A smart pumping system for a waterworks

Abstract

This invention discloses a smart pumping system for a waterworks, specifically relating to the field of automatic control of waterworks pumping stations. It includes: a data acquisition module comprising an outflow water flow meter, an outflow water pressure meter, an inlet pressure meter, an outlet pressure meter, a power acquisition unit, and a pump status acquisition terminal, used to collect data on the pump start / stop status, operating frequency, inlet pressure, outlet pressure, outflow water volume, inlet area, outlet area, shaft power, power consumption during sampling cycle, actual outflow water volume, and actual outflow water pressure, outputting pumping data acquisition records; and a control execution module comprising a host computer, a PLC, a PID controller, a frequency converter, and an edge computing controller. Through responsibility identification, hysteresis stripping, and structural reconstruction, the system decomposes average efficiency inefficiency into short-time response lag and responsibility mismatch, reducing the misjudgment of surface deviations as pump tripping demands, and enabling pump group scheduling to more closely align with the actual source of mismatch while meeting water supply targets.

Description

Technical Field

[0001] This invention relates to the field of automatic control technology for waterworks pumping stations, and more specifically, to a smart pumping system for waterworks. Background Technology

[0002] In the control of secondary pumping stations in waterworks, the existing pumping systems are mainly aimed at ensuring the outflow or pressure of the water. The constant flow or constant pressure mode is set by the host computer, and the PLC combined with the PID controller performs the same frequency variable frequency regulation of multiple water pumps. During low-carbon operation, the pump is increased or decreased based on the efficiency of a single pump, the average efficiency of the pump group, the average outflow, or the energy consumption ratio. In the scenario of continuous water supply where urban water supply peaks and troughs frequently switch, secondary pump stations must not only complete real-time control on the edge computing side to reduce reliance on human experience and remote computing, but also ensure that scheduling actions do not cause fluctuations in the outflow water pressure or deviation of the total outflow water flow from the target. Under these conditions, judging whether to add or remove pumps based solely on the average efficiency or average outflow of the pump set will mix the overload of the main pump, the low contribution of the follower pumps, the high consumption and low contribution of a single pump, and the hysteresis response after the preceding frequency increase, frequency decrease, pump addition, and pump reduction into the same inefficient state. This manifests as the total factory pressure or flow rate meeting the standard, but some pumps are at high frequency and high consumption for a long time, the energy consumption ratio is not improved after adding pumps, the pressure recovery is delayed after reducing pumps, and the same pump set is repeatedly switched in adjacent time periods. This makes it impossible for low-carbon scheduling to distinguish between the superficial deviation that can be eliminated by frequency fine-tuning and the structural mismatch that requires reconstructing the pump set's responsibility relationship. The technical problem to be solved by this application is: how to identify the mismatch in the internal bearing relationship of the pump set and generate a pump set reconfiguration scheduling result that does not destroy the water supply target, under the premise that the outflow or pressure of the outflow water continuously meets the water supply target. Summary of the Invention

[0003] To overcome the aforementioned deficiencies in the prior art, embodiments of the present invention provide a smart pumping system for a waterworks. This system collects operational data of the effluent pumps and main pipeline at the edge computing side, calculates the energy efficiency and load mismatch value of a single pump, identifies the primary load pump, the follower load pump, and the abnormal load pump, and generates a pump group reconfiguration scheduling command after distinguishing between hysteresis inefficiency and structural inefficiency states by combining the response of preceding adjustment actions. This solves the problem mentioned in the background art of difficulty in identifying the load mismatch within the pump group and generating pump group reconfiguration scheduling results that do not disrupt the water supply target under the condition of continuous satisfaction of the water supply target.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a smart pumping system for a waterworks, comprising: The data acquisition module includes an outflow water flow meter, an outflow water pressure meter, an inlet pressure meter, an outlet pressure meter, a power acquisition unit, and a pump status acquisition terminal. It is used to collect the pump start-stop status, operating frequency, inlet pressure, outlet pressure, outflow water flow, inlet area, outlet area, shaft power, power consumption during sampling cycle, actual outflow water flow, and actual outflow water pressure, and outputs pumping acquisition records. The control execution module, including a host computer, PLC, PID controller, frequency converter and edge computing controller, is used to read the basic operating mode, the target value of the 24-hour control segment and the pumping acquisition record. In constant flow mode, the flow deviation is generated by subtracting the actual flow from the target outflow. In constant pressure mode, the pressure deviation is generated by subtracting the actual outflow from the target outflow pressure. The PID controller outputs multi-pump synchronous frequency adjustment commands according to the flow deviation or pressure deviation. The single-pump calculation module is used to read the pumping acquisition records, subtract the inlet pressure from the outlet pressure and add the pressure compensation value to generate the static pressure difference, convert the water flow rate, inlet area and outlet area into velocity head difference, convert the static pressure difference into static head and add it to the velocity head difference to generate the total head, and then generate water power, pump efficiency and single pump output energy efficiency ratio from the total head, water flow rate, shaft power, outlet pressure and sampling cycle power consumption, and output the single pump energy efficiency record. The responsibility identification module is used to read pumping acquisition records and single pump energy efficiency records, normalize and sum the operating frequency, shaft power and sampling cycle power consumption to generate the input responsibility value, normalize and sum the outflow flow ratio, pressure flow output ratio and pump efficiency ratio to generate the target contribution value, subtract the target contribution value from the input responsibility value to generate the responsibility mismatch value, generate the main responsibility pump according to the target contribution value with the largest value, generate the abnormal responsibility pump according to the positive and largest value of the responsibility mismatch value, generate the other operating pumps as follow-up responsibility pumps, and output the responsibility record; The reconfiguration scheduling module is used to read basic control records, single pump energy efficiency records, undertaking records, and previous action records. It sequentially compares the changes in total outflow water flow, outflow water pressure, operating frequency, shaft power, and single pump output energy efficiency ratio before and after the previous frequency increase, frequency decrease, pump increase, and pump decrease actions with the current undertaking mismatch value, generates a hysteresis inefficiency state or a structural inefficiency state, and outputs a frequency fine-tuning command in the hysteresis inefficiency state. In the structural inefficiency state, it generates pump increase commands, pump decrease commands, replacement commands, or undertaking order reordering commands according to the main undertaking pump, follow-up undertaking pump, and abnormal undertaking pump, and outputs the pumping scheduling record.

[0005] In a preferred embodiment, the data acquisition module performs the following steps: The pump status acquisition terminal establishes a pump number field for the water pump in automatic frequency conversion adjustment mode, and reads the start-stop status and operating frequency under the pump number field according to the sampling period. The inlet pressure gauge reads the inlet pressure under the same pump number field, the outlet pressure gauge reads the outlet pressure under the same pump number field, and the power acquisition unit reads the shaft power and power consumption during the sampling period under the same pump number field to generate a single pump status item. The inlet pipe diameter, outlet pipe diameter, and outlet flow rate of the water pump are read from the pump number field. The inlet pipe diameter is converted into the inlet area, the outlet pipe diameter is converted into the outlet area, and the outlet flow rate, inlet area, and outlet area are written into the single pump status item to generate a single pump data acquisition item. The outflow meter reads the actual outflow water flow rate according to the sampling cycle, and the outflow water pressure meter reads the actual outflow water pressure according to the same sampling cycle. The actual outflow water flow rate, actual outflow water pressure, sampling cycle number and all single pump acquisition items are combined to generate a pumping acquisition record.

[0006] In a preferred embodiment, the execution of the control execution module includes: The host computer reads the basic operating mode and the target value of the 24-hour control segment, and sends the basic operating mode, the current control segment number, the target outflow water flow rate in constant flow mode or the target outflow water pressure in constant pressure mode to the edge computing controller. The edge computing controller reads the pumping acquisition records. In constant flow mode, it generates a flow deviation by subtracting the actual outflow from the target outflow. In constant pressure mode, it generates a pressure deviation by subtracting the actual outflow from the target outflow pressure. The flow deviation or pressure deviation is written into the basic control record and then sent to the PLC. The PLC sends the flow or pressure deviation to the PID controller. The PID controller generates a frequency adjustment amount based on the flow or pressure deviation. The PLC sends a synchronous frequency adjustment command to the frequency converter based on the frequency adjustment amount. The frequency converter adjusts the water pump in automatic frequency conversion adjustment mode according to the synchronous frequency adjustment command and outputs the execution result of the multi-pump synchronous frequency adjustment command. If the actual outflow water flow or actual outflow water pressure is missing for two consecutive sampling cycles, the control execution module will stop generating new flow deviations or pressure deviations and stop outputting the automatic scheduling commands generated by the intelligent pumping system. The PLC will then switch to the default running state or the manual takeover state.

