A coal-fired power generating unit circulating water pump optimization method, device and storage medium

Abstract

This invention relates to the field of energy conservation and consumption reduction in power plants, and discloses a method, equipment, and storage medium for optimizing circulating water pumps in coal-fired power generating units. It addresses problems such as insufficient integration of fluid dynamic response characteristics and inadequate matching of pump pipeline resistance in traditional circulating water pump operation. The method includes: acquiring unit test data to establish a back pressure dynamic response model; determining pump operating hydraulic characteristic parameters including flow rate and power consumption; determining the back pressure output correction relationship caused by pump flow rate changes under different loads based on unit operating data; constructing a pump operation control strategy, including a pump impeller speed control strategy, using the model, parameters, and relationships; and evaluating and matching the control strategy with pipeline network prediction data and measured state parameters to obtain a fluid resource allocation sequence. This invention effectively reduces the overall coal consumption cost of unit operation by controlling impeller speed while meeting equipment safety constraints.

Description

Technical Field

[0001] This invention relates to the field of energy conservation and consumption reduction in power plants, and in particular to a method, equipment and storage medium for optimizing the circulating water pump of a coal-fired power generation unit. Background Technology

[0002] In the thermodynamic cycle of a coal-fired power plant, condenser vacuum (back pressure) is one of the core parameters affecting the unit's thermal economy. Condenser vacuum is mainly determined by the operation mode of the circulating water system: the circulating water system delivers cooling water to the condenser through circulating water pumps, and changes in the circulating water flow rate will change the condenser heat exchange conditions, causing changes in back pressure.

[0003] Therefore, the operation and regulation of circulating water pumps is essentially a matching problem between fluid mechanical power consumption and system heat exchange benefits: increasing the circulating water volume can reduce back pressure and increase unit output (reducing coal consumption), but at the same time, it will increase the power consumption of the circulating water pumps themselves (increasing coal consumption). Finding the optimal combination of circulating water pumps (discrete fluid node start-stop matching, continuous fluid node flow regulation) to achieve an optimal balance between the coal consumption benefits from reduced back pressure and the coal consumption costs from increased power consumption throughout the entire cycle is the key to achieving deep energy saving and consumption reduction in the unit.

[0004] Although some research has attempted to improve circulating water pump control, existing technologies still have the following significant shortcomings: Traditional methods for pump scheduling often rely on steady-state models, assuming that the back pressure instantly reaches a new steady-state value after a change in circulating water flow. However, in reality, the condenser cooling water flow is a thermodynamic object with high inertia and significant time delay. After a step change in circulating water flow, the back pressure requires a period of pure delay and a time constant to gradually stabilize. Ignoring the dynamic process of fluid transport leads to a disconnect between control commands and actual conditions, easily resulting in erroneous start / stop commands under varying loads, and causing vacuum deterioration.

[0005] Furthermore, existing controls rarely delve deeply into the transient characteristics of the coupled water pumps. Since errors are inevitable in pipeline resistance, load forecasting, and circulating water inlet temperature forecasting, traditional open-loop control, when the deviations are large, can easily lead to condenser vacuum exceeding safety limits or frequent pump start-ups and shutdowns. Existing technologies lack effective feedback correction and dynamic parameter matching calculation mechanisms, making it difficult to guarantee strict constraint compliance and safe equipment operation.

[0006] Therefore, we propose an optimization method, equipment, and storage medium for circulating water pumps in coal-fired power generating units to solve the above problems. Summary of the Invention

[0007] This invention provides a method, equipment, and storage medium for optimizing circulating water pumps in coal-fired power generating units, which addresses problems such as neglecting dynamic characteristics, failing to consider coal consumption throughout the entire cycle, lack of modeling for switching transition processes, insufficient handling of prediction uncertainties, and coarse characteristic models in traditional circulating water pump control.

[0008] The first aspect of this invention provides a method for optimizing circulating water pumps in coal-fired power generating units. The method includes: acquiring unit test data and establishing a back pressure dynamic response model to the change in circulating water flow rate in response to condenser back pressure; determining the total circulating water flow rate and total pump power consumption under each cooling water delivery configuration mode based on pump performance data and pipeline characteristic data, and obtaining pump operating hydraulic characteristic parameters; determining the change in unit output caused by back pressure changes under different loads based on unit operating characteristic data, and obtaining a back pressure-output correction relationship; constructing a pump operation control strategy with the goal of optimizing the overall operating cost of the pump and system, and constrained by the mechanical and hydraulic operating boundaries of the equipment, using the back pressure dynamic response model, the pump operating hydraulic characteristic parameters, and the back pressure-output correction relationship, the pump operation control strategy includes a pump impeller speed control strategy; and performing dynamic parameter matching on the pump operation control strategy by combining pipeline network prediction data and measured state parameters to obtain a fluid resource allocation sequence and output the current time period's distribution node optimization decision.

