules, but they are set for different types of data.

[0086] Step 104 : When the abnormality check result indicates that the risk component has an abnormality, determine an abnormality handling method for the risk device based on the abnormal operation data and abnormality check data in the component operation data corresponding to the risk component.

[0087] Among them, the exception handling method refers to the disposal strategy for verified abnormal conditions, which may include parameter adjustment (such as frequency adjustment), equipment switching, manual maintenance, etc.

[0088] In one example, the most suitable exception handling method can also be found from an exception handling database corresponding to the risk component based on the abnormal operation data and the abnormal verification data. The exception handling database is generated based on historical maintenance records of components of the same type.

[0089] In another example, a pre-set relationship is established between the abnormal operation data and abnormal verification data and different exception handling methods. For example, different exception handling methods correspond to different data ranges. Thus, the corresponding exception handling method can be determined based on the matching relationship between the abnormal operation data and abnormal verification data and the data range. Furthermore, if the abnormal operation data and abnormal verification data correspond to different exception handling methods, both exception handling methods can be used simultaneously, which helps ensure the accuracy of the exception handling methods.

[0090] In actual implementation, the exception handling methods corresponding to different risk components may be different. In this case, the exception handling method used this time can be selected from all exception handling methods corresponding to the risk component based on the abnormal operation data and the abnormal verification data.

[0091] Step 105: Generate exception prompt information based on the exception handling method.

[0092] The exception prompt information is used to remind workers to handle equipment anomalies. In one example, the exception prompt information corresponds to the exception handling method. For example, if the exception handling method includes parameter adjustment, the exception prompt information may include generating parameter adjustment suggestions to prompt workers to adjust equipment parameters. For another example, if the exception handling method includes equipment switching, the exception prompt information may include generating equipment switching suggestions to prompt workers to switch equipment.

[0093] Furthermore, the system can automatically execute exception handling steps based on the exception handling method. For example, when the exception handling method includes adjusting the frequency, the system can generate a frequency adjustment instruction based on the actual situation or with a fixed adjustment range to adjust the operating frequency of the equipment. For example, when the exception handling method includes manual maintenance, the system can automatically generate and distribute an operation and maintenance work order to improve the efficiency of manual maintenance.

[0094] In order to more clearly understand the technical solution provided by this embodiment, an example is used to illustrate the following. For example: when the vibration value of the blower bearing exceeds the preset threshold, the risk monitoring rule automatically triggers abnormal monitoring. The system synchronously retrieves the three-phase current harmonic data of the motor corresponding to the bearing as abnormality verification data, and analyzes the current spectrum characteristics through Fourier transform. If the amplitude of the 2 times the rotation frequency component in the current spectrum increases synchronously, the risk verification rule determines that the mechanical looseness anomaly is established. At this time, based on the abnormal operation data and the abnormal verification data, the same type of fault handling solution is found from the historical maintenance records, and a maintenance work order containing the bearing tightening torque parameters and lubricant model is generated, and automatically pushed to the mobile terminal of the operation and maintenance personnel.

[0095] The implementation principle of a multi-source data fusion and remote diagnosis method for sewage treatment equipment in an embodiment of the present application is as follows: for each target component in the sewage treatment equipment, risk monitoring of the target component is performed based on the component operation data of the target component and the risk monitoring rules corresponding to the target component; when a risk component is monitored, abnormality verification data corresponding to the risk component is obtained; based on the abnormality verification data and the risk verification rules corresponding to the target component, whether the risk component has an abnormality is verified to obtain an abnormality verification result; when the abnormality verification result indicates that the risk component has an abnormality, the abnormal operation data and abnormality verification data corresponding to the component operation data of the risk component are determined to determine the abnormality handling method for the risk equipment; and abnormality prompt information is generated based on the abnormality handling method. By adopting the above technical solution, a multi-dimensional verification mechanism for abnormal states can be established, thereby reducing the risk of misjudgment of a single data source and ensuring the reliability of the fault diagnosis conclusion.

[0096] Furthermore, determining a reasonable abnormality handling method and generating prompt information based on abnormal operation data and verification data will help operation and maintenance personnel take quick measures to reduce the impact of equipment failures and ensure the stable operation and treatment efficiency of the sewage treatment system.

[0097] In some embodiments, reference Figure 2 Step 102, when a risk component is detected, obtains abnormality verification data corresponding to the risk component, including the following steps:

[0098] Step 201: When a risk component is detected, the risk type corresponding to the risk component is determined based on abnormal operation data.

