A sewage lifting equipment whole life cycle early warning system based on an AI algorithm

Abstract

The application provides a sewage lifting equipment whole life cycle early warning system based on an AI algorithm, comprising a data acquisition module, a data transmission module, an AI analysis module, a fault early warning and decision module and a user interaction module, and aims to solve the problems of inaccurate fault early warning, unscientific life assessment, extensive maintenance strategy and lack of whole life cycle management in the existing sewage lifting equipment management, so as to realize equipment operation attenuation quantification, accurate fault early warning, personalized maintenance decision and whole life cycle optimization through multi-source data acquisition and AI algorithm analysis, reduce enterprise operation cost, and improve equipment operation stability and industry intelligent level.

Description

Technical Field

[0001] This invention relates to the field of intelligent operation and maintenance technology, and in particular to an early warning system for the entire life cycle of sewage lifting equipment based on AI algorithms. Background Technology

[0002] With the large-scale development and intelligent transformation of the wastewater treatment industry, the operation stability of wastewater lifting equipment, as the core power unit of the wastewater treatment system, directly determines the wastewater treatment efficiency and environmental compliance. In the early days, the management of wastewater lifting equipment relied on manual inspection, which was inefficient and had a slow response to faults, making it difficult to meet the management needs of large-scale equipment clusters. In recent years, with the popularization of Internet of Things technology, some companies have introduced basic sensor monitoring systems to achieve the initial collection of operating parameters, but they are still in the primary stage of data collection and simple alarm.

[0003] In the course of industry development, equipment management technology has gradually evolved from passive maintenance to preventative maintenance. However, existing technologies still have significant shortcomings. For example, in fault early warning, current technologies can only trigger alarms based on a single parameter threshold, failing to comprehensively assess the equipment's operational degradation trend, leading to frequent false alarms and missed alarms. When assessing component lifespan, only single indicators such as operating time are relied upon, without considering factors such as load and environment, resulting in excessive replacement or operation with defects. In terms of maintenance strategies, a one-size-fits-all fixed-cycle maintenance model is typically adopted, leading to either wasted resources or insufficient maintenance. More importantly, existing technologies lack a full life-cycle management perspective, failing to integrate data from the entire process of equipment from break-in and stable operation to degradation and scrapping, thus failing to provide a scientific basis for equipment replacement and upgrades. These problems not only lead to high operating costs for enterprises but may also cause environmental risks such as wastewater overflows due to sudden equipment failures, hindering the high-quality development of the industry.

[0004] Therefore, there is an urgent need in this field for an AI-based early warning system for the entire lifecycle of wastewater lifting equipment to solve the above problems. Summary of the Invention

[0005] This invention provides an AI-based early warning system for the entire lifecycle of sewage lifting equipment, aiming to solve the problems of inaccurate fault warnings, unscientific life assessments, extensive maintenance strategies, and lack of full lifecycle management in the management of existing sewage lifting equipment. Through multi-source data collection and AI algorithm analysis, it realizes the quantification of equipment operation degradation, accurate fault warning, personalized maintenance decisions, and full lifecycle optimization, thereby reducing enterprise operating costs and improving equipment operation stability and the level of industry intelligence.

[0006] This invention provides an AI-based early warning system for the entire lifecycle of wastewater lifting equipment, comprising:

[0007] The data acquisition module is used to collect real-time monitoring parameters of the sewage lifting equipment, including the operating parameters of key components and the environmental parameters of the environment in which the equipment is located.

[0008] A data transmission module, which is connected to the data acquisition module, is used to transmit the acquired real-time monitoring parameters to a remote processing center;

[0009] The AI ​​analysis module, which is deployed in the remote processing center and communicates with the data transmission module, is used to receive and analyze the real-time monitoring parameters to perform equipment health status assessment, operational degradation quantification, and fault risk prediction.