[0009] Optionally, in a first implementation of the first aspect of the present invention, the method includes: extracting unit output data corresponding to different condenser back pressures from the unit operating characteristic data; determining the change in unit output caused by back pressure changes at multiple typical load points, and obtaining back pressure correction coefficients corresponding to each of the multiple typical load points; and organizing the back pressure correction coefficients corresponding to each typical load point according to the load range to form a back pressure output correction relationship.

[0010] Optionally, in the second implementation of the first aspect of the present invention, the method includes: discretizing the operation adjustment cycle into multiple time periods, using the fluid distribution mode identifier in each time period as the optimization decision variable; for each time period, converting the change in unit output in that time period into fuel and coal consumption increment, converting the total power consumption of the pump in that time period into coal consumption increment, and constructing a comprehensive energy consumption accounting model based on the fuel and coal consumption increments and the coal consumption increments in all time periods, with the minimum comprehensive energy consumption of the pump throughout the cycle as the comprehensive accounting index; and converting the mechanical and hydraulic operation boundaries of the equipment into constraints of the corresponding operation command variables, wherein the mechanical and hydraulic operation boundaries of the equipment include: frequency limit for switching state of distribution resources, minimum state dwell time limit, and upper limit of condenser vacuum.

[0011] Optionally, in a third implementation of the first aspect of the present invention, the method includes: A continuous sliding time-period adjustment method is adopted, and the measured back pressure value is collected at the current decision moment as the measured state parameter to update the initial state of the back pressure dynamic response model; The predicted load values ​​and circulating water inlet temperature values ​​for multiple future time periods are obtained as the predicted data, and the constraints are corrected in combination with the preset safety margin value. After updating the initial state and correcting the constraints, the pump operation control strategy is solved to obtain the fluid resource allocation sequence, and the optimal distribution node optimization decision output for the current time period is extracted.

[0012] Optionally, in a fourth implementation of the first aspect of the present invention, it further includes: Acquire the transitional characteristic data of circulating water flow rate and total pump power consumption over time during the switching of adjacent distribution mode states of the circulating water pump; When constructing the pump operation control strategy, if the transmission and distribution mode state switch occurs between adjacent time periods, the average transition value in the transition characteristic data is used to replace the steady-state value, and the change in unit output and the total power consumption of the pump during the corresponding time period are calculated.

[0013] Optionally, in a fifth implementation of the first aspect of the present invention, determining the change in unit output caused by back pressure changes at multiple typical load points, obtaining the back pressure correction coefficients corresponding to each of the multiple typical load points, and organizing the back pressure correction coefficients corresponding to each typical load point according to the load range to form the back pressure output correction relationship includes: Linear regression is performed on the unit output data extracted under each typical load point to obtain the back pressure correction coefficient for the corresponding typical load point; Using the adjacent typical load points as interval endpoints, linear interpolation is used to obtain the corresponding back pressure correction coefficient for any load value between adjacent endpoints, thus forming the back pressure output correction relationship.

[0014] Optionally, in a sixth implementation of the first aspect of the present invention, the step of converting the change in unit output during the time period into an increase in fuel consumption, and converting the total power consumption of the pump during the time period into an increase in coal consumption, includes: The turbine heat rate, the lower heating value of fuel, and the unit price are obtained. Based on the change in the unit output in each time period, the fuel consumption increment in each time period is calculated. Obtain the total plant power consumption rate and calculate the corresponding coal consumption increment based on the total power consumption of the pumps in each time period; The fuel consumption increment for each time period is added to the coal consumption increment for the corresponding time period to obtain the comprehensive operating energy consumption of the water pump for each time period.

[0015] Optionally, in the seventh implementation of the first aspect of the present invention, the fuel consumption increment of each time period is added to the coal consumption increment of the corresponding time period to obtain the comprehensive operating energy consumption of the water pump in each time period, and the total coal consumption cost increment for a single time period is calculated using the following formula. : ; In the formula, k is the time period number; Let k be the total power consumption of the pump during time period k; Let k be the change in unit output during time period k; This represents the plant's total power consumption rate. This refers to the turbine heat rate; The lower heating value of fuel; Price per unit; The duration of a single time period.

[0016] A second aspect of the present invention provides a circulating water pump optimization device for a coal-fired power generation unit, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor invokes the instructions in the memory to cause the circulating water pump optimization device for the coal-fired power generation unit to execute the above-described circulating water pump optimization method for the coal-fired power generation unit.

[0017] A third aspect of the present invention provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the above-described method for optimizing the circulating water pump of a coal-fired power generator set.

