A remote automatic control method applied to a pump station

Abstract

The application belongs to the technical field of industrial automation control, and specifically discloses a remote automation control method applied to a pump station, which comprises the following steps: acquiring multi-source sensor data composed of sand concentration data, pump set running current and rotating speed data, and pressure data; identifying a device chain reaction caused by water quality deterioration by analyzing the time sequence correlation between sand concentration abnormality and current fluctuation in real time; generating a dynamic regulation instruction based on the above, reducing the running load of the affected pump to reduce wear, and simultaneously adjusting the running parameters of the healthy pump to maintain the total output flow; collecting feedback data to evaluate the regulation effect after executing the instruction, and ensuring the stable operation of the system through iterative adjustment. Overload risk early warning is continuously carried out during the processing, and preventive maintenance suggestions for the equipment are generated by comparing and analyzing the historical data. The scheme realizes the transformation from passive response to active defense, and maintains the overall performance of the system while ensuring the safety of single equipment.

Description

Technical Field

[0001] This invention belongs to the field of industrial automation control technology and relates to a remote automated control method for pumping stations. Background Technology

[0002] Pumping stations, as key infrastructure in water conservancy, municipal, and industrial sectors, undertake the important task of fluid transportation and dispatching. To improve operational efficiency and management convenience, modern pumping stations generally adopt remote automated control systems. These systems utilize various sensors and actuators to remotely monitor functions such as pump start-up and shutdown, and adjustment of operating parameters. In systems with multiple pumps operating in parallel, the operating status of each pump influences the others, placing higher demands on the coordination capabilities and intelligence level of the control system.

[0003] Currently, remote control technology applied to pumping stations typically employs modular monitoring and protection mechanisms. For example, the system may have an independent water quality monitoring module to monitor indicators such as sediment concentration in the water and set corresponding alarm thresholds. Simultaneously, the system also equips each pump with an independent electrical protection system, such as overcurrent and overload protection. When the operating current exceeds a set value, corresponding protection actions are triggered, such as load reduction or shutdown. These monitoring and protection functions operate independently, responding to a single fault signal.

[0004] However, existing technologies lack effective data correlation and collaborative logic between various monitoring and control modules. When the sediment concentration in water increases, the system may only treat it as an abnormal water quality event, failing to correlate it in real time with the performance degradation caused by the event to pump impeller wear, and the resulting chain reaction of reduced single-pump efficiency, leading to increased load on other parallel pumps to compensate for flow, ultimately resulting in overload. This information silo and fragmented control logic make the system lack the ability to effectively identify and respond to such cross-domain, causally related, complex fault scenarios, resulting in delayed regulatory responses and making it difficult to fundamentally avoid systemic risks. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art and to achieve the above objectives, the present invention proposes the following technical solution: a remote automated control method for pumping stations, comprising: step 1, acquiring sand concentration data at the pumping station inlet, operating current data and speed data of each pump in a multi-pump parallel system, and pressure data at key locations of the pumping station, to obtain multi-source sensor data.

[0006] Step 2: Perform real-time collaborative analysis on multi-source sensor data. When the sand concentration data exceeds its normal range and the operating current data shows abnormal fluctuations consistent with the time of change of the sand concentration data, generate an impact confirmation result.

[0007] Step 3: Based on the impact confirmation results, generate a dynamic control instruction that includes adjusting the operating parameters of the affected pump and coordinating the control of the operating parameters of other pumps.

[0008] Step 4: Execute the dynamic control command and obtain the operating current data, speed data and pressure data after the command is executed to obtain feedback data.

[0009] Step 5: Evaluate the feedback data. When the feedback data indicates that the operating current is not within the preset safe operating range or the pressure is unstable, generate a parameter adjustment command and execute the command until the system returns to stable operation.

[0010] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) The present invention realizes in-depth perception and intelligent intervention of the pump station operation status by constructing a multi-source data collaborative analysis and closed-loop feedback control mechanism. It is not simply monitoring the sediment concentration or overload, but conducting causal correlation analysis between the two, which can accurately identify the equipment chain reaction caused by water quality deterioration, thereby providing early warning and taking targeted collaborative control measures, effectively preventing the spread and escalation of faults, and greatly enhancing the operational stability and anti-interference ability of the entire pump station system.