[0010] A fault warning and decision-making module, which is connected to the AI ​​analysis module, is used to generate warning information and maintenance decision suggestions based on the output results of the AI ​​analysis module;

[0011] The user interaction module is communicatively connected to both the fault warning and decision-making module and the AI ​​analysis module, and is used to display equipment status, warning information, and decision suggestions to the user.

[0012] And receive user feedback.

[0013] Compared with the prior art, the beneficial effects achieved by the present invention are as follows:

[0014] 1. This invention integrates multi-source data through AI algorithms to achieve quantitative assessment of equipment performance degradation and prediction of failure risks, thereby avoiding business interruptions and environmental risks caused by sudden failures and reducing economic losses.

[0015] 2. This invention is based on a life prediction model that considers multiple factors such as load and environment. It can accurately determine the remaining life of components, avoid excessive replacement or operation with defects, and, combined with personalized maintenance plans, significantly reduce maintenance costs and resource waste.

[0016] 3. This invention integrates data from the entire process of equipment break-in, stable operation, and wear-out / scrapping, providing a full-cycle perspective for equipment management, reducing manual inspection workload, shortening fault repair time, and improving overall equipment operating efficiency;

[0017] 4. This invention, through the collection of massive amounts of operational data, can provide manufacturers with a basis for product optimization, promote equipment iteration and upgrading, and at the same time provide data support for industry policy formulation, guiding the industry towards high efficiency, environmental protection and intelligence;

[0018] 5. This invention continuously enriches the AI ​​algorithm training data through user feedback, constantly improving the accuracy of early warnings and the scientific nature of decision-making, forming a cycle of monitoring-analysis-decision-feedback, and optimizing equipment management effectiveness in the long term.

[0019] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0020] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention, but do not constitute a limitation thereof; in the drawings:

[0021] Figure 1 This is a schematic diagram of the structure of a full life cycle early warning system for sewage lifting equipment based on AI algorithms provided by the present invention. Detailed Implementation

[0022] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0023] Example 1:

[0024] This invention provides an AI-based early warning system for the entire lifecycle of wastewater lifting equipment. Please refer to [link / reference]. Figure 1 ,include:

[0025] The data acquisition module is used to collect real-time monitoring parameters of the sewage lifting equipment. These parameters include the operating parameters of key components and the environmental parameters of the environment in which the equipment is located.

[0026] The data transmission module, which is connected to the data acquisition module, is used to transmit the acquired real-time monitoring parameters to the remote processing center.

[0027] The AI ​​analysis module, deployed in a remote processing center and connected to the data transmission module, is used to receive and analyze real-time monitoring parameters to assess equipment health status, quantify operational degradation, and predict fault risks.

[0028] The fault warning and decision-making module, which is connected to the AI ​​analysis module, is used to generate warning information and maintenance decision suggestions based on the output of the AI ​​analysis module;

[0029] The user interaction module communicates with the fault warning and decision-making module and the AI ​​analysis module respectively. It is used to display the device status, warning information, decision suggestions to the user, and receive user feedback.

[0030] Specifically, this embodiment comprehensively captures the operating data (current, temperature, speed, liquid level signal, etc.) of key components of the sewage lifting equipment (such as water pumps, crushers, electromagnetic level floats, etc.) and environmental data of the environment in which the equipment is located (humidity, pH, magnetic field strength, etc.) through the data acquisition module. The data is then securely and stably transmitted to the AI ​​analysis module in the remote processing center through the data transmission module. The AI ​​analysis module performs in-depth processing of the data through preset AI algorithms to achieve health status assessment, quantification of operational degradation, and fault risk prediction throughout the entire life cycle of the equipment. The fault warning and decision-making module generates targeted warning information and maintenance decision suggestions based on the analysis results of the AI ​​analysis module. The user interaction module establishes a two-way interaction channel between the user and the system, presenting the user with relevant equipment status, warning and decision information, and collecting user feedback to optimize system performance. All modules work together to form a closed-loop equipment life cycle warning and management system.

[0031] In one implementation, the data acquisition module includes:

[0032] The water pump monitoring unit includes a current sensor installed on the water pump power supply line and a temperature sensor installed on the water pump bearing position, used to collect the water pump operating current and operating temperature.