[0018] The mechanism of this invention is as follows: By closely combining the characteristics of the fluid pipeline network with the actual hydraulic performance of the pump, it can not only meet the hard constraints of electromechanical equipment such as the number of pump start-ups and shutdowns, minimum operating downtime, and upper vacuum limit, but also effectively cope with model mismatch and external disturbances. It organically combines pipeline hydraulic identification, multi-source data fusion, and continuous sliding regulation, so that the state switching decision of the circulating water distribution mode is no longer a static table lookup, but can reflect the changes in fluid dynamics and achieve global dynamic matching of parameters in all operating conditions and the whole cycle. Beneficial effects: Based on field step test data, the least squares method is used to identify the parameters of the first-order inertial plus pure delay model, and a back pressure dynamic response model parameter table is formed according to different loads and circulating water inlet temperature ranges, instead of using a single fixed model. This characterizes the large delay and large inertia dynamic characteristics of condenser back pressure to circulating water flow, optimizes the system's ability to predict the lag process of back pressure changes, and avoids common mistakes based on steady-state models, such as blindly starting pumps ahead of time or lagging vacuum reduction. This fundamentally improves the scientificity and safety of control decisions under varying operating conditions.

[0019] By summing the changes in unit output caused by back pressure changes with the total power consumption of circulating water pumps over the entire time period, a comprehensive energy consumption accounting model with the minimum increase in total coal consumption cost over the entire cycle is constructed. This achieves system-level, full-time-scale global energy conservation coordination, ensuring that every pump adjustment can bring the maximum net coal consumption benefit over a long period of time. In discrete-time optimization, the actual energy consumption of switching unsteady-state processes is quantified, filling the gap in dynamic switching modeling, eliminating the computational blind spot caused by ignoring the transition process, and making the optimization results highly consistent with the actual energy consumption, thus truly realizing fully dynamic rather than quasi-steady-state optimization.

[0020] This constitutes an operation and control framework that closely matches the physical characteristics of the water pump. It not only overcomes model mismatch in real time through feedback, but also effectively resists the uncertainty of load and water temperature prediction through hydraulic statistical margin. It strictly ensures the vacuum safety of the condenser and the start-up and shutdown restrictions of the equipment, and improves the anti-disturbance capability and mechanical operation reliability of the fluid pipeline system under complex working conditions.

[0021] Achieving high-precision modeling across the entire chain, from equipment mechanism to operational data, ensures that the optimization model is widely applicable to various complex circulating water systems that include a mix of constant-speed pumps and variable-frequency pumps, and the calculation results can be directly used for engineering guidance. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of an embodiment of the method for optimizing the circulating water pump of a coal-fired power generation unit according to the present invention; Figure 2 This is a schematic diagram of another embodiment of the method for optimizing the circulating water pump of a coal-fired power generation unit according to the present invention; Figure 3 This is a schematic diagram illustrating how changing the operating mode of the circulating water pump can cause a step change in the circulating water flow rate. Figure 4 A schematic diagram of the response curve for recording the change of condenser back pressure over time; Figure 5 This is a schematic diagram of one embodiment of the circulating water pump optimization equipment for coal-fired power generation units in this invention. Detailed Implementation

[0023] This invention provides a method, device, and storage medium for optimizing circulating water pumps in coal-fired power generating units. It addresses problems in traditional circulating water pump control, such as neglecting dynamic characteristics, failing to consider full-cycle coal consumption, lacking modeling for switching transition processes, insufficient handling of predictive uncertainties, and coarse characteristic models. The terms "first," "second," "third," "fourth," etc. (if present)," in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0024] For ease of understanding, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 1 One embodiment of the optimization method for circulating water pumps of coal-fired power generating units in this invention includes: 101. Based on the field step test data, establish a first-order inertial plus pure delay dynamic response model of condenser back pressure to circulating water flow rate change, and obtain the back pressure dynamic response model.

[0025] It is understood that the executing entity of this invention can be a circulating water pump optimization device for a coal-fired power generation unit, or it can be a terminal or a server; no specific limitation is made here. This embodiment of the invention will be described using a server as an example.

[0026] It should be noted that the executing entity (server) needs to control the field equipment to apply disturbances, collect actual operating data, and extract dynamic features to complete parameter identification.

[0027] A field step test was conducted during the stable operation of the coal-fired power generating unit. Taking a unit with a rated power of 300MW as an example, the unit load was maintained at a stable operation of 225MW, at which time the cooling water inlet temperature was measured to be stable at 20.0℃. Before the test began, the server recorded the initial balance data at this time: the total circulating water flow rate was stable at 32000t / h, and the condenser back pressure was stable at 5.20kPa.

[0028] A step disturbance in flow rate is applied to the system. At the 10-minute mark of data recording, the server issues a command to start a standby circulating water pump or adjust the operating frequency of the running pump, causing a step increase in the circulating water flow rate. The flow rate increases briefly and then stabilizes again at 40,000 t / h. At this point, the system receives a step change in circulating water flow rate of 8,000 t / h.

[0029] The server continuously monitors the dynamic response data of the condenser back pressure. Observation records show that within the first minute after the flow rate increase (i.e., from the 10th to the 11th minute), the condenser back pressure did not change immediately and remained at 5.20 kPa. Since pipeline transportation and heat exchange processes require time, this waiting time during which the back pressure does not respond to flow rate changes is the pure delay time parameter in the model, which was confirmed and recorded as 60 seconds in this embodiment.