[0099] Risk types refer to potential equipment failure categories identified by analyzing abnormal operating data and are used to guide the selection of verification data. In practice, risk types can be implemented based on fault signature library matching or pattern recognition algorithms.

[0100] In an example, taking a pump as an example, if the abnormal operation data shows that the temperature of the pump body increases abnormally, it may be classified as an overheating risk type.

[0101] Step 202: Determine a verification data acquisition method based on the risk type.

[0102] The verification data acquisition method refers to a strategy for dynamically selecting the data source of abnormal verification data based on the risk type. Specifically, the data acquisition method may include: own data and / or associated component data.

[0103] In this embodiment, the correspondence between the risk type and the verification data acquisition method is pre-set, which can be implemented through a pre-defined rule table or a decision tree model.

[0104] Step 203 : When the verification data acquisition method includes self-data, the associated operation data in the component operation data corresponding to the risk component that is associated with the abnormal operation data is determined as abnormal verification data.

[0105] The association relationship between the component operation data is pre-set. In one example, the association relationship is determined based on the logical association between the component operation data, and can be determined specifically through data correlation analysis.

[0106] In one example, continuing to use the risk component as an example of a pump, if the verification data acquisition method includes its own data, at least one of the parameters such as vibration frequency, current fluctuation, etc. that are related to temperature is screened out from the operating data of the pump and determined as abnormal verification data.

[0107] Step 204 : When the verification data acquisition method includes associated component data, abnormal verification data is determined from component operation data of associated components corresponding to the risk component.

[0108] The association relationship between the components is pre-set. In one example, the association relationship is set based on the structural association between the components in the device, and can be specifically determined by the structural topology relationship of the device.

[0109] In one example, continuing to use the risk component as an example of a pump, if the verification data acquisition method includes associated component data, the pipeline connected to the pump will be determined as an associated component, and at least one of the component operation data such as pipeline pressure, flow, etc. in the component operation data corresponding to the pipeline will be determined as abnormal verification data.

[0110] In actual implementation, according to actual needs, the verification data acquisition method can include both its own data and related component data. For example, in a motor overload risk scenario, by simultaneously acquiring its own current data and the gear wear data of the related reducer, it is possible to quickly distinguish between abnormal phenomena caused by equipment overload and mechanical wear, and avoid misjudgment.

[0111] By adopting the above technical solution, dynamic adaptation of the verification data acquisition method can be achieved, and the optimal data source can be automatically selected for different risk types, effectively reducing the processing volume of irrelevant data. At the same time, the accuracy of anomaly judgment is improved through multi-dimensional data cross-validation.

[0112] Specifically, when using its own data as verification data, it can effectively leverage related operating data associated with abnormal operating data for cross-validation, improving the accuracy of abnormality judgment. When using related component data as verification data, it can mine valuable information from the operating data of related components, further enhancing the comprehensiveness and reliability of anomaly detection. This allows for refined verification of abnormal conditions in risky components, improving the efficiency and accuracy of fault diagnosis.

[0113] Based on the above technical solution, refer to Figure 3 In the above step 204, determining abnormal verification data from the component operation data of the associated components corresponding to the risk component includes the following steps:

[0114] Step 301: When there are two or more risk components, determine whether the associated component is a risk component.

[0115] Step 302: When the associated component is a risk component, the device operation data with abnormalities in the operation data corresponding to the associated component is determined as abnormal verification data.

[0116] Specifically, when the associated component is a risk component, it also has abnormal operating data. At this time, the abnormal operating data of the associated component is determined as abnormal verification data. During the verification process, the abnormal data corresponding to the associated risk component can be comprehensively analyzed, which can help improve the accuracy of abnormal judgment.

[0117] Step 303: If the associated component is not a risk component, determine whether the associated component and the risk component are of the same type.

[0118] The classification method of the component types is pre-set. For example, if the component types include pumps, then when the associated component and the risk component are both pumps, the associated component and the risk component are determined to be of the same type.

[0119] Step 304 : When the associated component and the risk component are of the same type, data of the component operation data corresponding to the associated component that is of the same type as the abnormal operation data is determined as abnormal verification data.

[0120] Specifically, since the risk component is obtained by monitoring the component operation data, and when the associated component is of the same type as the risk component, it means that the associated component also has component operation data of the same type as the abnormal verification data. At this time, determining the component operation data as verification data can help avoid the impact of fluctuations in the component operation data on component abnormality judgment, and thus realize the verification of component abnormalities.

[0121] Step 305 : When the types of the associated component and the risk component are different, determine abnormality verification data from the component operation data corresponding to the associated component based on the risk type.