[0011] (2) By implementing differentiated and coordinated control of affected and healthy pumps, this invention can effectively protect damaged equipment while ensuring the stability of the total output flow of the system, thus avoiding the drawback of sacrificing the overall system performance to protect a single device in traditional emergency solutions. This dynamic and refined control method ensures that the pumping station can maintain a highly efficient and economical operating state when dealing with sudden working conditions, thereby improving the accuracy of the control strategy and the efficiency of system operation.

[0012] (3) Through deep learning and analysis of historical event data, this invention can continuously optimize early warning thresholds and control strategies, and generate forward-looking maintenance suggestions based on the actual impact of abnormal operating conditions on equipment performance. This data-driven preventive maintenance model makes equipment management more scientific, effectively reduces the wear of key components, reduces the risk of unplanned downtime, and realizes the transformation from passive response to proactive prevention, thereby ensuring the long-term reliable operation of the pumping station. Attached Figure Description

[0013] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0014] Figure 1 This is a schematic diagram of the implementation steps of the method of the present invention. Detailed Implementation

[0015] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0016] Please see Figure 1 As shown, the present invention proposes a remote automated control method for pumping stations, which includes: Step 1, acquiring sand concentration data at the pumping station inlet, operating current data and speed data of each pump in the multi-pump parallel system, and pressure data at key locations of the pumping station, to obtain multi-source sensor data.

[0017] In a preferred embodiment, obtaining multi-source sensor data includes: collecting sand concentration data through a sand concentration detector installed at the pump station inlet.

[0018] Operating current and speed data are collected by current sensors and speed sensors installed on each pump in a multi-pump parallel system.

[0019] Pressure data is collected by pressure sensors installed at key locations in the pumping station.

[0020] Data on sand concentration, operating current, rotational speed, and pressure are collected and transmitted via a wireless communication module to obtain multi-source sensor data.

[0021] Specifically, at the main inlet of the pumping station, i.e. the common pipeline from which all pumps draw water, a non-contact sediment concentration detector, such as an ultrasonic or optical scattering detector, is deployed. This detector continuously measures the turbidity or particulate matter concentration of the water flowing through it at a preset sampling frequency and converts the measured values ​​into standardized sediment concentration data.

[0022] Secondly, for each independent pump in a multi-pump parallel system, a Hall effect or current transformer-type current sensor is installed on the power supply circuit of its drive motor to monitor the current consumption of the motor in real time, thereby generating operating current data characterizing the load of each pump; at the same time, a photoelectric or magnetic induction speed sensor is installed on the rotating shaft or coupling of each pump to accurately measure the actual speed of the pump and generate speed data.

[0023] Furthermore, in order to understand the hydraulic conditions of the entire system, pressure sensors are installed at key locations such as the main outlet pipe or manifold of the pump station to collect the main pipeline pressure when the system delivers fluid to the outside, and obtain pressure data that reflects the overall working performance of the system.

[0024] All data collected by sensors, including sand concentration data, independent operating current and speed data for each pump, and system pressure data, are assigned precise timestamps and aggregated by a field data acquisition unit. This unit, using an industrial-grade wireless communication module and employing 4G, 5G, or other reliable wireless network protocols, periodically transmits the integrated structured data packets to a remote control center. The remote control center receives and parses these data packets, ultimately forming a synchronized, multi-dimensional time-series database—multi-source sensor data—providing a complete and correlated data foundation for subsequent collaborative analysis and intelligent control of operating conditions.

[0025] This method constructs a multi-source sensor data system that comprehensively reflects the operating status of pumping stations by physically deploying different types of sensors and synchronously integrating their collected data over time. This system overcomes the limitations of traditional single-monitoring methods, enabling independent monitoring of water quality, individual equipment status, and overall system performance. More importantly, through time synchronization and centralized transmission of data, it provides the necessary conditions for revealing the intrinsic causal relationship between changes in sediment concentration, fluctuations in single pump load, and system pressure response. This lays a solid foundation for subsequent cross-scenario collaborative control from the data acquisition perspective.