[0030] After the pure delay period ends (i.e., from the 11th minute), as the cooling water flow increases and the heat exchange effect improves, the condenser back pressure begins to show a downward trend. The back pressure gradually decreases until around the 25th minute, when the downward trend of the back pressure slows down and stabilizes at a new equilibrium value of 4.40 kPa. At this point, the total steady-state change in back pressure is a decrease of 0.80 kPa.

[0031] Based on the extracted dynamic test data, the server performs parameter calculations and constructs a first-order inertial plus pure delay model: When calculating the static gain parameter, the 0.80 kPa decrease in back pressure is divided by the 8000 t / h increase in flow rate; when calculating the time constant parameter, the moment when the back pressure change reaches 63.2% of the total change (i.e., decreases to 4.69 kPa) is found, the time taken is 5 minutes, and the time constant is confirmed to be 300 seconds. The server combines the pure delay time, static gain, and time constant to obtain the back pressure dynamic response model.

[0032] 102. Based on the performance curves of the circulating water pump and the characteristic curves of the pipeline, determine the total circulating water flow and total pump power consumption corresponding to each cooling water delivery configuration mode, and obtain the hydraulic characteristic parameters of the pump operation.

[0033] It should be noted that the circulating water system of the aforementioned 300MW unit is equipped with three fixed-frequency circulating water pumps of the same model, which are labeled as pump A, pump B and pump C respectively.

[0034] The system retrieves the factory performance data of these three circulating water pumps from the equipment management database, extracting the flow rate versus head curve and the flow rate versus shaft power curve for each pump. Simultaneously, the server reads the pipeline characteristic curve fitted from prior field surveys, which reflects the principle that resistance increases with total flow rate when water flows through the condenser.

[0035] For all possible pump combination operation modes, the operating point is calculated one by one: When operating with only one pump (pump A only), the server overlays and matches the flow rate and head curves of the single pump with the pipeline characteristic curves. The intersection of the two curves is the actual operating point. The server calculates that the total circulating water flow rate at this intersection is 20,000 t / h. This value is then located on the flow rate and power curve, determining that the total power consumption of the pump at this point is 2000 kW.

[0036] When two pumps are operated in parallel (pumps A and B are running), the server superimposes the flow rates of the two pumps at the same head to obtain the parallel performance curve, and then intersects it with the pipeline curve. Due to the increased pipeline resistance, the total circulating water flow rate is calculated to be 32000 t / h. Corresponding to the power curve, the total power consumption of the two pumps is confirmed to be 3800 kW.

[0037] When three pumps are operated in parallel (all three pumps, A, B, and C, are running), the pipeline resistance increases significantly, and the total circulating water flow at the operating point is calculated to be 40,000 t / h. Further calculations show that the total power consumption of the three pumps under this operating combination reaches 5,600 kW.

[0038] The server summarizes the calculated data in a structured format to obtain a table of pump operating hydraulic characteristic parameters. This table clearly maps the control decision results: 1 pump mode corresponds to 20000t / h and 2000kW; 2 pump modes correspond to 32000t / h and 3800kW; 3 pump modes correspond to 40000t / h and 5600kW.

[0039] 103. Based on the turbine heat balance diagram or on-site micro-output test data, determine the change in unit output caused by back pressure changes under different loads, and obtain the back pressure output correction relationship table.

[0040] It should be noted that the historical data of the on-site micro-output test of the 300MW unit under typical load conditions (150MW, 225MW, 300MW) and the turbine heat balance diagram were retrieved from the production system. The server extracted the data according to the set load range and quantified the absolute output increase or decrease of the unit corresponding to each unit change (1kPa) in back pressure under different load levels.

[0041] Taking the unit operating at a 225MW load as an example, the server retrieved test records: While keeping the turbine control valve opening constant, when the condenser back pressure decreased from the baseline 5.00 kPa to 4.00 kPa, the actual output increased from 225.0 MW to 229.0 MW due to the increased steam expansion work space. Based on this, the server concluded that at a 225MW load level, for every 1 kPa decrease in back pressure, the unit output will increase by 4.0 MW.

[0042] Data conversion for other load ranges confirmed that at a low load level of 150MW, the steam flow is relatively small, and the unit output increases by 3.0MW for every 1kPa decrease in back pressure; while at a full load level of 300MW, the steam flow is large, and the change in back pressure has a more significant impact on power output, with the unit output increasing by 5.0MW for every 1kPa decrease in back pressure.

[0043] The server integrates and performs boundary interpolation on the data of different load nodes and their corresponding back pressure output increment rates to obtain a 2D back pressure output correction table. This table uses the actual load as the horizontal index and the back pressure change as the vertical index; the values ​​within the grid represent the output correction increments. Once completed, when predicting back pressure changes caused by different pump combinations, the system can directly consult this table to obtain the unit's output benefits, which are then converted into standard coal consumption to support further economic calculations.