[0122] The correspondence between the risk type and the component operation data type is pre-set. For example, for a water pump, the risk type is current overload, and the associated component is the pipe connected to the water pump. In this case, the abnormal verification data can be the pressure data of the pipe.

[0123] In the above technical solution, by establishing a three-level screening mechanism of risk status judgment, type matching, and risk type adaptation, it can help solve the problem that traditional methods often adopt a unified data collection strategy when dealing with multiple risk components, resulting in redundant verification data or missing key related information, and thus effectively improve the pertinence and reliability of abnormal verification data.

[0124] Specifically, since the operating data of marked abnormal components are preferentially used as the verification benchmark in the process of determining abnormal verification data, the waste of resources caused by repeatedly collecting data of normal associated components is avoided; the consistency of the verification data and the operating characteristics of the target components is ensured by matching the equipment types, such as using the flow data of the same type of water pumps as the comparison benchmark; verification parameters are dynamically selected for different types of associated components, such as establishing a verification relationship between the water pump with abnormal vibration data and the pressure change data of the associated pipeline, thereby enhancing the effectiveness of data association between different types of equipment.

[0125] In some implementations, the anomaly verification data includes numerical verification data represented by numerical values, and the risk verification rule includes a change matching rule.

[0126] The change matching rule records the correspondence between changes in numerical verification data determined when a risk component anomaly occurs and changes in abnormal verification data, thereby capturing the abnormal characteristics of coordinated changes in multiple parameters. In actual implementation, the change matching relationship can be obtained through experiments or technical experience. In one example, the change matching rule is implemented using a correlation matrix of the change trends of different parameters in historical failure cases.

[0127] In actual implementation, the numerical verification data may include at least one of quantifiable data such as vibration frequency, current value, pressure value, etc.

[0128] Accordingly, reference Figure 4 In step 103, based on the abnormality verification data and the risk verification rules corresponding to the target component, the abnormality verification result is obtained by verifying whether the risk component has abnormalities.

[0129] Step 401: monitor the changes in the numerical verification data and the changes in the abnormal verification data.

[0130] The change situation may be a data representation of the rise, fall, fluctuation range, etc. extracted by a time series analysis method, which can reflect the change situation of the data over time.

[0131] In one example, the target component includes a water pump as an example. When the vibration data of the water pump continuously increases within a preset period of time, the water pump is determined to be a risk component, the current data and outlet pressure data of the water pump are determined to be abnormal verification data, and the changes in the current data and outlet pressure data of the water pump are monitored.

[0132] Vibration data refers to the physical parameters generated by the mechanical vibrations of a water pump during operation, reflecting the operational status of the pump's mechanical structure. In practice, vibration data can be collected using an acceleration sensor to collect the vibration frequency and amplitude of the pump's surface.

[0133] Current data refers to the current value driving the water pump motor and is used to represent changes in motor load. In practice, current data can be collected in real time using a frequency converter or current transformer.

[0134] Outlet pressure data refers to the fluid pressure parameter at the pump's output end, used to monitor the pump's delivery efficiency. In practice, this can be measured using a pressure transmitter.

[0135] Furthermore, the abnormal verification data includes operation and maintenance data of the risk component. Before monitoring the changes in the verification data and the changes in the abnormal verification data, it also includes: parsing the operation and maintenance data to obtain risk prediction data corresponding to the risk component; determining whether there is risk prediction data that matches the abnormal operation data; in the absence of risk prediction data that matches the abnormal operation data, executing the step of determining whether the numerical verification data falls within the verification range corresponding to the risk data; in the presence of risk prediction data that matches the abnormal operation data, directly generating an abnormal verification result indicating that there is an abnormality in the risk component.

[0136] Among them, operation and maintenance data refers to historical maintenance records, component replacement cycle data, fault handling records and other data generated during the equipment maintenance process, which are used to reflect the correlation between equipment maintenance status and potential risks.

[0137] Risk prediction data is used to establish correlations between equipment anomalies and historical maintenance records, and also to indicate potential equipment anomalies. In practical implementation, machine learning models or statistical analysis methods can be used to analyze operation and maintenance data.

[0138] Optionally, determining whether there is risk prediction data that matches the abnormal operation data includes: determining the abnormality type corresponding to the abnormal operation data; determining whether there is an abnormality in the risk prediction data that matches the abnormality type; if so, determining that there is risk prediction data that matches the abnormal operation data; if not, determining that there is no risk prediction data that matches the abnormal operation data.