[0033] The liquid level signal acquisition unit is connected to the signal output device of the electromagnetic liquid level float and is used to acquire liquid level signals in real time.

[0034] An environmental sensing unit is distributed around the device to collect parameters such as ambient humidity, temperature, pH, and magnetic field strength.

[0035] The crusher monitoring unit includes a speed sensor, a temperature sensor, and a running timer installed on the crusher motor, used to collect motor speed, temperature, and cumulative running time.

[0036] Specifically, in this embodiment, the water pump monitoring unit accurately collects current data during water pump operation by deploying a current sensor on the water pump power supply line, and a temperature sensor is installed on the water pump bearing to capture temperature changes in real time during water pump operation. Both sensors together reflect the water pump's operating status. The liquid level signal acquisition unit directly interfaces with the signal output device of the electromagnetic liquid level float to acquire liquid level signals in real time to monitor the sewage level. The environmental sensing unit is rationally distributed around the equipment to comprehensively collect parameters such as environmental humidity, temperature, pH, and magnetic field strength, providing data support for analyzing the impact of the environment on equipment operation. The crusher monitoring unit collects motor speed and temperature data through a speed sensor and a temperature sensor installed on the crusher motor, and records the cumulative running time of the motor using a timer, providing multi-dimensional data for evaluating the crusher motor's performance and lifespan. It should be noted that the types of sensors mentioned above are not limited in this embodiment, and the principles will not be further elaborated.

[0037] In one implementation, the data transmission module includes:

[0038] The local aggregation node is connected to each sensor unit of the data acquisition module and is used to perform preliminary sorting and packaging of the real-time monitoring parameters collected by each sensor unit.

[0039] The network transmission unit is used to transmit packaged data to the AI ​​analysis module in the remote processing center via wired network or wireless low-power wide area network technology.

[0040] Specifically, in this embodiment, the local aggregation node establishes connections with various sensing units of the data acquisition module, such as the water pump monitoring unit, liquid level signal acquisition unit, environmental sensing unit, and crusher monitoring unit. It receives the dispersed real-time monitoring parameters collected by each unit, performs preliminary data processing such as format regularization and redundancy removal, and then packages the data into standardized data packets. The network transmission unit selects the transmission method according to the equipment installation scenario. For equipment with fixed installation locations and close to the remote processing center, wired network (such as Ethernet) transmission is used to ensure the stability and accuracy of data transmission. For equipment installed in complex environments or with high mobility, wireless low-power wide area network technologies such as ZigBee and NB-IoT are used to achieve flexible and efficient data transmission. Finally, the packaged standardized data packets are transmitted to the AI ​​analysis module of the remote processing center.

[0041] In one implementation, the AI ​​analytics module is configured to execute a comprehensive health assessment function covering the entire life cycle. It is used to quantify and score the entire process of a device's operation from initial operation to the end of its performance at time t;

[0042] Evaluation function for:

[0043] ;

[0044] In the formula, For instantaneous performance health based on real-time multi-source data fusion; To characterize the overall operational degradation degree of performance deviating from the ideal state; To predict the remaining effective lifetime based on physical models and load history; Rated design life; This is the preset equipment break-in period constant; The process influence coefficient is preset based on equipment type and engineering experience; It is a unit step function, ensuring that risk accumulation only occurs when performance health deteriorates;

[0045] Evaluation function The value range is [0,1], and its dynamic change curve directly represents the health trajectory of the equipment throughout its entire life cycle and is output to the fault warning and decision-making module.