[0007] In a preferred embodiment, the execution of the single-pump calculation module includes: Read the inlet pressure, outlet pressure, water flow rate, inlet area, and outlet area under the same pump number field in the pumping acquisition record. Subtract the inlet pressure from the outlet pressure and add the pressure compensation value to generate the static pressure difference. Convert the water flow rate to the unit flow rate per second according to the sampling time unit. Divide the unit flow rate per second by the inlet area and outlet area respectively to generate the inlet velocity and outlet velocity. Subtract the square of the inlet velocity from the square of the outlet velocity and divide by the gravity head conversion term to generate the velocity head difference. The static head is generated by multiplying the static pressure difference by the pressure head conversion factor. The static head is generated by adding the static head and the velocity head difference. The total head is generated by multiplying the total head, the outlet flow rate and the gravitational acceleration and then converting the flow rate and time units to generate the water power. Read the shaft power and sampling cycle power consumption under the same pump number field, divide the water power by the shaft power to generate the pump efficiency, divide the product of the outlet pressure and the outlet flow rate by the sampling cycle power consumption to generate the single pump output energy efficiency ratio, and write the static pressure difference, velocity head difference, total head, water power, pump efficiency and single pump output energy efficiency ratio into the same pump number field to generate a single pump energy efficiency record.

[0008] In a preferred embodiment, the execution of the responsibility identification module includes: Read the operating frequency, shaft power, sampling cycle power consumption, outlet flow rate, outlet pressure, pump efficiency, and single pump output energy efficiency ratio under the same pump number field in the pump acquisition record and single pump energy efficiency record. Divide the operating frequency, shaft power, and sampling cycle power consumption by the sum of the corresponding fields of all operating pumps and then add them to generate the input responsibility value. Divide the product of outlet flow rate, outlet pressure and outlet flow rate, and pump efficiency by the sum of the corresponding fields of all operating pumps and then add them to generate the target contribution value. Subtract the target contribution value from the input responsibility value to generate the responsibility mismatch value. The operating frequency sequence, effluent flow rate sequence, shaft power sequence, and single pump output energy efficiency ratio sequence under the same pump number field are read sequentially according to the sampling period. Fourier decomposition is performed on the operating frequency sequence, and the frequency item with the first value in amplitude sorting is taken as the frequency modulation master item. The complex components of the effluent flow rate sequence, shaft power sequence, and single pump output energy efficiency ratio sequence are read at the frequency position of the frequency modulation master item, respectively, to obtain the flow rate frequency item, power frequency item, and energy efficiency frequency item. The phase order of the flow rate frequency item relative to the frequency modulation master item, the phase order of the power frequency item relative to the flow rate frequency item, and the amplitude increase / decrease direction of the energy efficiency frequency item relative to the power frequency item are written into the response item.

[0009] In a preferred embodiment, the execution of the responsibility identification module further includes: The operating pump ranked first by the target contribution value is generated as the main undertaking pump. The remaining operating pumps are arranged in descending order of their undertaking mismatch values ​​to generate a candidate pump chain. Starting from the first of the candidate pump chain, each candidate pump is read one by one. For each candidate pump read, the undertaking response item of the main undertaking pump is deducted from the undertaking response item of the candidate pump to obtain the net mismatch response item of the candidate pump. The net mismatch response item and the undertaking mismatch value are written together into the temporary abnormal pump item.

[0010] In a preferred embodiment, the execution of the responsibility identification module further includes: The temporary abnormal pump items are subjected to pruning search, including: randomly selecting two temporary abnormal pump items for pairwise comparison. If their frequency regulation main term is the same, the flow rate following frequency term phase order is the same, and the power following frequency term phase order is the same, the temporary abnormal pump item with the lower mismatch value ranking is deleted; if their frequency regulation main term is the same and the energy efficiency following frequency term amplitude increases or decreases in opposite directions, the temporary abnormal pump item with the higher mismatch value ranking is retained, and the other temporary abnormal pump item is added to the follower candidate list; if the input contribution value of the temporary abnormal pump item is not higher than the target contribution value, the temporary abnormal pump item is deleted from the candidate pump chain, resulting in a candidate abnormal chain.

[0011] In a preferred embodiment, the execution of the responsibility identification module further includes: The process of greedily selecting pumps from the candidate anomaly chain includes: first, reading the first candidate anomaly pump in the candidate anomaly chain as the current anomaly pump, and then sequentially reading the next candidate anomaly pump; if merging the next candidate anomaly pump with the current anomaly pump increases the total mismatch value of the anomaly pump set but does not increase the total target contribution value, then the next candidate anomaly pump is merged into the anomaly pump set; if the total mismatch value does not increase after merging, or the total target contribution value increases, then the next candidate anomaly pump is skipped; after traversal, the operating pump with the highest mismatch value in the anomaly pump set is generated as the anomaly bearing pump, and the operating pumps other than the main bearing pump and the anomaly bearing pump are generated as follower bearing pumps, and the main bearing pump, follower bearing pump, anomaly bearing pump, input bearing value, target contribution value, bearing mismatch value, bearing response item, and net mismatch response item are written into the bearing record.

[0012] In a preferred embodiment, the execution of the reconfiguration scheduling module includes: Read the basic control record, single pump energy efficiency record, load record, and preceding action record. Read the total outflow water flow, outflow water pressure, operating frequency, shaft power, and single pump output energy efficiency ratio before and after the preceding frequency increase, frequency decrease, pump increase, and pump decrease actions in sequence according to the action occurrence time. Generate flow action difference, pressure action difference, frequency action difference, power action difference, and energy efficiency action difference respectively. Compare the corresponding field difference between the current sampling period and the sampling period after the action with the change sequence of the load mismatch value of the abnormal load pump. If the flow or pressure still changes along the direction of the preceding action and the load mismatch value of the abnormal load pump is lower than the load mismatch value of the sampling period after the action, a hysteretic inefficient state is generated. If the flow or pressure has changed away from the direction of the preceding action and the load mismatch value of the abnormal load pump is not lower than the load mismatch value of the sampling period after the action, a structural inefficient state is generated.

[0013] In a preferred embodiment, the execution of the reconfiguration scheduling module further includes: In a hysteresis-inefficient state, the flow deviation or pressure deviation in the basic control record is read, the frequency adjustment value generated by the PID controller is written to the main pump and the follower pump, and the start / stop state of the abnormal pump remains unchanged, generating a frequency fine-tuning instruction; in a structurally inefficient state, if the input load value of the main pump is higher than the target contribution value and there is no running pump with a positive load mismatch value among the follower pumps, then the pump to be stopped is written to increase the pump instruction; if there is a running pump at the bottom of the target contribution value sorting among the follower pumps and the input load value is higher than the target contribution value, then the follower pump is written to decrease the pump instruction; if the abnormal pump has the first load mismatch value sorting and there is a stopped pump, then the abnormal pump is stopped and the stopped pump is started is written to replace the pump instruction; if there is no stopped pump, then the frequency allocation order of the main pump, the follower pump, and the abnormal pump is rearranged from high to low according to the target contribution value and a load order rearrangement instruction is generated. After executing the frequency fine-tuning command, pump increase command, pump decrease command, pump replacement command, or pump order reordering command, the actual outflow water flow rate, actual outflow water pressure, single pump output efficiency ratio, and pump mismatch value are reread. The system generates the post-execution flow deviation or post-execution pressure deviation, post-execution pump group efficiency ratio, and post-execution pump mismatch value. The system also writes the execution action, changes in flow deviation or pressure deviation before and after execution, changes in pump group efficiency ratio before and after execution, changes in pump mismatch value before and after execution, and a rollback flag into the pumping scheduling record.

[0014] The technical effects and advantages of this invention are as follows: By identifying the responsibility, stripping the hysteresis, and restructuring the structure, the inefficiency of average efficiency is broken down into short-term response lag and responsibility mismatch, reducing the situation of misjudging surface deviations as pump tripping demand, and making pump group scheduling closer to the actual source of mismatch while meeting water supply targets. By linking pump-side status, hydraulic data, electrical data, and main pipe feedback through the pump number field and sampling cycle number, a unified data source is formed, reducing the impact of multi-instrument sampling misalignment on static pressure difference, total head, single pump energy efficiency, and mismatch calculations. By continuously calculating static pressure difference, velocity head difference, total head, water power, pump efficiency and single pump output energy efficiency ratio, the relationship between single pump output and energy consumption is digitized, providing a traceable calculation basis for classifying main pumps, follow-up pumps and abnormal pumps. By recalculating the flow or pressure deviation, pump efficiency ratio, and mismatch value after execution, and writing the rollback flag, the pump addition, reduction, replacement of the assigned pump, and reordering of assigned pumps have a feedback correction process, which relatively alleviates water supply fluctuations and repeated switching after scheduling. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the system modules of the present invention. Detailed Implementation

[0016] 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.