[0044] 104. Using the back pressure dynamic response model, pump operating hydraulic characteristic parameters and back pressure output correction relationship table, construct a pump operation control strategy with the minimum total coal consumption cost increment over the whole cycle as the comprehensive accounting index, and the pump start-up and shutdown frequency limit, minimum pump operation downtime and condenser vacuum upper limit as mechanical and process safety boundaries. The pump operation control strategy includes a pump impeller speed control strategy.

[0045] It should be noted that time was discretized. The entire optimization cycle was set to the next 4 hours, which was then divided into 24 time steps, each lasting 10 minutes. The standard coal price was set at 1000 yuan / ton, and the coal consumption rate for power generation was 300g / kWh.

[0046] The target calculation logic is constructed. At any time step, if the pump combination is changed (from 2 pumps to 3 pumps), the server queries the pump operating hydraulic characteristic parameters to obtain the power consumption increase data. In this example, 3 pumps consume 1.8MW more than 2 pumps. The flow rate change is input into the back pressure dynamic response model to predict the future back pressure decrease trajectory. For the predicted back pressure, the back pressure output correction relationship table is consulted to convert it into the turbine output increase. When the back pressure steadily decreases by 0.80 kPa, the corresponding output increase at a 225MW load is 3.2MW. The output increase is subtracted from the pump power consumption increase (3.2MW minus 1.8MW, net increase of 1.4MW). The net output change is converted into the coal consumption cost increment. The server accumulates the cost increments over 24 time steps; the goal is to find the solution that minimizes the total coal consumption cost.

[0047] The server adds three constraints: First, a limit on the number of pump start-stop cycles; a single pump must not be started or stopped more than once throughout the entire cycle. Second, a minimum pump downtime; a single pump must run continuously for at least 60 minutes after starting before being allowed to stop, and must wait at least 30 minutes after stopping before being allowed to restart. Decisions violating this rule will be rejected. Third, a condenser vacuum upper limit limit; the absolute value of the predicted condenser back pressure must not be lower than 3.5 kPa or higher than 9.0 kPa.

[0048] The server uses the above logic as the operating condition matching principle and electromechanical protection boundary to construct a multi-variable pump operation control strategy with multiple mechanical and process safety boundaries in memory.

[0049] 105. Using a continuous sliding time-period regulation method, the initial state of the back pressure dynamic response model is updated with the measured back pressure at each decision time. Combined with the predicted load value, the predicted circulating water inlet temperature value, and the margin value determined by the statistical analysis of historical prediction errors, the pump operation control strategy model is solved to obtain the fluid resource allocation sequence for multiple future time periods. The optimized decision output of the transmission and distribution node for the current time period is then output. At the next decision time, the process returns to update the initial state of the back pressure dynamic response model with the measured back pressure and solves the problem again.

[0050] It should be noted that the implementing entity needs to combine real-time measurement and future forecast data to form a dynamic closed-loop control.

[0051] Set the current time as 10:00 and optimize the window to 24 time periods over the next 4 hours. Assume that pump C in the system was manually started at 09:20 and will reach the minimum running time constraint threshold of 60 minutes of continuous operation by 10:20.

[0052] The measured condenser back pressure is 4.45 kPa. Due to slight fluctuations in fouling thermal resistance, the measured value is slightly higher than the theoretical value. The server immediately replaced the original initial state baseline in the model with this value.

[0053] The predicted data was acquired and a margin was added. Based on the thermal inertia of water bodies, the circulating water inlet temperature will rise slowly with a gentle gradient; simultaneously, the dispatch plan requires a reduction from 225MW to 150MW at 10:30. The server also added a back pressure safety margin of 0.05kPa to the model results.

[0054] The above data was substituted into the pump operation control strategy model for solution. Under the premise of satisfying all hard constraints, the pump combination sequence with the minimum comprehensive operating energy consumption of the pumps in the next 4 hours was calculated. Data from the first 4 time periods of this sequence were extracted, and the optimization results obtained by the server are shown in Table 1 below: Table 1

[0055] (Note: When pump C was stopped at 10:30, the cumulative running time of pump C had reached 70 minutes, which meets the constraint logic of more than 60 minutes.) At this point, the server only outputs the operation decisions for the three pumps in the first time period (10:00 to 10:10) and does not execute subsequent actions. After waiting for 10 minutes until 10:10, the server rereads the latest measured back pressure from the field, updates the initial state, shifts the optimization window forward by 10 minutes, reads the updated weather and load forecasts, re-solves to obtain a new sequence, and issues a new decision for the first time period, thus achieving closed-loop rolling disturbance rejection.

[0056] Please see Figure 2 , Figure 3 and Figure 4 Another embodiment of the optimization method for circulating water pumps of coal-fired power generating units in this invention includes: 201. Based on the field step test data, establish a first-order inertial plus pure delay dynamic response model of condenser back pressure to circulating water flow rate change, and obtain the back pressure dynamic response model.