[0046] Specifically, the AI ​​analysis module provided in this embodiment executes a comprehensive health assessment function covering the entire life cycle. This enables a quantitative assessment of the health status of equipment throughout its entire lifecycle; among which, The data is obtained by fusing and calculating real-time multi-source data collected by the data acquisition module. Based on the deviation analysis between real-time monitoring parameters and ideal operating parameters, Calculated by combining physical models with historical equipment load data. These are the rated design life parameters set at the time of manufacture of the equipment. Based on the equipment type and industry-standard presets, this represents the break-in period from initial operation to stable operation of the equipment; Based on the structural characteristics, operating principles, and extensive engineering experience of different equipment types, this system is designed to adjust the weight of each influencing factor on the evaluation results, and can also be adjusted based on user feedback during equipment use. For the standard unit step function, when When >0 (i.e., performance health decreases), =1, when When ≤0 (i.e., performance health remains unchanged or increases), =0.

[0047] In the formula, This is the initial break-in impact term, used to quantify the impact of the break-in period on the health status of the equipment from initial operation to stable operation. Essentially, it simulates the performance evolution law of the equipment during the break-in period through an exponential function, reflecting the characteristics of gradual optimization of initial performance and tendency to stabilize after the break-in is completed.

[0048] when =0 (when the device is initially running), =0, =1, the formula result is 1−1=0, indicating that the equipment is in the initial break-in stage when it is first started, and its performance has not yet stabilized. The health status assessment needs to exclude the interference of initial unstable factors; when t is in (0, During the break-in period, It increases as t increases. The value decreases as t increases, while the formula result gradually increases from 0, reflecting the continuous optimization of equipment performance through break-in, and the gradual weakening of the negative impact of the break-in period on health assessment; when t≥ When the break-in period is complete, ≥1, ≤ ≈0.37 (assuming k=1), the formula result approaches 1, indicating that the equipment has entered the stable operation stage. The impact of the break-in period on the health status can be ignored. The health assessment mainly depends on subsequent performance degradation and lifespan consumption factors.

[0049] This is a performance degradation and risk accumulation term used to quantify the combined impact of the real-time performance health level and accumulated risk of the equipment during the stable operation phase. By subtracting the accumulated degradation risk from the real-time health status, the dynamic changes in equipment performance are characterized.

[0050] When the device performance is healthy Rise or remain stable ( When ≥0), =0, cumulative risk score is 0, formula result equals That is, only the real-time performance health status is used as the evaluation basis for this stage;

[0051] When the device performance is healthy decline( When <0), =1, cumulative risk score calculation begins, the formula result is: minus The product of the cumulative risk score and the real-time health score is the cumulative risk of the current stage. The faster the rate of decline and the longer the duration, the larger the cumulative risk score and the smaller the result of the formula, reflecting the decline in health status when the equipment performance deteriorates.

[0052] This is a lifespan consumption factor. Reflecting the proportion of remaining effective service life to the rated design life, through an index. The extent to which adjusted life expectancy affects health assessment results;

[0053] By multiplying the effects of initial break-in, performance degradation and risk accumulation, and lifespan consumption, and comprehensively considering the key influencing factors at different stages of the equipment's entire life cycle, a health assessment value with a range of [0,1] is finally obtained. , The closer the value is to 1, the better the health status of the equipment. Its dynamic change curve fully presents the health trajectory of the equipment from initial operation to the end of performance, providing an analytical basis for the fault early warning and decision-making module.

[0054] In one implementation, instantaneous performance health for:

[0055] ;

[0056] In the formula, Let be the i-th normalized real-time monitoring parameter; and Parameters The sample mean and standard deviation under historical normal operating conditions; The time-varying weight of this parameter is calculated using the following formula: ,in For static base weights, This is a sensitivity coefficient, used to give more attention to parameters that change drastically.

[0057] Specifically, in this embodiment The results are calculated using a multi-parameter adaptive fusion model and are used to quantify the real-time performance health level of the device; among which... It is the value of each real-time monitoring parameter (such as current, temperature, speed, etc.) collected by the data acquisition module after normalization. Normalization can eliminate the difference in the dimensions of different parameters. and The sample mean and standard deviation are obtained by statistical analysis of the sample data of the i-th parameter under the historical normal operating conditions of the equipment. The calculation process will not be described in detail here. The static base weights are preset based on the importance of the i-th parameter to the device performance. The sensitivity coefficients, which are preset based on parameter characteristics, are all set by administrators or users.