[0017] Refer to the instruction manual appendix Figure 1 The present invention provides a smart pumping system for a waterworks, comprising: The data acquisition module includes an outflow water flow meter, an outflow water pressure meter, an inlet pressure meter, an outlet pressure meter, a power acquisition unit, and a pump status acquisition terminal. It is used to collect the pump start-stop status, operating frequency, inlet pressure, outlet pressure, outflow water flow, inlet area, outlet area, shaft power, power consumption during sampling cycle, actual outflow water flow, and actual outflow water pressure, and outputs pumping acquisition records. In this embodiment, the data acquisition module links the pump-side status data, pump-side hydraulic data, electrical energy data, and main pipe factory data in the secondary pumping station with the pump number field and the sampling period number to form a pumping acquisition record for the control execution module, single pump calculation module, undertaking identification module, and reconfiguration scheduling module to read. The pump number field is the unique number of the water-discharging pump in the system, and the sampling period number is generated by the edge computing controller according to the system clock. When the sampling times of each instrument are different within the same sampling period, the effective sampled value closest to the end time of the sampling period is selected and written into the corresponding field based on the end time of the sampling period. The pump status acquisition terminal first reads the effluent pumps whose control status is in automatic frequency conversion adjustment mode and establishes a pump number field for each effluent pump. Within each sampling period, the pump status acquisition terminal reads the start / stop status and operating frequency under the pump number field; the inlet pressure gauge reads the inlet pressure under the same pump number field; the outlet pressure gauge reads the outlet pressure under the same pump number field; and the power acquisition unit reads the motor input power and sampling period power increment under the same pump number field. The motor input power is converted to motor efficiency or frequency converter efficiency and written into the shaft power field, and the sampling period power increment is written into the sampling period power consumption field. After the start / stop status, operating frequency, inlet pressure, outlet pressure, shaft power, and sampling period power consumption are written into the same pump number field, a single pump status item is generated. If multiple sampling values ​​exist in the same field within the same sampling period, the sampling value closest to the end of the sampling period is selected. If the current sampling period lacks sampling values, the valid value from the previous sampling period is used and a reuse flag is written. If the start / stop status is "stopped," the pump number field is not written into the set of running pumps. The edge computing controller reads the inlet and outlet pipe diameters from the equipment parameter table based on the pump number field. When the pipe diameter unit is millimeters, it is first converted to meters, and then the inlet and outlet areas are generated according to the circular pipe area calculation rules. When a single pump flow collection point is set under the same pump number field, the outlet flow rate is read from the sampled value of the single pump flow collection point. When no single pump flow collection point is set under the same pump number field, the actual outlet water flow rate is used as the total, and the operating frequency ratio, shaft power ratio, and outlet pressure ratio of the operating pumps are calculated separately. The three ratios are added together and then divided by the sum of the three ratios of all operating pumps. The sum of the values ​​is used to obtain the allocation ratio, and the actual outflow water flow is multiplied by the allocation ratio to generate the single pump outflow rate. After processing, the outflow rate, inlet area, and outlet area are written into the single pump status item under the same pump number field to generate a single pump acquisition item. If the inlet pipe diameter or outlet pipe diameter is missing, the previous valid pipe diameter in the same pump number field of the equipment parameter table is read. If the previous valid pipe diameter does not exist, the inlet area, outlet area, and velocity head difference related fields of the pump number field are written with a missing mark. The subsequent single pump calculation module does not read the missing mark field to participate in the velocity head difference calculation. The outgoing water flow meter reads the actual outgoing water flow at the main outlet pipe according to the sampling cycle, and the outgoing water pressure meter reads the actual outgoing water pressure at the main outlet pipe according to the same sampling cycle. The edge computing controller merges the actual outgoing water flow, actual outgoing water pressure, sampling cycle number, and all single-pump acquisition items to generate a pumping acquisition record, and writes the pumping acquisition record to the local cache. If the actual outgoing water flow or actual outgoing water pressure is missing in the current sampling cycle, the valid value of the previous sampling cycle is used and written to the main pipe reuse mark. If the actual outgoing water flow or actual outgoing water pressure is missing for two consecutive sampling cycles, the pumping acquisition record retains the single-pump acquisition item, and the control execution module does not generate new flow deviation or pressure deviation based on the missing field. Through the above processing, the data acquisition module writes the single pump status item, single pump acquisition item, and main pipe feedback field into the same pumping acquisition record, so that the single pump energy efficiency calculation and mismatch judgment have the same sampling period basis. In practical applications, when the three effluent pumps are in automatic frequency conversion regulation mode, the data acquisition module establishes three pump number fields respectively. The pump status acquisition end, inlet pressure gauge, outlet pressure gauge, and power acquisition unit form three sets of single pump status items. The edge computing controller then reads the inlet pipe diameter, outlet pipe diameter, and effluent flow rate of the three effluent pumps to form three sets of single pump acquisition items. After the effluent flow meter and effluent pressure gauge write the actual effluent flow rate, actual effluent pressure, and sampling period number, they are merged to generate a pumping acquisition record for subsequent calculation of static pressure difference, velocity head difference, basic control deviation, and mismatch value.

[0018] The control execution module, including a host computer, PLC, PID controller, frequency converter and edge computing controller, is used to read the basic operating mode, the target value of the 24-hour control segment and the pumping acquisition record. In constant flow mode, the flow deviation is generated by subtracting the actual flow from the target outflow. In constant pressure mode, the pressure deviation is generated by subtracting the actual outflow from the target outflow pressure. The PID controller outputs multi-pump synchronous frequency adjustment commands according to the flow deviation or pressure deviation. In this embodiment, the control execution module is used to convert the operating target in the host computer into a control task that the edge computing controller can read, and to convert the main pipe feedback value in the pump acquisition record into a frequency modulation input that the PLC and PID controller can execute. The basic operating modes include constant flow mode and constant pressure mode. The constant flow mode uses the actual outflow water flow rate following the target outflow water flow rate as the control target, and the constant pressure mode uses the actual outflow water pressure following the target outflow water pressure as the control target. The target value of the 24-hour control segment is called by the host computer according to the system clock, and the edge computing controller only reads the target value that matches the basic operating mode under the current control segment number. The implementation process includes the following steps: The host computer is used to determine the current control task, ensuring that the edge computing controller only receives target fields valid under the current basic operating mode. During execution, the host computer reads the system clock, generates the current control segment number based on the hour of the system clock, and reads the target outflow water flow or target outflow water pressure corresponding to the current control segment number from the 24-hour control segment target values. When the basic operating mode is constant flow mode, the host computer sends the basic operating mode, the current control segment number, and the target outflow water flow to the edge computing controller. When the basic operating mode is constant pressure mode, the host computer sends the basic operating mode, the current control segment number, and the target outflow water pressure to the edge computing controller. When the basic operating mode is switched, the edge computing controller uses the next sampling period after the switch as the first control period of the new basic operating mode. If the target outflow water flow or target outflow water pressure is missing under the current control segment number, the target value of the previous control segment under the same basic operating mode is used and a target reuse flag is written. Among them, the target values ​​of the 24-hour control segment are generated by the host computer based on the historical daily water consumption curve, the average water consumption of the same hour over multiple consecutive days, and the daily scheduling plan. In constant flow mode, 24 sets of target outflow water flow are formed, and in constant pressure mode, 24 sets of target outflow water pressure are formed. They are written into the control segment target table according to the hour number for the edge computing controller to call. The edge computing controller is used to convert the main pipe feedback value into a basic control deviation, providing a clear input source for subsequent frequency adjustment actions. During execution, the edge computing controller reads the pump acquisition record output by the data acquisition module and reads the corresponding fields according to the basic operating mode. In constant flow mode, the edge computing controller generates the flow deviation by subtracting the actual outflow rate from the target outflow rate. In constant pressure mode, the edge computing controller generates the pressure deviation by subtracting the actual outflow rate from the target outflow rate. After completing the calculation, the edge computing controller generates a basic control record, which includes the sampling period number, basic operating mode, current control segment number, target outflow rate, actual outflow rate, flow deviation, target outflow rate, actual outflow rate, and pressure deviation. In constant flow mode, the pressure field retains the sampled value but is not used as the PID controller input; in constant pressure mode, the flow field retains the sampled value but is not used as the PID controller input. If the pump acquisition record has a main pipe reuse mark, the edge computing controller still writes it into the basic control record but does not generate a new flow deviation or pressure deviation, and continues to send the frequency adjustment amount of the previous sampling period to the PLC. The PLC is used to convert basic control deviations into frequency adjustment commands that the frequency converter can execute, causing the water pumps in automatic frequency conversion regulation mode and in the start / stop state to operate in the same frequency adjustment direction. During execution, the PLC receives the basic control records sent by the edge computing controller and sends the flow or pressure deviation to the PID controller. The PID controller generates a frequency adjustment value based on the flow or pressure deviation. When the frequency adjustment value is positive, the PLC sends a same-frequency adjustment command in the frequency increase direction to the frequency converter; when the frequency adjustment value is negative, the PLC sends a same-frequency adjustment command in the frequency decrease direction to the frequency converter. When the frequency adjustment amount is zero, the PLC sends a frequency-matching adjustment command to the frequency converter to maintain the current operating frequency. After receiving the frequency-matching adjustment command, the frequency converter only adjusts the water pumps whose start / stop status is running and whose control status is in automatic frequency conversion adjustment mode. Water pumps that are stopped, in manual mode, under maintenance, or fault-locked do not receive the frequency-matching adjustment command. After the frequency converter executes the command, it returns the execution status to the PLC. The PLC then sends the execution status back to the edge computing controller. The edge computing controller writes the frequency adjustment amount, the frequency-matching adjustment command, and the execution status into the basic control record, forming the execution result of the multi-pump frequency-matching adjustment command. Through the above processing, the control execution module writes the time-sharing target from the host computer, the main pipe feedback value from the pump acquisition record, the frequency adjustment amount output by the PID controller, and the inverter execution status into the same basic control record, enabling subsequent single-pump calculation, responsibility identification, and reconfiguration scheduling to read the executed basic control results. In practical applications: when the secondary pumping station is in constant flow mode and the target outflow rate corresponding to the current control segment number is already stored in the host computer, the host computer sends the target outflow rate to the edge computing controller. The edge computing controller reads the actual outflow rate from the pump acquisition record and calculates the flow deviation. The PLC sends the flow deviation to the PID controller to obtain the frequency adjustment amount. The inverter performs the same-frequency increase, same-frequency decrease, or frequency hold on the running automatic variable frequency water pump according to the frequency adjustment amount and writes the execution status back to the basic control record.