[0057] Specifically, under stable unit load conditions, the operating mode of the circulating water pump is manually changed to cause a step change in the circulating water flow rate, and the response curve of the condenser back pressure over time is recorded. Based on the response curve, the steady-state gain, time constant, and pure delay time in the first-order inertial plus pure delay model are identified using the least squares method. The above parameters are identified in different load ranges and circulating water inlet temperature ranges respectively, forming a back pressure dynamic response model parameter table, which is used to characterize the back pressure dynamic response model.

[0058] It should be noted that the following is a specific implementation example for a 600MW rated power unit: A step test was conducted within a specific load and water temperature range to obtain physical response data. The current unit load was set to be stable at 450MW, and the circulating water inlet temperature was set to be stable at 25.0℃. The system is currently operating with two fixed-frequency circulating water pumps in parallel. The server recorded the initial steady-state data at this time: the total circulating water flow rate was stable at 60000t / h, and the condenser back pressure was stable at 5.80kPa.

[0059] A step disturbance was applied to the system. Five minutes into the test, the operator manually changed the operating mode and started the third standby circulating water pump. The total circulating water flow rate quickly increased and then stabilized at 80,000 t / h, introducing a step flow disturbance of 20,000 t / h into the system.

[0060] Record the response curve. The server continuously records the trajectory of the condenser back pressure change with a sampling period of 1 second. Field data shows that in the first 50 seconds after the flow rate increases, the back pressure remains at 5.80 kPa without significant change due to the physical processes of water flow transmission in the pipeline and surface heat exchange; subsequently, the back pressure begins to decrease exponentially and gradually stabilizes at a new equilibrium point of 4.90 kPa around the 25th minute.

[0061] The least squares method was used to identify and calculate the three core parameters of the first-order inertial plus pure delay model: 1. Pure delay time identification: The time difference between the occurrence of the flow step and the first continuous downward trend of the back pressure curve is identified and confirmed to be 50s.

[0062] 2. Steady-state gain identification: Extract the steady-state change in back pressure (decreased by 0.90 kPa) and the step change in flow rate (increased by 20,000 t / h). The algorithm calculates the steady-state gain under the current operating condition by fitting the steady-state data range, quantifying the impact of unit flow rate change on back pressure.

[0063] 3. Time constant identification: For the dynamic descent segment after the delay ends, the least squares method is used to fit the first-order exponential decay curve to find the time constant solution that minimizes the sum of squared errors, and the time constant is found to be 240s.

[0064] The server repeats the above identification under typical loads such as 300MW, 450MW, and 600MW and different water temperature ranges, and structurally correlates the parameter data groups obtained under all operating conditions to form a multi-dimensional back pressure dynamic response model parameter table.

[0065] 202. Based on the performance curves of the circulating water pump and the characteristic curves of the pipeline, determine the total circulating water flow and total pump power consumption corresponding to each cooling water delivery configuration mode, and obtain the hydraulic characteristic parameters of the pump operation.

[0066] Specifically, the flow rate and head performance curves and the circulation water pipeline resistance characteristic curves of each circulating water pump are obtained. For pump combinations containing only constant-speed pumps, the intersection points of the total performance curves of each pump running in parallel and the pipeline resistance characteristic curves are found to obtain the total circulating water flow rate and total head of the pump combination. For pump combinations containing variable-frequency pumps, the performance curves of the variable-frequency pumps are first converted according to the speed ratio range of the variable-frequency pumps using the similarity law. Then, the intersection points of the converted variable-frequency pump performance curves with the total performance curves of the remaining constant-speed pumps running in parallel and the pipeline resistance characteristic curves are found to obtain the total circulating water flow rate and total head of the pump combination. Based on the flow rate and head of each pump at its respective operating point, the operating efficiency of each pump is obtained by referring to the pump efficiency curve, and then the total power consumption of the pump combination is calculated. The total circulating water flow rate and total power consumption of each pump combination are summarized into pump operating hydraulic characteristic parameters. This table is used to construct the comprehensive energy consumption accounting model and constraints of the pump operation control strategy model.

[0067] Furthermore, based on the flow rate and head of each pump at its respective operating point, the operating efficiency of each pump is obtained from the pump efficiency curve. Then, the total power consumption of the pump combination is calculated, including: reading the pump hydraulic efficiency of each pump at its operating point from the pump efficiency curve; obtaining the efficiency curve of the motor matched with each pump and finding the corresponding motor efficiency based on the shaft power of each pump; obtaining the transmission efficiency between the pump and the motor; multiplying the pump hydraulic efficiency, motor efficiency, and transmission efficiency to obtain the overall unit efficiency of each pump; calculating the input electrical power of each pump based on the flow rate, head, and overall unit efficiency of each pump; adding the input electrical power of all operating pumps in the combination to obtain the total power consumption of the pump combination. This total power consumption is used as the power consumption item in the pump operating hydraulic characteristic parameters of the pump combination, and is used to calculate the total coal consumption cost increment throughout the entire cycle in the pump operation control strategy model.