[0139] For example: During the operation of sewage treatment equipment, when the vibration data of the water pump fluctuates abnormally, the system first retrieves the maintenance records of the water pump for the past three months, including bearing replacement records and seal maintenance times. By analyzing the maintenance records, it is determined that the water pump may have cavitation. When it is detected that the vibration data of the water pump rises continuously within a preset period of time, since the increase in vibration data may be caused by cavitation, and the maintenance records indicate the risk of cavitation, it can be directly determined that there is an abnormality in combination with the maintenance records. On the contrary, if the maintenance records do not indicate any abnormalities that may cause the vibration data to rise, the numerical value-based abnormality verification step will be continued.

[0140] In this technical solution, risk prediction data is obtained by analyzing operation and maintenance data, effectively determining whether a matching prediction record exists for abnormal operation data. If no matching risk prediction data exists, the abnormality is precisely located by comparing the numerical verification data with the verification range. This approach not only improves the accuracy of fault diagnosis but also reduces the false alarm rate, thereby enhancing the reliability and efficiency of the entire sewage treatment equipment remote diagnosis system.

[0141] Step 402: Determine whether the change of the numerical verification data matches the change of the abnormal verification data based on the change matching rule.

[0142] In one example, continuing with the example that the risk component includes a water pump, the numerical verification data includes the vibration data of the water pump, and the abnormal verification data includes the current data and outlet pressure data of the water pump, accordingly, based on the change matching rule, it is determined whether the change of the numerical verification data matches the change of the abnormal verification data, including: when the vibration data continuously rises within a preset time period, determining whether the fluctuation amplitude of the current data reaches a preset fluctuation amplitude threshold, and whether the outlet pressure data decreases; when the fluctuation amplitude reaches the fluctuation amplitude threshold and the outlet pressure data decreases, determining that the change of the numerical verification data matches the change of the abnormal verification data.

[0143] The preset duration refers to a continuous monitoring time window set according to the type of water pump. In one example, the preset duration is determined within a time interval of 1 to 3 minutes.

[0144] The fluctuation amplitude threshold refers to the allowable range of current fluctuations set according to the rated power of the pump and is used to identify abnormal current fluctuations.

[0145] Step 403: When the change of the verification data matches the change of the validation data, an abnormality verification result is generated indicating that an abnormality exists in the risk component.

[0146] Optionally, when the change of the verification data does not match the change of the validation data, an abnormality verification result indicating that there is no abnormality in the risk component may be directly generated, or abnormality verification may be continued based on other methods.

[0147] Specifically, taking the water pump as an example of a risky component, when cavitation occurs in the water pump, the mechanical vibration will continue to increase, and the vibration sensor will detect an upward trend in amplitude. At the same time, cavitation causes abnormal motor load, and the current sensor captures fluctuations that exceed the normal range. In addition, cavitation affects the pump's suction capacity and fluid delivery efficiency, resulting in a decrease in outlet pressure. The system continuously tracks data within a set time window (for example, 2 minutes). When these three conditions are met simultaneously, the system will eliminate single data anomalies caused by environmental interference or temporary changes in operating conditions and accurately determine the presence of cavitation in the water pump.

[0148] This technical solution accurately identifies potential cavitation during pump operation. When pump vibration data continuously increases within a specific timeframe, a comprehensive assessment is made based on current fluctuations and changes in outlet pressure, effectively improving the accuracy of fault diagnosis. This multi-dimensional data linkage analysis method not only reduces false positives but also provides early warning of equipment anomalies, thereby preventing equipment damage caused by cavitation, extending pump life, and ensuring the stable operation of the sewage treatment system.

[0149] In this implementation, the matching relationship between changes in numerical verification data and changes in abnormal verification data can be used to perform anomaly verification, effectively identifying potential faults in risky components. Compared to traditional single-threshold judgment methods, this implementation significantly improves the accuracy and reliability of anomaly verification results by introducing change matching rules and combining dynamic data features for comprehensive analysis, thus providing a strong guarantee for the stable operation of the equipment.

[0150] Furthermore, the risk verification rule includes a verification range corresponding to the numerical verification data, and the verification range is determined when the risk component is abnormal.

[0151] The verification range refers to the pre-defined valid range of numerical verification data, used to initially determine whether the data conforms to an abnormal pattern. In actual implementation, risk verification rules can be determined by adding a safety margin to the historical abnormal data statistics. For example, the normal fluctuation range of the water pump vibration amplitude can be set to 2.5-3.8mm / s.