[0019] The single-pump calculation module is used to read the pumping acquisition records, subtract the inlet pressure from the outlet pressure and add the pressure compensation value to generate the static pressure difference, convert the water flow rate, inlet area and outlet area into velocity head difference, convert the static pressure difference into static head and add it to the velocity head difference to generate the total head, and then generate water power, pump efficiency and single pump output energy efficiency ratio from the total head, water flow rate, shaft power, outlet pressure and sampling cycle power consumption, and output the single pump energy efficiency record. In this embodiment, the single-pump calculation module converts the single-pump pressure, flow rate, pipe diameter area, and electrical energy fields in the pumping acquisition records into single-pump energy efficiency records, enabling the subsequent identification module to read the hydraulic output, energy consumption input, and energy efficiency results based on the same pump number field. The pressure compensation value, pressure head conversion factor, gravitational acceleration, and gravity head conversion item are read from the calculation parameter table by the edge computing controller. The pressure head conversion factor is determined according to the acquisition units of the inlet and outlet pressures, and the gravity head conversion item is generated from the gravitational acceleration relationship required for velocity head calculation. This implementation process includes the following steps: The single-pump calculation module first generates static pressure difference and velocity head difference to separate pressure changes and flow velocity changes and write them separately into the same pump number field. During execution, the edge computing controller reads the inlet pressure, outlet pressure, effluent flow rate, inlet area, and outlet area under the same pump number field from the pump acquisition record. It subtracts the inlet pressure from the outlet pressure and adds a pressure compensation value to generate the static pressure difference. The pressure compensation value is read from the calculation parameter table according to the pump number field. It is only used to prevent the subsequent total head from entering the zero value boundary when the difference between the inlet pressure and the outlet pressure is close to zero, and does not participate in the effluent water treatment process. Pressure control; the edge computing controller then converts the outlet flow rate into a flow rate per second based on the sampling time unit, divides the flow rate per second by the inlet area to generate the inlet velocity, divides the flow rate per second by the outlet area to generate the outlet velocity, subtracts the square of the inlet velocity from the square of the outlet velocity, and divides the result by the gravity head conversion term to generate the velocity head difference; when the outlet velocity is lower than the inlet velocity, the velocity head difference is written as a negative value and participates in the subsequent total head calculation; when the inlet area or outlet area has a missing mark, the velocity head difference is not calculated and is written as a missing mark; The single-pump calculation module generates the total head and water power to convert the hydraulic output of the single pump into the power field required for energy efficiency calculation. During execution, the edge computing controller multiplies the static pressure difference by the pressure head conversion factor to generate the static pressure head. The pressure head conversion factor is determined by the conversion relationship between the pressure acquisition unit and the head unit and is written into the calculation parameter table along with the pressure acquisition unit. The edge computing controller adds the static pressure head and the velocity head difference to generate the total head. When the velocity head difference has a missing marker, the edge computing controller only generates the total head using the static pressure head and writes the velocity missing marker into the total head field. After completing the total head calculation, the edge computing controller multiplies the total head, the outlet flow rate, and the gravitational acceleration, and converts the outlet flow rate into a time unit to generate the water power. When the outlet flow rate is already in units of flow per second, the time unit conversion is not performed again. When the outlet flow rate is in units of flow per hour, it is first converted into units of flow per second before participating in the water power calculation. The single-pump calculation module ultimately generates the pump efficiency and single-pump output energy efficiency ratio to distinguish between the pump's own efficiency and the pressure-flow output capacity formed by unit power consumption. During execution, the edge computing controller reads the shaft power and sampling period power consumption under the same pump number field, and divides the water power by the shaft power to generate the pump efficiency. When the shaft power is zero, has a missing flag, or is in the shutdown state, the pump efficiency is not generated, and the pump efficiency is written to an invalid flag. The edge computing controller multiplies the outlet pressure and the outlet flow rate to obtain the pressure-flow output, and then divides the pressure-flow output by the sampling period power consumption to generate the single-pump output energy efficiency ratio. When the sampling period power consumption is zero or has a missing flag, the single-pump output energy efficiency ratio is not generated, and the single-pump output energy efficiency ratio is written to an invalid flag. After processing, the edge computing controller writes the static pressure difference, velocity-head difference, total head, water power, pump efficiency, and single-pump output energy efficiency ratio into the same pump number field to generate a single-pump energy efficiency record for the identification module to read. Through the above processing, the single-pump calculation module converts static pressure difference, velocity head difference, total head, water power, pump efficiency, and single-pump output energy efficiency ratio into readable fields linked to the pump number field. It also sets processing rules for pressure compensation, unit conversion, missing area, zero power value, and zero power consumption value to prevent invalid data from entering the responsibility identification module. In practical applications: after the inlet pressure, outlet pressure, flow rate, inlet area, and outlet area of ​​a running pump are written into the pumping acquisition record, the single-pump calculation module first generates static pressure difference and velocity head difference, then generates total head and water power, and subsequently combines shaft power and sampling cycle power consumption to generate pump efficiency and single-pump output energy efficiency ratio. The single-pump energy efficiency record formed under the same pump number field is read by the responsibility identification module and used to calculate the target contribution value, input responsibility value, and responsibility mismatch value.