[0026] Step 2: Perform real-time collaborative analysis on multi-source sensor data. When the sand concentration data exceeds its normal range and the operating current data shows abnormal fluctuations consistent with the time of change of the sand concentration data, generate an impact confirmation result.

[0027] In a preferred embodiment, generating the impact confirmation result includes: extracting sand concentration data and operating current data from multi-source sensor data.

[0028] Anomaly markers for sand concentration are generated based on sand concentration data.

[0029] A current anomaly flag is generated based on the operating current data.

[0030] The temporal correlation between abnormal sand concentration indicators and abnormal current indicators is determined. When a temporal correlation exists between the two, an impact confirmation result is generated.

[0031] Specifically, after receiving continuous multi-source sensor data, the remote control center initiates a real-time collaborative analysis process. This process first separates two key types of information from the data stream: sediment concentration data from the inlet and operating current data from each pump.

[0032] For sediment concentration data, the system compares it in real time with a preset normal range, which is usually defined by an upper threshold. This threshold is determined based on the pump's design wear resistance and historical safe operating experience. Once the real-time sediment concentration data exceeds this upper threshold, the system immediately generates a Boolean-type sediment concentration anomaly flag, indicating that the current water quality is at risk.

[0033] Meanwhile, the system performs dynamic abnormal fluctuation analysis on the operating current data of each pump. This analysis is not a simple threshold judgment, but rather calculates the fluctuation index of the operating current within a short time window, such as the standard deviation or the deviation from a dynamic baseline (e.g., a moving average based on recent operating conditions). When the calculated fluctuation index exceeds a preset current fluctuation threshold, the system generates an independent current anomaly flag for the corresponding pump.

[0034] Finally, the system executes its core collaborative judgment logic, which correlates the abnormal sediment concentration flag with the abnormal current flags of each pump over time. The system checks for a time window in which the current anomaly flag is generated immediately after the sediment concentration anomaly flag. Only when both flags not only coexist but also satisfy this close temporal correlation does the system finally generate an impact confirmation result. This result is structured information that clearly indicates which pump's abnormal operation is directly caused by excessively high sediment concentration.

[0035] This method achieves a leap from phenomenon monitoring to cause diagnosis by conducting in-depth time-series correlation analysis between sediment concentration monitoring and pump operating current monitoring. It overcomes the ambiguity of isolated condition assessments in traditional methods, accurately distinguishing between pump load anomalies caused by elevated sediment concentration and similar phenomena caused by other factors such as mechanical failures or power grid fluctuations. This causal correlation-based collaborative analysis method generates impact confirmation results with extremely high confidence, providing a reliable basis for subsequent precise and effective protection and control measures, thereby significantly improving the intelligence level and response accuracy of the entire automated control system.

[0036] Step 3: Based on the impact confirmation results, generate a dynamic control instruction that includes adjusting the operating parameters of the affected pump and coordinating the control of the operating parameters of other pumps.

[0037] In a preferred embodiment, generating a dynamic control instruction based on the impact confirmation result, which includes adjusting the operating parameters of the affected pump and coordinating the control of the operating parameters of other pumps, includes: identifying the affected pump based on the impact confirmation result.

[0038] By combining the anomalies in sand concentration data and operating current data, adjustment instructions for the operating parameters of the affected pumps are generated.

[0039] Based on the overall operating status of the multi-pump parallel system, including the current total outlet pressure and the available margin of other pumps, coordinated control commands are generated for the other pumps.

[0040] During the generation of operating parameter adjustment instructions and coordinated control instructions, the operating current data is continuously monitored. When the operating current data of any pump approaches its overload limit, an overload risk indicator is generated.

[0041] Specifically, upon receiving the impact confirmation results, the remote control center immediately initiates the dynamic control command generation program. This program first analyzes the impact confirmation results, extracts the unique identifier of the affected pump, and thus identifies the specific target requiring protective intervention.