[0152] Accordingly, after determining whether the change of the numerical verification data matches the change of the abnormal verification data based on the change matching rule in step 402, the following steps are also included:

[0153] Step 404 : If the change in the verification data does not match the change in the check data, determine whether the numerical verification data falls within the corresponding verification range.

[0154] Step 405 : When the numerical verification data includes more than two pieces and some of the numerical verification data do not fall within the corresponding verification range, determine a verification adjustment parameter based on the number of pending verification data that do not fall within the corresponding verification range.

[0155] The calibration adjustment parameter is related to the number of pending calibration data and is used to determine the magnitude of the adjustment to the calibration range. Specifically, the number of pending calibration data indirectly reflects the probability of anomalies in risky components. Specifically, the fewer the number of pending calibration data, the more numerical calibration data falls within the calibration range, and the greater the probability of anomalies in the corresponding risky component, and vice versa.

[0156] In one example, the verification adjustment parameter is determined based on the number of pending verification data and the total number of verification data. For example, the ratio of the number of pending verification data to the total number of verification data is used as the verification adjustment parameter. In actual implementation, the number of pending verification data can also be directly used as the verification adjustment parameter.

[0157] Optionally, when the numerical verification data does not fall within the corresponding verification range, an abnormal verification result is generated indicating that there is no abnormality in the risk component; when the data verification data falls within the corresponding verification range, an abnormal verification result is generated indicating that there is an abnormality in the risk component.

[0158] Step 406 : Adjust the verification range corresponding to the pending verification data based on the verification adjustment parameter to obtain an adjusted range.

[0159] The adjusted range refers to the data determination interval expanded by verifying the adjustment parameters, and the adjusted range is larger than the verified range.

[0160] Optionally, adjusting the verification range corresponding to the pending verification data based on the verification adjustment parameter to obtain an adjusted range includes: determining an adjustment amplitude based on the verification adjustment parameter; and adjusting the verification range based on the adjustment amplitude to obtain the adjusted range. The adjustment amplitude is negatively correlated with the amount of pending verification data.

[0161] In one example, adjusting the calibration range based on the adjustment amplitude includes increasing an upper limit of the calibration range by the adjustment amplitude and / or decreasing a lower limit of the calibration range by the adjustment amplitude.

[0162] In one example, the verification adjustment parameter is the ratio of the number of pending verification data to the total amount of verification data. Accordingly, the adjustment amplitude is determined based on the verification adjustment parameter, including: multiplying the difference between 1 and the verification adjustment parameter by the length of the verification range as the verification adjustment amplitude.

[0163] In another example, the verification adjustment parameter is the number of pending verification data. Accordingly, determining the adjustment amplitude based on the verification adjustment parameter includes: determining the ratio of the maximum adjustment amplitude to the number of verification adjustment parameters as the verification adjustment amplitude.

[0164] Step 407: Determine whether the pending verification data falls within the corresponding adjusted range.

[0165] Step 408 : When all pending verification data fall within the corresponding adjusted range, an abnormality verification result indicating that the risk component is abnormal is generated.

[0166] Optionally, when there is verification data that falls within the corresponding adjusted range, an abnormality prompt message is generated indicating that there is no abnormality in the risk component.

[0167] By adopting this technical solution, changes in numerical verification data can be verified at multiple levels, not only considering the matching of changing trends but also introducing a dynamic adjustment mechanism for the verification range. This design effectively addresses the potential for misjudgment caused by data fluctuations during actual operation and improves the accuracy and reliability of anomaly judgments. In particular, when there are partial deviations in multiple sets of numerical verification data, adjusting the verification range ensures the comprehensiveness and rationality of anomaly verification results, thus providing a more scientific basis for the status assessment of risky components.

[0168] In some embodiments, reference Figure 5 In the above step 104, the abnormal operation data and abnormal verification data corresponding to the risk component are used to determine the abnormal handling method for the risk device, including:

[0169] Step 501: Input abnormal operation data and abnormal verification data into a pre-trained expert model to obtain an abnormality handling method.

[0170] The expert model is built based on an expert rule base. In one example, the expert model is implemented using an algorithm model based on a decision tree or a rule engine, automatically generating an exception handling method through rule matching and logical reasoning.

[0171] The expert rule base refers to a database that stores device anomaly diagnosis rules. It is used to store diagnostic rules formed by historical operation and maintenance experience. It can be constructed in the form of a relational database or a knowledge graph.

[0172] Accordingly, after generating the exception prompt information based on the exception handling method in step 105, the above step further includes:

[0173] Step 502: Obtain maintenance records of the sewage treatment equipment.