[0020] The responsibility identification module is used to read pumping acquisition records and single pump energy efficiency records, normalize and sum the operating frequency, shaft power and sampling cycle power consumption to generate the input responsibility value, normalize and sum the outflow flow ratio, pressure flow output ratio and pump efficiency ratio to generate the target contribution value, subtract the target contribution value from the input responsibility value to generate the responsibility mismatch value, generate the main responsibility pump according to the target contribution value with the largest value, generate the abnormal responsibility pump according to the positive and largest value of the responsibility mismatch value, generate the other operating pumps as follow-up responsibility pumps, and output the responsibility record; In this embodiment, the responsibility identification module is used to separately calculate the consumption-side responsibility and water supply-side contribution of the operating pumps, based on the water supply target already adjusted by the control execution module, and to identify the main responsibility pump, abnormal responsibility pumps, and follow-up responsibility pumps by combining the frequency response relationship in the sampling sequence. The responsibility identification module is executed by the edge computing controller, and the operating pumps are the water pumps whose start / stop state is running and whose control state is in automatic frequency conversion adjustment mode. The responsibility record is written after the candidate abnormal chain and abnormal pump set processing is completed. The responsibility response item, net mismatch response item, temporary abnormal pump item, and candidate abnormal chain are temporarily stored as intermediate fields in the edge computing controller cache area. The execution of this embodiment includes the following: The responsibility identification module first calculates the input responsibility value, target contribution value, and responsibility mismatch value to distinguish the difference between the responsibility on the consumption side and the contribution on the water supply side of the operating pump. During execution, the edge computing controller reads the operating frequency, shaft power, power consumption during the sampling cycle, outlet flow rate, outlet pressure, pump efficiency, and single pump output energy efficiency ratio under the same pump number field in the pump acquisition record and single pump energy efficiency record. The operating frequency ratio is calculated by dividing the operating frequency of a single operating pump by the sum of the operating frequencies of all operating pumps; the shaft power ratio is calculated by dividing the shaft power of a single operating pump by the sum of the shaft power of all operating pumps; and the power consumption ratio is calculated by dividing the power consumption during the sampling cycle of a single operating pump by the sum of the power consumption during the sampling cycle of all operating pumps. The input contribution value is generated by directly adding the proportions of each item without assigning additional weights. The proportion of water flow rate is calculated by dividing the water flow rate of a single operating pump by the sum of the water flow rates of all operating pumps. The proportion of pressure and flow output is calculated by multiplying the outlet pressure of a single operating pump by the water flow rate and dividing the sum of the pressure and flow output of all operating pumps. The proportion of pump efficiency is calculated by dividing the pump efficiency of a single operating pump by the sum of the pump efficiencies of all operating pumps. The three proportions are directly added to generate the target contribution value. The input contribution value is subtracted from the target contribution value to generate the contribution mismatch value. If any denominator is zero, the corresponding proportion is written as zero and marked with a zero denominator. Pump efficiency and single pump output energy efficiency ratio with invalid marks are not included in the calculation of the sum of the corresponding fields. The responsibility identification module then extracts the responsibility response item from the sampling sequence to determine the order of flow, power, and energy efficiency responses of the operating pump after frequency adjustment. During execution, the edge computing controller reads the operating frequency sequence, outlet flow sequence, shaft power sequence, and single pump output energy efficiency ratio sequence under the same pump number field according to the sampling period number. The sequence length is read from the frequency domain calculation parameter table of the edge computing controller. The frequency domain calculation parameter table is set according to the number of Fourier decomposition points. If the historical sampling period is insufficient, Fourier decomposition is temporarily not performed and a sequence insufficiency mark is written. The edge computing controller performs Fourier decomposition on the operating frequency sequence and reads the frequency item with the first value in the amplitude sort as the frequency adjustment main item. The frequency adjustment main item includes frequency position, amplitude, and other parameters. Values ​​and phases; the edge computing controller reads the complex components of the outflow rate sequence, shaft power sequence, and single pump output energy efficiency ratio sequence at the frequency position of the frequency modulation master term, and generates flow rate frequency-following term, power frequency-following term, and energy efficiency frequency-following term, respectively; the phase order is generated according to the phase sequence of the frequency modulation master term, flow rate frequency-following term, and power frequency-following term within the same sampling window; the increase or decrease direction of the amplitude of the energy efficiency frequency-following term is generated by comparing the amplitude of the energy efficiency frequency-following term in the current sampling window with the amplitude of the energy efficiency frequency-following term in the previous sampling window. When the amplitude of the current sampling window is greater than that of the previous sampling window, it is written as rising; when the amplitude of the current sampling window is less than that of the previous sampling window, it is written as falling; when the two are equal, it is written as flat; the above results are written into the response term; The responsibility identification module then generates a primary responsible pump and a temporary abnormal pump item. This is used to take the operating pump with the highest target contribution as a response reference, while retaining the mismatch characteristics of the remaining operating pumps that distinguish them from the primary responsible pump. During execution, the edge computing controller sorts the target contribution values ​​in descending order, and the operating pump at the top of the sorted list becomes the primary responsible pump. If the target contribution values ​​are the same, they are selected in ascending order of input responsibility values; if still the same, they are selected according to the pump number field order. The remaining operating pumps are sorted in descending order of responsibility mismatch values ​​to generate a candidate pump chain. The edge computing controller reads the candidate pumps one by one from the first in the candidate pump chain, deducting the responsibility response item of the primary responsible pump from the candidate pump's responsibility response item. The deduction method is to delete the frequency position, phase sequence, and amplitude increase / decrease direction fields of the candidate pump's responsibility response item that are the same as those of the primary responsible pump, retaining the remaining response fields to generate a net mismatch response item. If there are no remaining response fields after deduction, the net mismatch response item is written with a null value. The edge computing controller writes the net mismatch response item and the responsibility mismatch value together into the temporary abnormal pump item for pruning search reading. The responsibility identification module performs a pruning search on temporary abnormal pump items to delete those that repeatedly point to the same source of response mismatch and to add operating pumps with opposite energy efficiency changes to the follower candidate range. During execution, the edge computing controller randomly selects two temporary abnormal pump items for pairwise comparison. If the frequency positions of their frequency modulation master items, the phase order of their flow rate frequency-following items, and the phase order of their power frequency-following items are the same, the edge computing controller deletes the temporary abnormal pump item with the lower responsibility mismatch value. If the responsibility mismatch values ​​are the same, the temporary abnormal pump item with the higher single-pump output energy efficiency ratio is retained. If the single-pump output energy efficiency ratios are still the same, the pump number is retained. When two temporary abnormal pump items are ranked first in the segment order, and their frequency modulation main terms have the same frequency position and their energy efficiency frequency-dependent terms increase or decrease in opposite directions, the edge computing controller retains the temporary abnormal pump item ranked first in terms of bearing mismatch value and writes the other temporary abnormal pump item into the follower candidate list; when the input bearing value of the temporary abnormal pump item is not higher than the target contribution value, the edge computing controller removes the temporary abnormal pump item from the candidate pump chain; after completing the pairwise comparison, the remaining temporary abnormal pump items are arranged in descending order of bearing mismatch value to generate a candidate abnormal chain; the follower candidate list does not participate in the generation of abnormal bearing pumps, and is written into the follower bearing pump range first when generating follower bearing pumps; The responsibility identification module finally performs a greedy pump selection on the candidate anomaly chain to extract the operating pumps in the responsibility mismatch set without increasing the total target contribution. During execution, when the candidate anomaly chain is empty, the edge computing controller does not generate an abnormal responsibility pump and generates follow responsibility pumps for all operating pumps except the main responsibility pump. When the candidate anomaly chain is not empty, the edge computing controller reads the first candidate anomaly pump in the candidate anomaly chain as the current anomaly pump, initializes the anomaly pump set with the current anomaly pump, and then reads the next candidate anomaly pump in sequence. The edge computing controller uses the sum of the responsibility mismatch values ​​and the sum of the target contribution values ​​of all operating pumps in the anomaly pump set before merging as the benchmark, and the sum of the responsibility mismatch values ​​and the sum of the target contribution values ​​after adding the next candidate anomaly pump as the merged result. If the total mismatch value after merging is greater than the total mismatch value before merging and the total target contribution value after merging is not greater than the total target contribution value before merging, the next candidate abnormal pump will be merged into the abnormal pump set. If the total mismatch value after merging is not greater than the total mismatch value before merging, or the total target contribution value after merging is greater than the total target contribution value before merging, the next candidate abnormal pump will be skipped. After traversal, the edge computing controller will generate the operating pump with the highest mismatch value in the abnormal pump set as the abnormal bearing pump, and generate the operating pumps other than the main bearing pump and the abnormal bearing pump as the follower bearing pump. The main bearing pump, the follower bearing pump, the abnormal bearing pump, the input bearing value, the target contribution value, the bearing mismatch value, the bearing response item, and the net mismatch response item will be written into the bearing record. Through the above processing, the responsibility identification module does not directly determine the cause of inefficiency based on the average efficiency of the pump group. Instead, it first calculates the input responsibility value and target contribution value of each operating pump, then isolates the response fields identical to the main responsibility pump through frequency domain response relationships, and uses pruning search and greedy pump selection to identify abnormal responsibility pumps. This allows the subsequent reconfiguration and scheduling module to clearly identify the main responsibility pump, following responsibility pumps, and abnormal responsibility pumps. In practical applications: when three operating pumps participate in constant pressure water supply simultaneously, the target contribution value of the first operating pump is ranked first and it is used to generate the main responsibility pump; the operating frequency of the second operating pump... The proportion of the pump is high, but the proportion of the outflow rate, the proportion of the pressure and flow output, and the proportion of the pump efficiency are low, and the mismatch value is positive. The edge computing controller further reads the frequency regulation main item, flow rate frequency item, power frequency item and energy efficiency frequency item of the second running pump. If the net mismatch response item still retains the field that the power response precedes the flow response and the amplitude of the energy efficiency frequency item decreases, after pruning search and greedy pump selection, the second running pump is written as the abnormal undertaking pump and the third running pump is written as the following undertaking pump. The undertaking record is used by the reconfiguration scheduling module to determine whether to perform replacement of undertaking pump or reorder of undertaking order.