[0058] The squared deviation of the i-th real-time monitoring parameter from the mean of the parameter under historical normal operating conditions is divided by the standard deviation. Standardize the deviation to reflect the degree to which the parameter deviates from the normal state; For time-varying weights of parameters, through calculate, The absolute value of the rate of change of the i-th parameter; the larger the rate of change, the greater the absolute value of the rate of change. The larger the value, the higher the weight of parameters that change drastically in the health assessment, reflecting the focus on the dynamic changes of the parameters; The weighted sum of squared deviations of all parameters comprehensively reflects the overall deviation of multiple parameters from the normal state; through... The weighted sum of squared deviations is converted into an instantaneous performance health value with a range of (0,1] in the form of [formula missing]. The smaller the weighted sum of squared deviations, the better. The closer it is to 1, the higher the real-time performance health level of the device.

[0059] In one implementation, the overall operating attenuation for:

[0060] ;

[0061] In the formula, For a moment The equipment's equivalent load rate is obtained by comprehensively calculating the current and speed. This is the rated load; The environmental stress factor is calculated by combining temperature, humidity, and pH. The instability factor is calculated from the variance of the operating parameters of key components within a short time window; The damage index is determined by the material and structural properties.

[0062] Specifically, in this embodiment A cumulative damage model considering load and environment is used to characterize the cumulative degree of deviation of equipment performance from the ideal state. By comprehensively calculating the parameters such as current and speed collected by the data acquisition module, load parameters of different dimensions are converted into a unified equivalent load rate. The calculation method is conventional and will not be elaborated further. The rated load parameters set during equipment design; Temperature, humidity, and pH parameters collected by the environmental sensing unit are weighted and calculated (the specific calculation process can be flexibly based on actual applications) to reflect the stress impact of the environment on the equipment. The variance of key component operating parameters (such as current, speed, temperature, etc.) within a short time window (e.g., 5 minutes) is calculated by selecting a short time window. The larger the variance, the more unstable the operation. The characteristics of the key components of the equipment are determined based on the material properties (such as wear resistance and corrosion resistance) and structural design features. They are derived through experiments and engineering experience, preset by managers or users, and can be adjusted based on feedback.

[0063] in the formula For load-related damage items, This is the ratio of the actual load rate to the rated load rate. When the actual load rate exceeds the rated load rate, this value is greater than 1, and the damage caused by overload is amplified by the exponent m. For environmental damage impact items, A larger value indicates greater environmental stress; a larger value reflects the aggravating effect of environmental stress on equipment damage. (The coefficient is...) Adjusting the weighting of the impact of environmental stress; For the damage caused by unstable operation, The larger the value, the more unstable the operation; a larger value reflects the additional damage caused by unstable operation to the equipment. Adjust the weighting of the impact of operational instability;

[0064] By integrating the product of the above three terms over the time interval [0, t], the overall operational attenuation is obtained. Integration results The larger the value, the more severe the cumulative degradation of the equipment.

[0065] In one implementation, the remaining effective lifespan is predicted. for:

[0066] ;

[0067] In the formula, This represents the overall operational attenuation at the current moment. The instantaneous decay rate at the current moment is obtained by linear fitting through a sliding time window; It is a very small positive number used to prevent the denominator from being zero, ensuring the numerical stability of the formula when the decay rate is close to zero.

[0068] Specifically, in this embodiment, The remaining effective lifespan of the equipment is predicted by calculating using a dynamic model extrapolated from the current attenuation rate; where, The overall operational attenuation rate calculated using the above formula is... By selecting a fixed-length sliding time window (e.g., 1 hour), the time within the window is... The data is linearly fitted, and the slope of the fitted line is calculated to obtain the instantaneous decay rate at the current moment. For a preset very small positive number (such as 10) −6 ), used to avoid instantaneous decay rate The meaningless situation caused by the denominator approaching zero when the value is close to zero.