[0042] Next, the system quantifies the severity of the chemical condition based on real-time sand concentration data and the operating current data of the affected pumps. For example, the higher the sand concentration exceeds the normal range, or the greater the fluctuation in operating current or the greater the deviation from the normal baseline, the higher the risk level is determined. Based on this risk level, the system calculates a specific adjustment amount for operating parameters using a preset control strategy model, such as a lookup table or a multivariable function, and generates an operating parameter adjustment command for the affected pump. This command typically manifests as a precise percentage or numerical reduction in its target speed, aiming to decrease the relative velocity between the impeller and the sand particles, thereby reducing abrasive impact.

[0043] Simultaneously with generating the aforementioned protection commands, the system enters the collaborative control logic. To compensate for the flow loss caused by the reduced speed of the affected pumps, the system can predict the theoretical flow gap using pump performance curves or experimental data. Based on the overall operating status of the multi-pump parallel system, including the current total outlet pressure and the available margin of the remaining healthy pumps, the system allocates this flow gap to one or more unaffected pumps proportionally or using an optimization algorithm. For example, methods such as proportional allocation or allocation based on optimal efficiency can be used. The system generates collaborative control commands for these healthy pumps, instructing them to moderately increase their operating speed to ensure that the total output flow of the pumping station remains near the preset target value, thereby stabilizing the total system flow.

[0044] The current total outlet pressure refers to the actual pressure value measured on the main pipeline, such as the main outlet pipeline or the manifold, when the pumping station system is transporting fluid outward.

[0045] The available margin of the remaining healthy pumps refers to the remaining capacity of the pumps that are not currently affected by abnormal sand concentrations and can still safely increase flow or power under their existing operating conditions.

[0046] Throughout the entire instruction generation and calculation process, a parallel monitoring thread continuously monitors the real-time operating current data of all pumps. The system compares this data with a preset overload threshold that is below the trip protection value. Once it detects that the operating current data of any pump, whether it is a pump being decelerated or accelerated, rises and approaches this overload threshold, the system immediately marks the status of that pump and includes it in the overload monitoring range, preparing for possible overload protection control.

[0047] This method achieves a significant shift from passively protecting individual devices to actively maintaining the overall performance of the system by generating composite dynamic control commands that incorporate individual protection and system compensation. It not only effectively protects pumps affected by sediment concentration from rapid wear and tear, extending equipment lifespan, but also, through intelligent collaborative control, avoids the problem of reduced service capacity of the entire pumping station caused by adjustments to a single pump, ensuring the continuity and stability of system operation. Simultaneously, the proactive overload risk monitoring mechanism seamlessly integrates abnormal sediment concentration handling with overload protection, constructing a safety barrier against cascading failures and demonstrating comprehensive optimization control capabilities under multiple objectives and constraints.

[0048] Step 4: Execute the dynamic control command and obtain the operating current data, speed data and pressure data after the command is executed to obtain feedback data.

[0049] In a preferred embodiment, the step of executing the dynamic control command and obtaining the operating current data, speed data, and pressure data after the command execution to obtain feedback data includes: parsing the dynamic control command and separating the operating parameter adjustment command and the collaborative control command.

[0050] Issue and execute instructions to adjust operating parameters to the affected pumps.

[0051] Issue and execute coordinated control commands to other pumps.

[0052] After the command is executed, updated operating current data, speed data, and pressure data are collected to obtain feedback data.

[0053] Specifically, after generating dynamic control commands, the remote control center enters the command execution and feedback phase. First, the system parses the complex dynamic control commands, breaking them down into two independent but interconnected parts: commands to adjust the operating parameters of pumps affected by sand concentration, and commands for coordinated control of other healthy pumps in the system.

[0054] Subsequently, the system sends the parsed operating parameter adjustment commands to the corresponding field-programmable logic controllers (FPGAs) or frequency converters of the affected pumps via a secure remote communication link. Upon receiving the commands, the controller precisely adjusts the pump motor's drive frequency or controls the actuators of the electric valves to achieve the required reduction in speed or valve opening. Almost simultaneously, coordinated control commands are also distributed to the corresponding controllers of the remaining pumps, instructing them to moderately increase speed or adjust valve opening. After these control actions are executed, the system waits for a brief stabilization period to allow the entire pump station's hydraulic system to reach a new dynamic equilibrium.