[0021] The reconfiguration scheduling module is used to read basic control records, single pump energy efficiency records, responsibility records, and previous action records. It sequentially compares the changes in total outflow water flow, outflow water pressure, operating frequency, shaft power, and single pump output energy efficiency ratio before and after previous frequency increase, frequency decrease, pump increase, and pump decrease actions with the current responsibility mismatch value. It generates a hysteresis inefficiency state or a structural inefficiency state. In the hysteresis inefficiency state, it outputs a frequency fine-tuning command. In the structural inefficiency state, it generates pump increase commands, pump decrease commands, replacement commands, or responsibility order rearrangement commands according to the main responsible pump, follow-up responsible pump, and abnormal responsible pump. It outputs the pumping scheduling record. In this embodiment, the reconfiguration scheduling module is used to break down inefficient states into short-term inefficiency caused by action lag and structural inefficiency caused by mismatch in responsibility relationships. Based on the breakdown results, it selects frequency fine-tuning, pump increase, pump decrease, replacement of responsible pumps, or rearrangement of responsibility order. The preceding action record is generated by extracting the frequency increase, frequency decrease, pump increase, and pump decrease actions already executed in the pumping scheduling record, and includes at least the action number, action type, action target pump number, action occurrence time, total outflow water before action, total outflow water after action, outflow water pressure before action, outflow water pressure after action, operating frequency before action, operating frequency after action, shaft power before action, shaft power after action, single pump output energy efficiency ratio before action, and single pump output energy efficiency ratio after action. The pre-action field takes the field value of the previous sampling period before the action command was issued, and the post-action field takes the field value of the first sampling period after the action command was executed. The implementation process includes the following steps: The reconfiguration scheduling module first identifies the sources of inefficiency to prevent the adjustment response still in the release process from being misjudged as a pump group structural error. The edge computing controller reads the basic control record, single pump energy efficiency record, load record, and previous action record. It sequentially reads the pre-action and post-action fields of the preceding frequency increase, frequency decrease, pump increase, and pump decrease actions according to the action occurrence time. It generates flow action difference, pressure action difference, frequency action difference, power action difference, and energy efficiency action difference by subtracting the pre-action field value from the post-action field value. Then, it generates the current flow difference, current pressure difference, current frequency difference, current power difference, and current energy efficiency difference by subtracting the post-action field value from the current sampling period field value. If the flow action difference has the same sign as the current flow difference, or the pressure action difference has the same sign as the current pressure difference, it is determined that the actual effluent flow rate or effluent pressure is still changing along the direction of the preceding action. If the flow action difference has a different sign than the current flow difference and the pressure difference is different... If the sign of the action difference is different from the current pressure difference, it is determined that the actual outflow water flow and pressure have deviated from the direction of the preceding action. Fields with an action difference of zero are not involved in the direction determination. The edge computing controller then reads the load mismatch value of the abnormal pump in the current sampling period and the load mismatch value in the sampling period after the action. If the actual outflow water flow or pressure still changes along the direction of the preceding action, and the current load mismatch value of the abnormal pump is lower than the load mismatch value in the sampling period after the action, a hysteretic inefficient state is generated. If the actual outflow water flow and pressure deviate from the direction of the preceding action, and the current load mismatch value of the abnormal pump is not lower than the load mismatch value in the sampling period after the action, a structural inefficient state is generated. When the preceding action record is empty, a hysteretic inefficient state is not generated, and the load record is read directly. If an abnormal pump exists, a structural inefficient state is generated; if no abnormal pump exists, no new scheduling instruction is generated. The reconfigured scheduling module then generates scheduling instructions based on the inefficient state, linking the adjustment actions with the mismatch type of the undertaking relationship. In the hysteresis-induced inefficient state, the edge computing controller reads the flow deviation or pressure deviation from the basic control record, writes the frequency adjustment value generated by the PID controller into the main undertaking pump and the follower undertaking pump, and keeps the abnormal undertaking pump's start / stop status and the operating frequency of the previous sampling period unchanged, generating a frequency fine-tuning instruction. In the structural inefficiency state, the edge computing controller first determines the replacement undertaking pump instruction: if an abnormal undertaking pump exists and a stopped pump exists, the stopped pump with the highest single-pump output energy efficiency ratio in the most recent ranking is selected as the replacement pump; if the stopped pump lacks historical single-pump output energy efficiency ratios, the stopped pump with the lowest cumulative running time is selected; if they are still tied, they are selected according to the pump number field order, and the abnormal undertaking pump's shutdown and the replacement pump's start are written into the replacement undertaking pump instruction; if no replacement undertaking pump instruction is generated, then the pump reduction instruction is determined: among the follower undertaking pumps, there is an operating pump with the lowest target contribution value and an input undertaking value higher than the target contribution value. When the target contribution value is the same, the pump with the lowest target contribution value is written into the pump reduction command; when the target contribution values ​​are the same, the pump with the highest input contribution value is selected, and if they are still the same, the pump with the lowest single pump output energy efficiency ratio is selected; when no pump reduction command is generated, the pump increase command is then determined: when the input contribution value of the main pump is higher than the target contribution value and there is no operating pump with a positive contribution mismatch value among the follower pumps, the pump to be stopped is written into the pump increase command, and the pump to be stopped selection rule is the same as the pump to be replaced selection rule; when no pump replacement command, pump reduction command, or pump increase command is generated, the edge computing controller arranges the frequency allocation order of the main pump, follower pumps, and abnormal pumps according to the target contribution value from high to low, and writes the frequency adjustment amount output by the PID controller into the operating pump with the highest target contribution value, and places the abnormal pump at the end of the frequency allocation order, and generates a contribution order reordering command; the pump to be stopped only includes the water pump with the start / stop status being stopped, the control status being automatic frequency conversion adjustment, and not in the fault lockout or maintenance status; The reconstructed scheduling module finally writes back and verifies the scheduling results, ensuring that the pumping scheduling record can serve as the source of the preceding action record for the next sampling cycle. After the execution of frequency fine-tuning instructions, pump increase instructions, pump decrease instructions, replacement of assigned pump instructions, or assignment order rearrangement instructions, the edge computing controller rereads the actual outflow water flow, actual outflow water pressure, single pump output energy efficiency ratio, and assignment mismatch value. In constant flow mode, the flow deviation after execution is generated by subtracting the actual outflow water flow after execution from the target outflow water flow; in constant pressure mode, the pressure deviation after execution is generated by subtracting the actual outflow water pressure after execution from the target outflow water pressure. The energy efficiency ratio of the pump group after execution is generated by dividing the sum of the products of the outlet pressure and outflow water flow of all operating pumps by the sum of the power consumption of all operating pumps during the sampling cycle. When the sum of the power consumption of all operating pumps during the sampling cycle is zero, the energy efficiency ratio of the pump group after execution is written to zero and marked with a zero denominator. The assignment mismatch value after execution is read from the assignment identification module. The sampling cycle after execution regenerates the undertaking record; when the flow deviation or pressure deviation after execution increases compared to the deviation before execution, and the pump group energy efficiency ratio after execution is not higher than the pump group energy efficiency ratio before execution, a rollback flag is written indicating that rollback is required; when the flow deviation or pressure deviation after execution does not increase, or the pump group energy efficiency ratio after execution is higher than the pump group energy efficiency ratio before execution, a rollback flag is written indicating that execution is retained; when the rollback flag indicates that rollback is required, the next sampling cycle sends a reverse action command to the PLC: pump increase rollback stops newly added pumps, pump decrease rollback starts stopped pumps, replacement undertaking pump rollback starts stopped abnormal undertaking pumps and stops replacement pumps, and undertaking order rearrangement rollback restores the frequency allocation order of the previous sampling cycle; the edge computing controller writes the executed action, the pump number of the action object, the change in flow deviation or pressure deviation before and after execution, the change in pump group energy efficiency ratio before and after execution, the change in undertaking mismatch value before and after execution, and the rollback flag into the pumping scheduling record; When the rollback flag indicates that rollback is required, it means that the execution result of the current scheduling action will not be written into the priority retention state of the next sampling period. The edge computing controller will reread the pumping acquisition record, single pump energy efficiency record and load record in the next sampling period, and generate a new scheduling instruction based on the regenerated flow deviation or pressure deviation, pump group energy efficiency ratio and load mismatch value. The rollback flag does not indicate that the pump group state before the scheduling action was executed will be restored immediately. Through the above processing, the reconfigured scheduling module does not directly convert the average efficiency decline into pump-cutting action. Instead, it first reads the changes in flow rate, pressure, frequency, power, and energy efficiency caused by previous actions, and then combines this with the changes in the load mismatch value of the abnormally operating pump to determine the source of inefficiency. Subsequently, it selects a scheduling instruction that matches the hysteretic inefficiency state or the structural inefficiency state, and verifies the scheduling result by checking the post-execution deviation, pump group energy efficiency ratio, and load mismatch value. In practical applications: three effluent pumps operate in constant pressure mode. After the frequency increase action was performed in the previous sampling cycle, the effluent water pressure in the current sampling cycle... If the force continues to change along the direction of the frequency increase action, and the mismatch value of the abnormal pump is lower than the mismatch value of the sampling period after the action, the edge computing controller generates a hysteretic inefficient state and only writes frequency fine-tuning instructions to the main pump and the follower pump. If the effluent water pressure has deviated from the direction of the frequency increase action and the mismatch value of the abnormal pump has not decreased, the edge computing controller generates a structural inefficient state and, if there is a stopped pump, prioritizes generating a replacement pump instruction. After execution, it recalculates the pressure deviation, pump group energy efficiency ratio, and mismatch value, and then writes the verification results into the pumping scheduling record.