[0069] In the formula, the numerator The remaining healthy capacity reflects the current health level of the equipment; the denominator is the current effective decay rate, calculated by taking the instantaneous decay rate and the minimum normal value. The maximum value is used to ensure that the denominator is always positive and the value is stable. The formula is as follows: the remaining effective lifespan is equal to the remaining healthy capacity divided by the current effective decay rate, which is to predict how long the device can maintain normal operation based on the current degree and rate of decay.

[0070] In one implementation, the fault warning and decision-making module is configured as follows:

[0071] Receive the output of the AI ​​analysis module and ;

[0072] Settings based on The three-level dynamic early warning threshold, when When a warning is triggered at the attention level; and A level alert is triggered when <0; when An emergency-level warning is triggered at this time;

[0073] according to Main contribution parameters and The low-scoring parameters are used to locate the faulty components and causes by combining them with preset judgment rules, and to generate decision instructions that include maintenance, adjustment or replacement suggestions.

[0074] Specifically, in this embodiment To meet the three-level early warning threshold preset according to equipment operation safety requirements, maintenance costs, and industry standards (meeting) ),when When this occurs, it indicates a slight abnormality in the device's health status, triggering a level-of-concern alert to remind the user to monitor the device's operating status; when and When the health status is <0 (i.e., the health status continues to deteriorate), an action-level warning is triggered, requiring the user to take timely countermeasures; when When this occurs, it indicates that the equipment's health status is seriously abnormal, the risk of failure is extremely high, triggering an emergency warning that requires immediate action.

[0075] In terms of fault location and decision command generation, analysis The calculation process determines the parameter that contributes the most to the overall operational attenuation (i.e., the main contributing parameter), and simultaneously identifies... The calculation uses low-scoring parameters with high weight and large deviation. Combined with preset fault judgment rules (such as the correspondence between parameters and faulty components, the matching rules between deviation range and fault type, etc.), the faulty component and fault cause are accurately located. Then, targeted maintenance, adjustment or replacement suggestions are generated, forming a decision instruction. The process of locating the faulty component and generating the decision instruction is not limited in this embodiment. The preset judgment rules can be flexibly adjusted by managers or users based on the actual application environment and equipment components.

[0076] In one implementation, the fault warning and decision-making module further includes:

[0077] Combination Predictive trajectory and equipment maintenance economic model, dynamic programming of optimal preventive maintenance timing. Its objective function is to minimize the total expected cost per unit time, and in Drop to a preset threshold or near At that time, maintenance work orders and spare parts lists are automatically generated and pushed to the user interaction module.

[0078] Specifically, in this embodiment, combined with Predictive trajectory and equipment maintenance economic model, dynamic programming of optimal preventive maintenance timing. This can be achieved using the following formula:

[0079] ;

[0080] In the formula, The total expected cost per unit of time; For candidate preventive maintenance time points (i.e., the ones to be solved) ); Fixed costs for preventative maintenance, including labor costs, spare parts costs, and downtime losses, are preset based on equipment model, maintenance procedures, and industry cost standards. Additional costs incurred in repairing a failure, including emergency repair costs, business interruption losses, and environmental penalty risk costs, are determined based on historical failure loss data. The unit time loss cost of continuous equipment operation includes energy consumption loss, implicit costs of accelerated component decay, etc., and is calculated based on equipment energy consumption parameters and component life cycle costs.

[0081] For the equipment at all times The probability of a failure occurring is determined by The calculation shows that, among which For a moment The predicted remaining effective lifespan, i.e. The predicted trajectory; The average downtime for fault repairs is calculated based on historical repair records. For the current time ( The predicted remaining effective lifetime is based on (=0), based on Calculated using the formula; The minimum health threshold corresponding to preventive maintenance is preset based on equipment maintenance requirements and engineering experience to ensure that the equipment still has maintainable value during maintenance.

[0082] In the formula, The predicted trajectory is continuously calculated and updated by the AI ​​analysis module, reflecting the changing trend of the equipment's remaining effective lifespan at different points in time. Based on this trajectory, the fault warning and decision-making module... The formula calculates the failure probability at different candidate time points $T$. The smaller the value, the higher the risk of failure per unit time. The faster the growth.