[0068] It should be noted that in this embodiment, the circulating water system of the 600MW unit is equipped with three circulating water pumps, of which pumps A and B are fixed-frequency and fixed-speed pumps, and pump C is a variable-frequency pump with variable-frequency regulation capability.

[0069] Retrieve the flow rate and head performance curves of pumps A and B at rated speeds, and the baseline performance curve of pump C at the reference speed from the equipment database. Simultaneously, retrieve the resistance characteristic curve of the circulating water pipeline, verified by on-site resistance testing. This curve reflects the physical law that resistance increases quadratically with increasing flow rate.

[0070] The current system needs to calculate the total flow rate and total power consumption of a specific combination of pump A (fixed frequency) and pump C (variable frequency) operating in parallel (at 90% speed ratio): Because of the presence of a variable frequency pump, the server uses the similarity law to convert the curve of pump C based on its 90% speed ratio. Then, under the same head coordinate system, the curves of pump A and pump C are superimposed to fit the overall performance curve. The server then calculates the intersection of the overall performance curve and the pipeline resistance characteristic curve. Considering that the parallel flow rate of two full-speed fixed-frequency pumps is 60,000 t / h, the total flow rate at the intersection of the parallel resistance of one full-speed pump and one decelerating pump is calculated to be 55,000 t / h, and the total system head is 20 m. Based on the principle of equal head and reverse distribution, pump A handles a flow rate of 30,000 t / h, and pump C handles a flow rate of 25,000 t / h.

[0071] After obtaining the actual flow rate and head, calculate the efficiency and power consumption of each pump: For pump A, using a working point of 30000t / h and 20m, the hydraulic efficiency is found to be 86%, the motor efficiency to be 95%, and the transmission efficiency to be 99%. Multiplying these three together, the total efficiency of the unit is approximately 80.8%. Based on the flow rate, head, and total efficiency, the input power of pump A is calculated to be 2100kW.

[0072] For pump C, using a working point of 25000t / h and 20m, the hydraulic efficiency at 90% speed is found to be 84%, the motor efficiency is 94%, and the transmission efficiency is 99%. Multiplying these three together, the total efficiency is approximately 78.1%. Therefore, the calculated input power of pump C is 1800kW.

[0073] The server adds the 2100kW of pump A to the 1800kW of pump C, resulting in a total power consumption of 3900kW for the combined pumps.

[0074] The server executes the calculation process cyclically for all discrete operating combinations, summarizing the corresponding total circulating water flow and total pump power consumption to obtain the pump operating hydraulic characteristic parameters, providing retrieval support for building a comprehensive energy consumption accounting model.

[0075] 203. Based on the turbine heat balance diagram or on-site micro-output test data, determine the change in unit output caused by back pressure changes under different loads, and obtain the back pressure output correction relationship table.

[0076] Specifically, the turbine heat balance diagram or the unit output data corresponding to different condenser back pressures in the field micro-output test are obtained; under multiple typical load points, the back pressure and unit output data are analyzed to determine the change in unit output for each unit change in back pressure at each load point, and the back pressure correction coefficient corresponding to each load point is obtained; the back pressure correction coefficients of each load point are segmented and organized according to the load interval, and the corresponding back pressure correction coefficients are obtained by linear interpolation of the load values ​​between adjacent typical load points, forming a back pressure output correction relationship table; the back pressure output correction relationship table is used to determine the change in unit output under arbitrary load and back pressure changes in the pump operation control strategy model.

[0077] Furthermore, the unit output data corresponding to different condenser back pressures in the turbine heat balance diagram or on-site micro-output test are obtained, including: under the condition that the unit load is stable at the typical load point, keeping the main steam parameters and reheat steam parameters unchanged, changing the condenser back pressure by adjusting the circulating water flow rate or the vacuum breaker valve opening, recording the measured unit output value corresponding to each back pressure value, and obtaining a set of back pressure and output corresponding data for the typical load point; performing linear regression on the set of back pressure and output corresponding data to obtain the correction coefficient of unit output with back pressure change at the typical load point, and using this correction coefficient as the back pressure correction coefficient for the typical load point; repeating the above operation for each typical load point to obtain the back pressure correction coefficient for each typical load point; The back pressure correction coefficients for each typical load point are segmented and organized according to the load interval. Linear interpolation is then used to obtain the corresponding back pressure correction coefficients for load values ​​between adjacent typical load points. This includes: using the load values ​​of two adjacent typical load points as interval endpoints, and using the back pressure correction coefficients of each of the two typical load points as the back pressure correction coefficient values ​​at the interval endpoints; for any load value within the interval, linear interpolation is performed on the back pressure correction coefficient values ​​at the two endpoints based on the relative position of the load value and the load values ​​at the two interval endpoints to obtain the back pressure correction coefficient corresponding to the load value; the endpoint load values, endpoint back pressure correction coefficients, and interpolation rules for each load interval are combined to form a back pressure output correction relationship table. This back pressure output correction relationship table is used to determine the corresponding back pressure correction coefficient based on any load value in the construction of the pump operation control strategy model, and to obtain the unit output change by combining the back pressure change under that load.