[0022] Working Principle: This solution first collects data from the effluent pump and main pipe using the effluent flow meter, effluent pressure meter, inlet pressure meter, outlet pressure meter, power acquisition unit, and pump status acquisition terminal, generating a pumping acquisition record. Then, the host computer determines the constant flow or constant pressure basic operating mode, the edge computing controller calculates the flow deviation or pressure deviation, and the PLC and PID controller drive the frequency converter to perform multi-pump synchronous frequency regulation. Subsequently, based on the single pump pressure, flow rate, pipe diameter area, shaft power, and power consumption, the static pressure difference, velocity head difference, total head, water power, pump efficiency, and single pump output energy efficiency ratio are calculated to further identify the main pump, follower pump, and abnormal pump. Finally, combining the response changes after the preceding frequency increase, frequency decrease, pump increase, and pump decrease actions, the solution distinguishes between hysteresis inefficiency and structural inefficiency states, and generates frequency fine-tuning, pump increase, pump decrease, pump replacement, or pump order reordering commands respectively. After execution, the deviation, pump group energy efficiency ratio, and pump mismatch value are recalculated to form a closed-loop scheduling. In a continuous water supply scenario at a secondary pumping station in a waterworks, when three effluent pumps are operating at constant pressure, the system first ensures that the effluent pressure follows the target value, and then calculates the actual output and energy consumption of each pump. If one of the pumps has a high frequency and power, but low effluent contribution and energy efficiency, the system will not immediately increase or decrease the pump based on the low average efficiency. Instead, it will first determine whether this inefficiency is due to a short-term response lag after the frequency increase. If the pressure is still recovering and the abnormal load value is decreasing, only a minor frequency adjustment will be made to avoid frequent pump switching. If the pressure response has ended but the pump is still high in consumption and low in contribution, it is determined to be structurally inefficient. The abnormal pump will be replaced with a shut-down pump first, or the priority of each pump will be rearranged, thereby adjusting the pump group operation structure without disrupting the water supply target.

[0023] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A smart pumping system for a waterworks, characterized in that, include: The data acquisition module includes an outflow water flow meter, an outflow water pressure meter, an inlet pressure meter, an outlet pressure meter, a power acquisition unit, and a pump status acquisition terminal. It is used to collect the pump start-stop status, operating frequency, inlet pressure, outlet pressure, outflow water flow, inlet area, outlet area, shaft power, power consumption during sampling cycle, actual outflow water flow, and actual outflow water pressure, and outputs pumping acquisition records. The control execution module, including a host computer, PLC, PID controller, frequency converter and edge computing controller, is used to read the basic operating mode, the target value of the 24-hour control segment and the pumping acquisition record. In constant flow mode, the flow deviation is generated by subtracting the actual flow from the target outflow. In constant pressure mode, the pressure deviation is generated by subtracting the actual outflow from the target outflow pressure. The PID controller outputs multi-pump synchronous frequency adjustment commands according to the flow deviation or pressure deviation. The single-pump calculation module is used to read the pumping acquisition records, subtract the inlet pressure from the outlet pressure and add the pressure compensation value to generate the static pressure difference, convert the water flow rate, inlet area and outlet area into velocity head difference, convert the static pressure difference into static head and add it to the velocity head difference to generate the total head, and then generate water power, pump efficiency and single pump output energy efficiency ratio from the total head, water flow rate, shaft power, outlet pressure and sampling cycle power consumption, and output the single pump energy efficiency record. The responsibility identification module is used to read pumping acquisition records and single pump energy efficiency records, normalize and sum the operating frequency, shaft power and sampling cycle power consumption to generate the input responsibility value, normalize and sum the outflow flow ratio, pressure flow output ratio and pump efficiency ratio to generate the target contribution value, subtract the target contribution value from the input responsibility value to generate the responsibility mismatch value, generate the main responsibility pump according to the target contribution value with the largest value, generate the abnormal responsibility pump according to the positive and largest value of the responsibility mismatch value, generate the other operating pumps as follow-up responsibility pumps, and output the responsibility record; The reconfiguration scheduling module is used to read basic control records, single pump energy efficiency records, undertaking records, and previous action records. It sequentially compares the changes in total outflow water flow, outflow water pressure, operating frequency, shaft power, and single pump output energy efficiency ratio before and after the previous frequency increase, frequency decrease, pump increase, and pump decrease actions with the current undertaking mismatch value, generates a hysteresis inefficiency state or a structural inefficiency state, and outputs a frequency fine-tuning command in the hysteresis inefficiency state. In the structural inefficiency state, it generates pump increase commands, pump decrease commands, replacement commands, or undertaking order reordering commands according to the main undertaking pump, follow-up undertaking pump, and abnormal undertaking pump, and outputs the pumping scheduling record.

2. The intelligent pumping system for a waterworks according to claim 1, characterized in that: The execution of the data acquisition module includes: The pump status acquisition terminal establishes a pump number field for the water pump in automatic frequency conversion adjustment mode, and reads the start-stop status and operating frequency under the pump number field according to the sampling period. The inlet pressure gauge reads the inlet pressure under the same pump number field, the outlet pressure gauge reads the outlet pressure under the same pump number field, and the power acquisition unit reads the shaft power and power consumption during the sampling period under the same pump number field to generate a single pump status item. The inlet pipe diameter, outlet pipe diameter, and outlet flow rate of the water pump are read from the pump number field. The inlet pipe diameter is converted into the inlet area, the outlet pipe diameter is converted into the outlet area, and the outlet flow rate, inlet area, and outlet area are written into the single pump status item to generate a single pump data acquisition item. The outflow meter reads the actual outflow water flow rate according to the sampling cycle, and the outflow water pressure meter reads the actual outflow water pressure according to the same sampling cycle. The actual outflow water flow rate, actual outflow water pressure, sampling cycle number and all single pump acquisition items are combined to generate a pumping acquisition record.

3. The intelligent pumping system for a waterworks according to claim 2, characterized in that: The execution of the control execution module includes: The host computer reads the basic operating mode and the target value of the 24-hour control segment, and sends the basic operating mode, the current control segment number, the target outflow water flow rate in constant flow mode or the target outflow water pressure in constant pressure mode to the edge computing controller. The edge computing controller reads the pumping acquisition records. In constant flow mode, it generates a flow deviation by subtracting the actual outflow from the target outflow. In constant pressure mode, it generates a pressure deviation by subtracting the actual outflow from the target outflow pressure. The flow deviation or pressure deviation is written into the basic control record and then sent to the PLC. The PLC sends the flow or pressure deviation to the PID controller. The PID controller generates a frequency adjustment amount based on the flow or pressure deviation. The PLC sends a synchronous frequency adjustment command to the frequency converter based on the frequency adjustment amount. The frequency converter adjusts the water pump in automatic frequency conversion adjustment mode according to the synchronous frequency adjustment command and outputs the execution result of the multi-pump synchronous frequency adjustment command. If the actual outflow water flow or actual outflow water pressure is missing for two consecutive sampling cycles, the control execution module will stop generating new flow deviations or pressure deviations and stop outputting the automatic scheduling commands generated by the intelligent pumping system. The PLC will then switch to the default running state or the manual takeover state.