[0174] Maintenance records refer to the causes and solutions to equipment failures recorded during actual maintenance. Specifically, the fault codes, repair steps, and replacement parts information recorded in the work order system can be used to extract effective O&M experience data.

[0175] In one example, the maintenance record is uploaded to the equipment operation and maintenance system by maintenance personnel during the equipment maintenance process or after the equipment maintenance is completed.

[0176] Step 503: extracting operation and maintenance data from the maintenance record, and updating the expert rule base based on the operation and maintenance data.

[0177] Specifically, when the verification results indicate an anomaly in a risky component, the abnormal operating data and verified abnormal verification data are input into the pre-trained expert model. The expert model can then perform rule matching and logical reasoning based on the built-in expert rule library, automatically generating corresponding anomaly handling methods, such as recommending shutdown for maintenance or adjusting operating parameters. After the equipment is repaired, the system extracts the actual cause of the failure, treatment measures, and treatment effect data from the maintenance record. Using a rule extraction algorithm, this data is converted into new expert rules, and the expert rule library is dynamically updated. In this way, by continuously collecting maintenance data and updating the rule library, the diagnostic accuracy of the expert model can be gradually improved.

[0178] In one example, natural language processing (NLP) technology is used to extract keywords related to fault descriptions in maintenance records to obtain operation and maintenance data.

[0179] By adopting the above-mentioned technical solution, it is possible to automatically determine the precise handling method for abnormal situations of sewage treatment equipment, and optimize the handling strategy through continuous learning. Specifically, the solution uses a pre-trained expert model to analyze abnormal operation data and abnormal verification data, thereby generating targeted abnormal handling methods, improving the accuracy and efficiency of fault handling. At the same time, after generating the abnormal prompt information, the maintenance record is further obtained and the operation and maintenance data therein is extracted to update the expert rule base, realizing the self-learning and self-optimization capabilities of the system, making subsequent abnormality handling more intelligent and efficient. This not only improves the level of automation of equipment operation and maintenance, but also effectively reduces the risk of misjudgment caused by insufficient human experience, providing reliable protection for the stable operation of sewage treatment equipment.

[0180] In some embodiments, reference Figure 6 In the above step 104, if the abnormality check result indicates that the risk component has an abnormality, the following steps are also included:

[0181] Step 601: perform abnormality prediction on the sewage treatment equipment based on abnormal operation data to obtain abnormality prediction results.

[0182] The anomaly prediction results are used to indicate possible anomalies in the equipment. Specifically, because the equipment is composed of components, the status of each component will affect the equipment's status. Therefore, if an anomaly is identified in a risky component, it is necessary to perform an anomaly prediction for the sewage treatment equipment based on the abnormal operating data.

[0183] In one example, performing abnormality prediction on sewage treatment equipment based on abnormal operation data includes: determining an abnormality type corresponding to a risk component based on the abnormal operation data; and performing abnormality prediction on the sewage treatment equipment based on the abnormality type corresponding to the risk component.

[0184] Among them, the abnormal type of the component is a correspondence between abnormal types of the sewage treatment equipment set in advance.

[0185] In actual implementation, the abnormality type of the sewage treatment equipment may also directly correspond to the type of component operation data of each component, so that the abnormality type of the sewage treatment equipment can be directly deduced based on the type of abnormal operation data.

[0186] Furthermore, in order to improve the accuracy of the abnormality prediction results, abnormality prediction of sewage treatment equipment can be performed in combination with abnormality verification data.

[0187] Step 602: Update the device status of the associated devices based on the abnormality prediction result and the position of the sewage treatment equipment in the sewage treatment process.

[0188] Associated equipment refers to equipment whose performance is affected by the sewage treatment equipment. The relationships between these equipment are reflected in the sewage treatment process flow. For example, equipment located downstream of the sewage treatment equipment in the sewage treatment process is considered associated equipment. This can be achieved through process topology analysis, such as the impact chain of an aeration tank pump failure on dissolved oxygen concentration.

[0189] Specifically, considering that sewage treatment is a process in which a series of equipment work together according to a certain process flow, the abnormality of one device may affect the working performance of subsequent equipment, and then affect the final sewage treatment effect. Therefore, in the process of impact judgment, it is necessary to consider it in conjunction with the process flow.