[0055] Immediately following, the system initiates a new round of real-time data acquisition. Using previously deployed sensors, it captures and records the operating current data, actual operating speed data, and total outlet pressure data of each pump after the control action is completed. All this data, collected after the control action and bearing a new timestamp, is integrated to form a complete dataset, namely feedback data, which is used for subsequent evaluation of the control effect.

[0056] This method constructs a complete closed-loop control system by closely integrating central decision-making with on-site execution and immediately collecting the system status after execution. It goes beyond simply issuing commands; it translates control intentions into actual physical actions and quantifies the resulting effects in real time. The crucial step of generating feedback data provides objective evidence for system self-evaluation and subsequent iterative optimization, transforming the entire automated control process from a static, pre-set mode into a dynamic, verifiable, and self-correcting intelligent response system, thereby ensuring the effectiveness and accuracy of regulation.

[0057] In a further preferred embodiment, after generating the overload risk identifier, the method further includes: analyzing the cause of the overload by combining the speed data and pressure data, and generating an overload handling instruction.

[0058] Execute overload handling instructions to reduce the load on pumps at risk of overload.

[0059] Continuously acquire and evaluate feedback data after load reduction operations until the overload risk indicator is cleared.

[0060] Specifically, when the system generates parameter adjustment commands to address overload risks, it internally executes a refined closed-loop control process. This process begins with continuous dynamic monitoring of operating current data. The system does not wait for the current to reach a hard trip limit, but rather triggers intervention immediately when the current continues to rise and enters a preset warning zone close to the overload threshold. Once triggered, the system first identifies and confirms the unique identity of the pump experiencing the situation, namely the overloaded pump's identifier.

[0061] Upon receiving the flag, the system does not take immediate action but instead enters a diagnostic analysis phase. It retrieves real-time speed and system pressure data associated with the overloaded pump flag and performs multivariate causal analysis to determine the root cause of the overload. For example, the system might analyze whether abnormal sand concentration caused a performance degradation in other pumps, thus shifting the load to the current pump; or whether an abnormal increase in system outlet pressure passively increased the pump's load. Based on this diagnostic conclusion, the system generates a targeted overload handling instruction that specifies the exact control method and magnitude.

[0062] Subsequently, the system executes this overload handling command and sends it to the local controller of the overload pump via a remote communication link, instructing it to perform precise operations, such as reducing the inverter output frequency by a calculated value to reduce its speed, or driving the electric valve actuator to reduce its valve opening.

[0063] After the command is executed, the system does not end the process but enters a continuous verification state. It monitors the adjusted operating current and pressure data in real time at a higher frequency and compares them with the safety target. This process will continue until the monitored operating current data stably falls back to within the safe range, and the system confirms that the overload risk has been eliminated.

[0064] This method achieves intelligent and refined management of overload risk by implementing a complete closed-loop control system encompassing four stages: early warning, diagnosis, execution, and verification. It transforms traditional, precipitous overload protection into a smooth, proactive load regulation. By intervening before overload occurs and selecting the optimal control measures based on the root cause of the fault, this method not only effectively avoids system interruptions and hydraulic shocks caused by equipment tripping but also ensures the ultimate effectiveness of the control measures through continuous feedback verification, greatly enhancing the operational resilience and stability of the entire multi-pump parallel system.

[0065] Step 5: Evaluate the feedback data. When the feedback data indicates that the operating current is not within the preset safe operating range or the pressure is unstable, generate a parameter adjustment command and execute the command until the system returns to stable operation.

[0066] In a preferred embodiment, the evaluation feedback data, when the feedback data indicates that the operating current is not within a preset safe operating range or the pressure is unstable, generates a parameter adjustment instruction, including: extracting operating current data and pressure data from the feedback data.

[0067] Determine whether the operating current data of all pumps are within their preset safe operating range, and generate a current safety indicator.

[0068] Determine whether the pressure data meets the requirements for stable system operation and generate a pressure stability indicator.

[0069] When the current safety indicator and the pressure stability indicator are not simultaneously met, a parameter adjustment command is generated.