4. The intelligent pumping system for a waterworks according to claim 3, characterized in that: The execution of the single-pump calculation module includes: Read the inlet pressure, outlet pressure, water flow rate, inlet area, and outlet area under the same pump number field in the pumping acquisition record. Subtract the inlet pressure from the outlet pressure and add the pressure compensation value to generate the static pressure difference. Convert the water flow rate to the unit flow rate per second according to the sampling time unit. Divide the unit flow rate per second by the inlet area and outlet area respectively to generate the inlet velocity and outlet velocity. Subtract the square of the inlet velocity from the square of the outlet velocity and divide by the gravity head conversion term to generate the velocity head difference. The static head is generated by multiplying the static pressure difference by the pressure head conversion factor. The static head is generated by adding the static head and the velocity head difference. The total head is generated by multiplying the total head, the outlet flow rate and the gravitational acceleration and then converting the flow rate and time units to generate the water power. Read the shaft power and sampling cycle power consumption under the same pump number field, divide the water power by the shaft power to generate the pump efficiency, divide the product of the outlet pressure and the outlet flow rate by the sampling cycle power consumption to generate the single pump output energy efficiency ratio, and write the static pressure difference, velocity head difference, total head, water power, pump efficiency and single pump output energy efficiency ratio into the same pump number field to generate a single pump energy efficiency record.

5. The intelligent pumping system for a waterworks according to claim 4, characterized in that: The execution of the identification module includes: Read the operating frequency, shaft power, sampling cycle power consumption, outlet flow rate, outlet pressure, pump efficiency, and single pump output energy efficiency ratio under the same pump number field in the pump acquisition record and single pump energy efficiency record. Divide the operating frequency, shaft power, and sampling cycle power consumption by the sum of the corresponding fields of all operating pumps and then add them to generate the input responsibility value. Divide the product of outlet flow rate, outlet pressure and outlet flow rate, and pump efficiency by the sum of the corresponding fields of all operating pumps and then add them to generate the target contribution value. Subtract the target contribution value from the input responsibility value to generate the responsibility mismatch value. The operating frequency sequence, effluent flow rate sequence, shaft power sequence, and single pump output energy efficiency ratio sequence under the same pump number field are read sequentially according to the sampling period. Fourier decomposition is performed on the operating frequency sequence, and the frequency item with the first value in amplitude sorting is taken as the frequency modulation master item. The complex components of the effluent flow rate sequence, shaft power sequence, and single pump output energy efficiency ratio sequence are read at the frequency position of the frequency modulation master item, respectively, to obtain the flow rate frequency item, power frequency item, and energy efficiency frequency item. The phase order of the flow rate frequency item relative to the frequency modulation master item, the phase order of the power frequency item relative to the flow rate frequency item, and the amplitude increase / decrease direction of the energy efficiency frequency item relative to the power frequency item are written into the response item.

6. The intelligent pumping system for a waterworks according to claim 5, characterized in that: The execution of the identification module also includes: The operating pump ranked first by the target contribution value is generated as the main undertaking pump. The remaining operating pumps are arranged in descending order of their undertaking mismatch values ​​to generate a candidate pump chain. Starting from the first pump in the candidate pump chain, each candidate pump is read one by one. For each candidate pump read, the undertaking response item of the main undertaking pump is deducted from the undertaking response item of the candidate pump to obtain the net mismatch response item of the candidate pump. The net mismatch response item and the undertaking mismatch value are written together into the temporary abnormal pump item.

7. The intelligent pumping system for a waterworks according to claim 6, characterized in that: The execution of the identification module also includes: The temporary abnormal pump items are subjected to pruning search, including: randomly selecting two temporary abnormal pump items for pairwise comparison. If their frequency regulation main term is the same, the flow rate following frequency term phase order is the same, and the power following frequency term phase order is the same, the temporary abnormal pump item with the lower mismatch value ranking is deleted; if their frequency regulation main term is the same and the energy efficiency following frequency term amplitude increases or decreases in opposite directions, the temporary abnormal pump item with the higher mismatch value ranking is retained, and the other temporary abnormal pump item is added to the follower candidate list; if the input contribution value of the temporary abnormal pump item is not higher than the target contribution value, the temporary abnormal pump item is deleted from the candidate pump chain, resulting in a candidate abnormal chain.

8. The intelligent pumping system for a waterworks according to claim 7, characterized in that: The execution of the identification module also includes: The process of greedily selecting pumps from the candidate anomaly chain includes: first, reading the first candidate anomaly pump in the candidate anomaly chain as the current anomaly pump, and then sequentially reading the next candidate anomaly pump; if merging the next candidate anomaly pump with the current anomaly pump increases the total mismatch value of the anomaly pump set but does not increase the total target contribution value, then the next candidate anomaly pump is merged into the anomaly pump set; if the total mismatch value does not increase after merging, or the total target contribution value increases, then the next candidate anomaly pump is skipped; after traversal, the operating pump with the highest mismatch value in the anomaly pump set is generated as the anomaly bearing pump, and the operating pumps other than the main bearing pump and the anomaly bearing pump are generated as follower bearing pumps, and the main bearing pump, follower bearing pump, anomaly bearing pump, input bearing value, target contribution value, bearing mismatch value, bearing response item, and net mismatch response item are written into the bearing record.

9. A smart pumping system for a waterworks according to claim 8, characterized in that: The execution of the reconfiguration scheduling module includes: Read the basic control record, single pump energy efficiency record, load record, and preceding action record. Read the total outflow water flow, outflow water pressure, operating frequency, shaft power, and single pump output energy efficiency ratio before and after the preceding frequency increase, frequency decrease, pump increase, and pump decrease actions in sequence according to the action occurrence time. Generate flow action difference, pressure action difference, frequency action difference, power action difference, and energy efficiency action difference respectively. Compare the corresponding field difference between the current sampling period and the sampling period after the action with the change sequence of the load mismatch value of the abnormal load pump. If the flow or pressure still changes along the direction of the preceding action and the load mismatch value of the abnormal load pump is lower than the load mismatch value of the sampling period after the action, a hysteretic inefficient state is generated. If the flow or pressure has changed away from the direction of the preceding action and the load mismatch value of the abnormal load pump is not lower than the load mismatch value of the sampling period after the action, a structural inefficient state is generated.

10. A smart pumping system for a waterworks according to claim 9, characterized in that: The execution of the reconfiguration scheduling module also includes: In a hysteresis-inefficient state, the flow deviation or pressure deviation in the basic control record is read, the frequency adjustment value generated by the PID controller is written to the main pump and the follower pump, and the start / stop state of the abnormal pump remains unchanged, generating a frequency fine-tuning instruction; in a structurally inefficient state, if the input load value of the main pump is higher than the target contribution value and there is no running pump with a positive load mismatch value among the follower pumps, then the pump stop is written to the pump increase instruction; if there is a running pump at the bottom of the target contribution value sorting among the follower pumps and the input load value is higher than the target contribution value, then the pump decrease instruction is written to the follower pump; if the load mismatch value of the abnormal pump is at the top of the sorting and there is a stopped pump, then the abnormal pump is stopped and the stopped pump is started, writing a replacement pump instruction; if there is no stopped pump, then the frequency allocation order of the main pump, the follower pump, and the abnormal pump is rearranged from high to low according to the target contribution value and a load order rearrangement instruction is generated. After executing the frequency fine-tuning command, pump increase command, pump decrease command, pump replacement command, or pump order reordering command, the actual outflow water flow rate, actual outflow water pressure, single pump output efficiency ratio, and pump mismatch value are reread. The system generates the post-execution flow deviation or post-execution pressure deviation, post-execution pump group efficiency ratio, and post-execution pump mismatch value. The system also writes the execution action, changes in flow deviation or pressure deviation before and after execution, changes in pump group efficiency ratio before and after execution, changes in pump mismatch value before and after execution, and a rollback flag into the pumping scheduling record.

Images