[0083] In the objective function, the numerator comprehensively considers the fixed costs of preventive maintenance, the additional costs of failures, and the costs of continuous operation losses, while the denominator is the effective operating time of the equipment (including actual operating time and deductions for downtime due to failures). The objective function aims to minimize... To achieve optimal economic efficiency. Meanwhile, constraints are limited. It must be within the current remaining effective lifespan, and the equipment health level must not be lower than [a certain threshold] at the time of maintenance. This avoids premature maintenance or maintenance when the equipment is beyond repair.

[0084] Find the solution through numerical solutions or optimization algorithms (such as gradient descent) that make smallest This is the optimal time for preventative maintenance. .when The value drops to a preset maintenance trigger threshold (e.g.) +0.1), or the current running time and When the difference is less than a preset time interval (such as 24 hours), the system automatically generates a maintenance work order (which specifies the maintenance time, process, and technical requirements) and a spare parts list (determined based on the wear and tear of key equipment components and maintenance needs), and pushes it to the user interaction module.

[0085] In one implementation, the user interaction module includes:

[0086] The full life cycle health curve unit is used for dynamic drawing and display. and The curve shows how the curve changes over time and is compared with the design baseline curve.

[0087] The intelligent early warning dashboard unit is used to display early warning information at all levels, faulty components and causes, and decision-making instructions;

[0088] The maintenance collaboration center unit is used to receive maintenance plans in real time, provide feedback on maintenance results, and upload on-site multimedia data to form a closed-loop maintenance record.

[0089] Specifically, this embodiment allows users to intuitively compare the differences between the actual curve and the baseline curve, and comprehensively grasp the changing trends of equipment health status. The intelligent early warning dashboard unit adopts a structured display method, clearly presenting early warning information at all levels (early warning level, trigger time, associated equipment, etc.), faulty components and fault cause analysis results, and corresponding decision instructions (maintenance, adjustment, or replacement suggestions), facilitating users to quickly obtain key information. The maintenance collaboration center unit provides users with maintenance-related interactive functions. Users can receive maintenance plans (maintenance work orders, spare parts lists) pushed by the system in real time through this unit, and provide feedback on maintenance results (such as maintenance completion status, fault resolution effect, etc.) after completing the maintenance work. They can also upload on-site pictures, videos, and other multimedia data during the maintenance process. The system records and archives this information, forming a complete maintenance closed loop, providing data support for subsequent equipment management and system optimization. It should also be noted that the maintenance results and actual fault conditions reported by users can be used as incremental training samples for the adjustable coefficients in the various algorithms of the AI ​​analysis module, continuously optimizing model parameters and improving early warning accuracy.

[0090] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An AI algorithm-based sewage lifting equipment full-life cycle early warning system, characterized in that, include: The data acquisition module is used to collect real-time monitoring parameters of the sewage lifting equipment, including the operating parameters of key components and the environmental parameters of the environment in which the equipment is located. A data transmission module, which is connected to the data acquisition module, is used to transmit the acquired real-time monitoring parameters to a remote processing center; The AI ​​analysis module, which is deployed in the remote processing center and communicates with the data transmission module, is used to receive and analyze the real-time monitoring parameters to perform equipment health status assessment, operational degradation quantification, and fault risk prediction. A fault warning and decision-making module, which is connected to the AI ​​analysis module, is used to generate warning information and maintenance decision suggestions based on the output results of the AI ​​analysis module; The AI analysis module is configured to execute a whole life cycle health comprehensive evaluation function for quantitatively scoring the whole process state of the device from initial operation to performance termination at time t; The evaluation function is: ； In the formula, is the instantaneous performance health degree based on real-time multi-source data fusion; is the comprehensive operating degradation degree representing the performance deviation from the ideal state; is the predicted remaining effective life based on the physical model and load history; is the rated design life; is the preset equipment running-in period constant; is the process influence coefficient preset according to the equipment type and engineering experience; is a unit step function, ensuring that risk accumulation is only performed when the performance health degree decreases. The evaluation function The value range of the evaluation function is [0, 1], the dynamic curve of which directly represents the health trajectory of the equipment in the whole life cycle and is output to the fault early warning and decision module. The user interaction module is communicatively connected to the fault warning and decision-making module and the AI ​​analysis module, respectively, and is used to display the device status, warning information, decision suggestions to the user, and receive user feedback.