[0070] Specifically, upon receiving the feedback data, the remote control center initiates a control effect evaluation program. This program first extracts two key time series from the feedback dataset: the operating current data of all pumps and the pressure data of the system's total outlet.

[0071] For the operating current data, the system compares the real-time current value of each pump with its respective preset safety range. This safety range is an interval defined by an upper and lower limit; the upper limit is usually below the overload protection threshold, while the lower limit prevents the pump from running dry or operating inefficiently. Only when the operating current data of all pumps falls within this safety range will the system generate a Boolean current safety flag with a true value. If the current of any pump exceeds the range, the flag is false.

[0072] Simultaneously, the system performs stability analysis on the pressure data. This analysis aims to detect whether there are drastic, unexpected pressure fluctuations, which may indicate risks such as hydraulic instability or cavitation within the system. The analysis method can be to calculate the standard deviation or coefficient of variation of the pressure data within a short time window and compare it with a preset stability threshold. If the pressure fluctuation is less than the threshold, the system generates a Boolean-type pressure stability flag with a true value; otherwise, it is considered false.

[0073] Finally, the system performs a logical check: if and only if both the current safety flag and the pressure stability flag are not simultaneously true, it indicates that the previous adjustment failed to fully achieve the expected safety and stability goals. The system then generates a new parameter adjustment command. This command is based on iterative optimization of the current evaluation results; for example, if the current is still too high, the command might further reduce the speed of the relevant pump.

[0074] This method upgrades automated control from a single command execution to a closed-loop, self-correcting iterative optimization process by establishing an effectiveness evaluation mechanism based on multi-dimensional feedback data. This ensures the ultimate effectiveness of control behavior and the robustness of system operation. It no longer relies on idealized model predictions but verifies and continuously corrects control strategies through real-time evaluation of real-world responses. This mechanism can promptly identify and correct new problems caused by insufficient or excessive regulation, ensuring that the system can cope with initial risks such as abnormal sediment concentrations without introducing secondary risks such as overload or system instability. Ultimately, it guides the entire pumping station system to converge to a truly safe, stable, and efficient operating state.

[0075] In a further preferred embodiment, the system further includes: after the system resumes stable operation, continuously acquiring multi-source sensor data to form long-term monitoring data.

[0076] By comparing and analyzing long-term monitoring data with historical data before system recovery, changes in equipment performance caused by abnormal operating conditions are identified, and an equipment performance evaluation report is generated.

[0077] When an equipment performance assessment report indicates a potential risk of damage, equipment maintenance recommendations are generated.

[0078] Specifically, once the remote automated control system successfully handles an abnormal sediment concentration or overload risk event, and the pump station's operation returns to normal and stable, the system will switch to long-term monitoring and preventative maintenance mode. In this mode, the system will not stop data acquisition but will continuously and uninterruptedly record sediment concentration data from the sediment concentration detector, as well as operating current and speed data from each pump. This data, collected during normal operation after the event, is aggregated, tagged, and stored to form a long-term monitoring dataset with a long time span.

[0079] Subsequently, the system initiates an analysis module that retrieves two core datasets for comparative analysis: one is short-term historical data containing the complete process before, during, and after the most recent anomaly event; the other is newly generated long-term monitoring data. The analysis focuses on assessing whether the performance of critical pumping station equipment, especially pumps subjected to high-sediment-laden water flow impacts, has undergone irreversible changes. For example, the system compares whether there is a permanent, slight increase in the pump's operating current before and after the event, under the same speed and system pressure. This increase is often a direct reflection of decreased hydraulic efficiency due to wear on flow components such as the impeller and pump casing. When the system identifies, through this comparative analysis, that the performance degradation exceeds a preset maintenance threshold, it automatically generates a structured equipment maintenance recommendation. This recommendation clearly indicates the equipment number with potential damage risk, the specific quantitative indicators of performance degradation, and a preliminary diagnostic conclusion based on a wear model, such as "It is recommended to check the impeller leading edge wear."

[0080] Finally, this equipment inspection and repair council can trigger and guide professionals to conduct regular, targeted inspections, repairs, or replacements of severely worn parts on designated equipment through the alarm interface of the system workstation or directly to the mobile terminals of maintenance personnel.