[0190] In one example, the device status of associated equipment is updated based on the abnormality prediction result and the position of the sewage treatment equipment in the sewage treatment process flow, including: updating the device status of equipment directly related to the sewage treatment equipment among the associated equipment based on the abnormality prediction result, and determining the equipment after the device status is updated as the updated equipment; updating the device status of equipment directly related to the updated equipment among the associated equipment based on the device status of the updated equipment, and determining the updated equipment as the updated equipment, until all associated equipment are determined as updated equipment.

[0191] In actual implementation, the status of each related equipment can also be directly deduced based on the process flow chart.

[0192] Step 603 : predicting the sewage treatment effect based on the equipment status of each reference equipment in the sewage treatment process to obtain a predicted treatment result.

[0193] Reference devices are those that directly impact wastewater treatment results. Generally, they are located near the end of the wastewater treatment process, for example, the last three devices in the process. Since the status of reference devices may be updated during status updates, and the status of reference devices directly impacts wastewater treatment results, it is necessary to re-predict wastewater treatment results based on the status of the reference devices after the device status update.

[0194] The method for predicting the sewage treatment effect based on the device status of the reference device is pre-set. In one example, the sewage treatment effect prediction can be performed based on a trained effect prediction model, which is trained based on the device status of the reference device and the corresponding sewage treatment effect collected historically.

[0195] Furthermore, after the sewage treatment effect is predicted, prompts can be given to equipment operation and maintenance personnel based on actual conditions to assist them in making operation and maintenance decisions.

[0196] In the above technical solution, abnormalities of sewage treatment equipment can be predicted based on abnormal operation data, and the equipment status of the equipment in the sewage treatment process can be updated based on the abnormal prediction results. Finally, the sewage treatment effect can be predicted based on the equipment status of the reference equipment. In this way, when an abnormality is detected, the impact of the abnormality on the sewage treatment effect can be accurately judged, which can help assist operation and maintenance personnel in making operation and maintenance decisions.

[0197] This application also provides a multi-source data fusion and remote diagnosis system for sewage treatment equipment, Figure 7 The multi-source data fusion and remote diagnosis system for sewage treatment equipment includes a data acquisition component 710 and a cloud server 720 that is communicatively connected to the data acquisition component 710.

[0198] Data acquisition component 710 is used to acquire operational data from various components within the sewage treatment equipment and transmit it to cloud server 720. In one example, data acquisition component 710 includes at least one of a current sensor, a voltage sensor, a temperature sensor, and a vibration sensor. Accordingly, data acquisition component 710 can be installed on the component to be monitored.

[0199] A cloud operation and maintenance platform runs on the cloud server 720, and the cloud operation and maintenance platform is used to implement the multi-source data fusion and remote diagnosis method of the sewage treatment equipment provided by the above method embodiment.

[0200] The above are all preferred embodiments of the present application and are not intended to limit the scope of protection of this application. Unless otherwise specified, any feature disclosed in this specification (including the abstract and drawings) may be replaced by other equivalent or similar features. In other words, unless otherwise specified, each feature is merely an example of a series of equivalent or similar features.

Claims

1. A multi-source data fusion and remote diagnosis method for sewage treatment equipment, characterized in that: The method comprises: For each target component in the sewage treatment equipment, risk monitoring is performed on the target component based on the component operation data of the target component and the risk monitoring rules corresponding to the target component; When a risk component is detected, obtaining abnormality verification data corresponding to the risk component; Verifying whether the risk component has an abnormality based on the abnormality verification data and the risk verification rule corresponding to the target component, and obtaining an abnormality verification result; When the abnormality check result indicates that the risk component has an abnormality, determining an abnormality handling method for the risk device based on abnormal operation data with an abnormality in the component operation data corresponding to the risk component and the abnormality check data; Generate exception prompt information based on the exception handling method; The abnormal operation data includes numerical verification data represented by numerical values. The risk verification rule includes a change matching rule and a verification range corresponding to the numerical verification data. The change matching rule is used to record the correspondence between the change of the numerical verification data determined when the risk component is abnormal and the change of the abnormal verification data. The verification range is determined when the risk component is abnormal. The abnormal verification data and the risk verification rule corresponding to the target component are used to verify whether the risk component is abnormal, and obtain an abnormal verification result, including: Monitoring the changes in the numerical verification data and the abnormal verification data; Determining whether a change in the numerical verification data matches a change in the abnormal verification data based on the change matching rule; In a case where the change in the numerical verification data does not match the change in the abnormal verification data, determining whether the numerical verification data falls within a corresponding verification range; In a case where the numerical verification data includes more than two pieces and some of the numerical verification data do not fall within the corresponding verification range, determining a verification adjustment parameter based on a ratio of the number of pending verification data that does not fall within the corresponding verification range to the total amount of the numerical verification data; Adjusting the verification range corresponding to the pending verification data based on the verification adjustment parameter to obtain an adjusted range, wherein the adjusted range is larger than the verification range; Determining whether the pending verification data falls within the corresponding adjusted range; When all pending verification data fall within the corresponding adjusted range, generating an abnormality verification result indicating that the risk component is abnormal; When a risk component is detected, obtaining abnormality verification data corresponding to the risk component includes: In the case where the risk component is monitored, determining a risk type corresponding to the risk component based on the abnormal operation data; Determining a verification data acquisition method based on the risk type; In a case where the verification data acquisition method includes self-data, determining, among the component operation data corresponding to the risk component, associated operation data that has an associated relationship with the abnormal operation data as the abnormal verification data; In a case where the verification data acquisition method includes associated component data, the abnormal verification data is determined from component operation data of associated components corresponding to the risk component.