2. The system of claim 1, wherein, The data acquisition module includes: The water pump monitoring unit includes a current sensor installed on the water pump power supply line and a temperature sensor installed on the water pump bearing position, used to collect the water pump operating current and operating temperature. The liquid level signal acquisition unit is connected to the signal output device of the electromagnetic liquid level float and is used to acquire liquid level signals in real time. An environmental sensing unit is distributed around the device to collect parameters such as ambient humidity, temperature, pH, and magnetic field strength. The crusher monitoring unit includes a speed sensor, a temperature sensor, and a running timer installed on the crusher motor, used to collect motor speed, temperature, and cumulative running time.

3. The system of claim 1, wherein, The data transmission module includes: The local aggregation node is connected to each sensing unit of the data acquisition module and is used to perform preliminary sorting and packaging of the real-time monitoring parameters collected by each sensing unit. The network transmission unit is used to transmit the packaged data to the AI ​​analysis module of the remote processing center via a wired network or wireless low-power wide area network technology.

4. The system according to claim 1, characterized in that, The instantaneous performance health for: ； In the formula, Let be the i-th normalized real-time monitoring parameter; and Parameters The sample mean and standard deviation under historical normal operating conditions; The time-varying weight of this parameter is calculated using the following formula: ,in For static base weights, This is a sensitivity coefficient, used to give more attention to parameters that change drastically.

5. The system according to claim 1, characterized in that, The overall operational attenuation for: ； In the formula, For a moment The equipment's equivalent load rate is obtained by comprehensively calculating the current and speed. This is the rated load; The environmental stress factor is calculated by combining temperature, humidity, and pH. The instability factor is calculated from the variance of the operating parameters of the key components within a short time window; The damage index is determined by the material and structural properties.

6. The system according to claim 1, characterized in that, The predicted remaining effective lifespan for: ； In the formula, This represents the overall operational attenuation at the current moment. The instantaneous decay rate at the current moment is obtained by linear fitting through a sliding time window; It is a very small positive number used to prevent the denominator from being zero, ensuring the numerical stability of the formula when the decay rate is close to zero.

7. The system according to claim 1, characterized in that, The fault warning and decision-making module is configured as follows: Receive the output of the AI ​​analysis module and ; Settings based on The three-level dynamic early warning threshold, when < A warning at the attention level will be triggered at this time; when < and A level-one alert is triggered when the value is <0. when < An emergency-level warning is triggered at this time; according to Contribution parameters and The low-scoring parameters are used to locate the faulty components and causes by combining them with preset judgment rules, and to generate decision instructions that include maintenance, adjustment or replacement suggestions.

8. The system according to claim 7, characterized in that, The fault warning and decision-making module also includes: Combination Predictive trajectory and equipment maintenance economic model, dynamic programming of optimal preventive maintenance timing. Its objective function is to minimize the total expected cost per unit time, and in Drop to a preset threshold or near At that time, maintenance work orders and spare parts lists are automatically generated and pushed to the user interaction module.

9. The system according to claim 1, characterized in that, The user interaction module includes: The full life cycle health curve unit is used for dynamic drawing and display. and The curve shows how the curve changes over time and is compared with the design baseline curve. The intelligent early warning dashboard unit is used to display early warning information at all levels, faulty components and causes, and decision-making instructions; The maintenance collaboration center unit is used to receive maintenance plans in real time, provide feedback on maintenance results, and upload on-site multimedia data to form a closed-loop maintenance record.

Images