[0081] This method combines short-term emergency response with long-term health management, achieving a paradigm shift from passively dealing with faults to proactively preventing failures. It utilizes each anomaly as a "stress test" of equipment performance, accurately identifying minute damage that is difficult to detect visually but accumulates over long-term operation and eventually leads to major failures. By generating data-driven, highly targeted equipment maintenance recommendations, it significantly improves the efficiency and effectiveness of maintenance work, avoiding the resource waste caused by indiscriminate periodic maintenance, and allowing for intervention at the optimal time to extend the effective service life of equipment at minimal cost, thereby ensuring the long-term reliability and economy of the entire pumping station system.

[0082] In a further preferred embodiment, the method further includes storing multi-source sensor data, impact confirmation results, and dynamic control commands as event records in a historical database.

[0083] By analyzing historical databases, we can identify the correlation patterns between abnormal sediment concentrations and overload occurrences, and generate an optimization model for control strategies.

[0084] Based on the control strategy optimization model, the internal parameters used to generate dynamic control commands are adjusted to generate an optimization strategy.

[0085] Specifically, after each automated control cycle, this method archives and stores the complete data chain of that event. Specifically, the system stores the multi-source sensor data at the time of the event trigger, the impact confirmation results generated by collaborative analysis, the issued dynamic control commands, the feedback data collected after execution, and any subsequent possible parameter adjustment commands as a complete record with a unique event identifier in a structured time-series database. This accumulated data forms a historical database over time.

[0086] Building upon this foundation, the system periodically, or after accumulating a certain number of events, activates a deep analysis engine. This engine applies data mining and statistical analysis techniques to process the historical database, aiming to discover potential patterns and optimize control parameters. For example, to optimize the sediment concentration warning threshold, the engine analyzes all records of events that affect the confirmation result and statistically analyzes the distribution of sediment concentration data in the short period before these events occur, thereby finding a critical value that can provide earlier and more accurate warnings of actual risks. Similarly, to optimize the overload warning threshold, the engine analyzes the dynamic characteristics of the operating current data that leads to overload, such as the rate of change and duration, to establish a more complex warning model, rather than relying solely on a static current upper limit. The new thresholds or model parameters derived from the analysis are automatically updated into the system's real-time control logic.

[0087] Furthermore, this analysis may reveal complex relationships between multiple variables, such as discovering that a specific range of sediment concentration combined with a particular pump is more likely to trigger overload. Based on such findings, the system can generate a comprehensive optimization strategy, which may include adjusting the default operating limits of certain pumps under specific seasonal or hydrological conditions, or developing a set of preventative control measures to intervene early under specific risk modes.

[0088] This method endows remote automation control systems with a self-evolving capability by constructing a complete closed loop from data acquisition and execution to feedback and learning optimization. The system is no longer a static rule executor, but an intelligent agent capable of learning and growing from its own experience. Each abnormal event becomes a valuable learning opportunity, enabling the system's early warning capabilities, control precision, and strategy rationality to continuously improve over time and with the accumulation of experience. This iterative optimization based on historical data achieves a fundamental shift in control logic from preset to adaptive, thus maintaining optimal operating performance and highest reliability under constantly changing environments and equipment conditions, realizing the long-term intelligent evolution of the entire control system.

[0089] The above content is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, and all such modifications and additions should fall within the protection scope of the present invention.