2. The method according to claim 1, characterized in that The determining the abnormality verification data from the component operation data of the associated component corresponding to the risk component includes: In a case where the risk components include more than two, determining whether the associated component is a risk component; In a case where the associated component is a risk component, determining abnormal device operation data in the operation data corresponding to the associated component as the abnormal verification data; If the associated component is not the risk component, determining whether the associated component and the risk component are of the same type; In a case where the associated component and the risk component are of the same type, determining data of the same type as the abnormal operation data in the component operation data corresponding to the associated component as the abnormality verification data; In a case where the associated component and the risk component are of different types, abnormality verification data is determined from component operation data corresponding to the associated component based on the risk type.

3. The method according to claim 1, characterized in that After determining whether the change of the numerical verification data matches the change of the abnormal verification data based on the change matching rule, the method further includes: When the change of the numerical verification data matches the change of the abnormality verification data, an abnormality verification result indicating that the risk component is abnormal is generated.

4. The method according to claim 1, wherein The abnormality verification data includes operation and maintenance data of the risk component. Before monitoring the change of the numerical verification data and the change of the abnormality verification data, the method further includes: Analyzing the operation and maintenance data to obtain risk prediction data corresponding to the risk components; determining whether there is the risk prediction data matching the abnormal operation data; In the absence of the risk prediction data matching the abnormal operation data, the step of determining whether the numerical verification data falls within a verification range corresponding to the risk prediction data is performed.

5. The method according to claim 1, wherein The risk component includes a water pump, the numerical verification data includes vibration data of the water pump, and the abnormal verification data includes current data and outlet pressure data of the water pump. The determining whether a change in the numerical verification data matches a change in the abnormal verification data based on the change matching rule includes: In the case where the vibration data continuously rises within a preset time period, determining whether the fluctuation amplitude of the current data reaches a preset fluctuation amplitude threshold, and whether the outlet pressure data decreases; When the fluctuation amplitude reaches the fluctuation amplitude threshold and the outlet pressure data decreases, it is determined that the change of the numerical verification data matches the change of the abnormal verification data.

6. The method according to claim 1, characterized in that The determining of the abnormality handling method for the risky device based on the abnormal operation data and the abnormal verification data including the abnormal operation data in the component operation data corresponding to the risky component, includes: Inputting the abnormal operation data and the abnormal verification data into a pre-trained expert model to obtain the abnormality handling method; Wherein, the expert model is constructed based on training using an expert rule base; After generating the exception prompt information based on the exception handling method, the method further includes: Obtaining maintenance records for the sewage treatment equipment; Operation and maintenance data are extracted from the maintenance records, and an expert rule base is updated based on the operation and maintenance data.

7. The method according to claim 1, characterized in that When the abnormality check result indicates that the risk component is abnormal, the method further includes: Performing abnormality prediction on the sewage treatment equipment based on the abnormal operation data to obtain an abnormality prediction result; updating the device status of associated equipment based on the abnormality prediction result and the position of the sewage treatment equipment in the sewage treatment process; The sewage treatment effect is predicted based on the equipment status of each preset reference equipment in the sewage treatment process to obtain a predicted treatment result.

8. A multi-source data fusion and remote diagnosis system for sewage treatment equipment, characterized in that: The system includes a data acquisition component and a cloud server in communication with the data acquisition component; The data acquisition component is used to acquire component operation data of each component in the sewage treatment equipment and send it to the cloud server; A cloud operation and maintenance platform runs on the cloud server, and the cloud operation and maintenance platform is used to implement the multi-source data fusion and remote diagnosis method of the sewage treatment equipment according to any one of claims 1 to 7.

Images