Claims

1. A method for remote automated control of a pump station, characterized in that, The method comprises the following steps: Step 1, obtaining the sediment concentration data of the pump station inlet, the running current data and the rotating speed data of each pump in the multi-pump parallel system, and the pressure data of the key positions of the pump station, to obtain multi-source sensor data; Step 2, real-time collaborative analysis of the multi-source sensor data, when the sediment concentration data exceeds its normal range and the running current data appears abnormal fluctuation consistent with the sediment concentration data change time, an influence confirmation result is generated; Step 3, based on the influence confirmation result, a dynamic control instruction containing the adjustment of the running parameters of the affected pump and the collaborative control of the running parameters of other pumps is generated; Step 4, executing the dynamic control instruction, and obtaining the running current data, the rotating speed data and the pressure data after the execution of the instruction, to obtain feedback data; Step 5, evaluating the feedback data, when the feedback data indicates that the running current is not in the preset safe working interval or the pressure is unstable, a parameter adjustment instruction is generated, and the instruction is executed until the system returns to stable operation; The generation of the influence confirmation result comprises: extracting the sediment concentration data and the running current data from the multi-source sensor data; generating a sediment concentration anomaly mark based on the sediment concentration data; generating a current anomaly mark based on the running current data; judging the time correlation of the sediment concentration anomaly mark and the current anomaly mark, and when there is time correlation, an influence confirmation result is generated; The generation of the dynamic control instruction containing the adjustment of the running parameters of the affected pump and the collaborative control of the running parameters of other pumps based on the influence confirmation result comprises: identifying the affected pump according to the influence confirmation result; generating a running parameter adjustment instruction for the affected pump in combination with the abnormal degree of the sediment concentration data and the running current data; generating a collaborative control instruction for other pumps according to the overall running state of the multi-pump parallel system, including the current total water outlet pressure and the available margin of other pumps; In the process of generating the running parameter adjustment instruction and the collaborative control instruction, the running current data is continuously monitored, and when the running current data of any pump approaches its overload limit value, an overload risk mark is generated; The execution of the dynamic control instruction and the obtaining of the running current data, the rotating speed data and the pressure data after the execution of the instruction to obtain feedback data comprise: analyzing the dynamic control instruction to separate the running parameter adjustment instruction and the collaborative control instruction; issuing and executing the running parameter adjustment instruction to the affected pump; issuing and executing the collaborative control instruction to other pumps; After the execution of the instruction, the updated running current data, the rotating speed data and the pressure data are collected to obtain feedback data.

2. A method for remote automated control of a pump station as claimed in claim 1, characterized in that, The obtaining of the multi-source sensor data comprises: collecting the sediment concentration data through the sediment concentration detector installed at the inlet of the pump station; collecting the running current data and the rotating speed data through the current sensor and the rotating speed sensor respectively installed on each pump in the multi-pump parallel system; collecting the pressure data through the pressure sensor arranged at the key positions of the pump station; collecting and transmitting the sediment concentration data, the running current data, the rotating speed data and the pressure data through the wireless communication module to obtain the multi-source sensor data.

3. The method for remote automation control applied to a pump station according to claim 1, characterized in that, After the generation of the overload risk mark, it further comprises: analyzing the overload reason in combination with the rotating speed data and the pressure data to generate an overload processing instruction; The overload processing instruction is executed to perform a load reduction operation on the pump at risk of overload; The feedback data after the load reduction operation is continuously acquired and evaluated until the overload risk indicator is cleared.

4. The method for remote automation control applied to a pump station according to claim 1, characterized in that, The evaluation of the feedback data generates a parameter adjustment instruction when the feedback data indicates that the operating current is not within the preset safe operating range or the pressure is unstable, including: Extracting operating current data and pressure data from the feedback data; Judging whether the operating current data of all pumps are within their preset safe operating range to generate a current safety flag; Judging whether the pressure data meets the system stable operation requirement to generate a pressure stability flag; Generating a parameter adjustment instruction when the current safety flag and the pressure stability flag are not met simultaneously.

5. The method for remote automated control of a pump station of claim 1, wherein, Further comprising: After the system resumes stable operation, continuously acquiring multi-source sensor data to form long-term monitoring data; Comparing and analyzing the long-term monitoring data with the historical data before the system resumes stable operation to identify the equipment performance changes caused by abnormal working conditions and generate an equipment performance evaluation report; Generating an equipment maintenance recommendation when the equipment performance evaluation report indicates a potential damage risk.

6. The method for remote automated control of a pump station of claim 1, wherein, Further comprising: Storing the multi-source sensor data, impact confirmation results, and dynamic control instructions as event records in a historical database; Analyzing the historical database to identify the correlation mode of the sand concentration anomaly and the overload occurrence to generate a control strategy optimization model; Adjusting the internal parameters used to generate the dynamic control instruction based on the control strategy optimization model to generate an optimized strategy.

Images