[0078] It should be noted that the server collaborative control system conducted micro-output increase tests at three typical load points of the unit: 300MW, 450MW, and 600MW.

[0079] Taking a typical 450MW load point as an example, the total heat input of the main steam entering the turbine was kept stable. The condenser back pressure was changed by adjusting the speed of the variable frequency circulating water pump, and the measured output data was recorded. The data collected on site were as follows: when the back pressure was stable at 4.50 kPa, the measured unit output was 451.8 MW; when the back pressure rose to 5.00 kPa, the output was 450.0 MW; when the back pressure was 5.50 kPa, the output was 448.2 MW; and when the back pressure was 6.00 kPa, the output was 446.4 MW.

[0080] After acquiring the data, the server performed linear regression analysis on the data set. The calculation showed that for every 1.00 kPa change in condenser back pressure, the unit output changed by 3.6 MW. Therefore, 3.6 MW / kPa was determined as the back pressure correction factor for a typical load point of 450 MW.

[0081] Repeating the above steps, since the total steam flow is small at 300MW low load, the sensitivity of the last-stage blades to vacuum changes is low. The regression analysis yields a back pressure correction factor of 2.4MW / kPa for 300MW. At 600MW full load, the steam flow reaches its rated maximum, and the calculated back pressure correction factor is 4.8MW / kPa.

[0082] After obtaining the endpoint correction coefficients for three typical load intervals, the server segments the load according to its magnitude. For any load value between adjacent typical load points, a linear interpolation rule is used for conversion. When the unit is operating at a 400MW load, this value falls within the 300MW to 450MW range, with the distance from 300MW accounting for two-thirds of the entire range. The server performs linear interpolation between 2.4MW / kPa and 3.6MW / kPa, calculating the back pressure correction coefficient for 400MW to be 3.2MW / kPa.

[0083] The server encapsulates the endpoint load values, endpoint correction coefficients, and linear interpolation rules for each interval to form a back pressure output correction relationship table.

[0084] 204. Using the back pressure dynamic response model, pump operating hydraulic characteristic parameters and back pressure output correction relationship table, construct a pump operation control strategy with the minimum total coal consumption cost increment over the whole cycle as the comprehensive accounting index, and the pump start-up and shutdown frequency limit, minimum pump operation downtime and condenser vacuum upper limit as mechanical and process safety boundaries. The pump operation control strategy includes a pump impeller speed control strategy.

[0085] Specifically, the operation of constructing the pump operation control strategy model includes: discretizing the operation control cycle into multiple equal-length time periods, using the fluid distribution mode identifier (including the start-stop status of the fixed-frequency pump and the impeller speed command of the variable-frequency pump) in each time period as the optimization decision variable to obtain the sequence of operation command variables; for each time period, based on the load forecast value and the circulating water inlet temperature forecast value of that time period, using the back pressure dynamic response model to calculate the change process of the condenser back pressure with time in that time period, and combining the back pressure output correction relationship table to determine the change in unit output caused by the back pressure change in that time period; according to the pump combination number of that time period, looking up the corresponding total pump power consumption from the pump operation hydraulic characteristic parameters; converting the change in unit output in each time period into the fuel coal consumption increment, and converting the total pump power consumption in each time period into the coal consumption increment, forming a comprehensive energy consumption accounting model with the minimum total coal consumption cost increment of the whole cycle as the comprehensive accounting index; The limit on the number of pump start-stop cycles is transformed into an upper limit constraint on the number of changes in the operating command variables in adjacent time periods. The minimum pump operation and downtime is transformed into a constraint on the continuous operation time and continuous downtime of each pump. The upper limit of condenser vacuum is transformed into an upper limit constraint on the condenser pressure calculated by the back pressure response model. Specifically, the pump operation control strategy includes a pump impeller speed control strategy, the specific construction and implementation of which are as follows: (1) Establishing the speed calculation optimization logic: In the comprehensive energy consumption calculation model, the impeller speed ratio of the variable frequency pump in the variable frequency pump combination is used as the optimization variable for continuous or discrete growth. For any given impeller speed ratio, the system uses the similarity law of fluid mechanics to calculate the circulating water flow and total pump power consumption under the current pipeline resistance; the circulating water flow is input into the back pressure dynamic response model to obtain the back pressure change, and then combined with the back pressure output correction relationship table to obtain the unit output change; finally, the net coal consumption cost increment corresponding to the current impeller speed ratio is calculated, and the optimal impeller speed target value that minimizes the total coal consumption cost is locked through the optimization algorithm; (2) Setting speed control boundary constraints: In the mechanical and hydraulic operation boundary of the equipment, speed upper and lower limit constraints and speed change rate constraints are introduced for impeller speed control. The speed upper and lower limit constraints are used to limit the pump impeller speed to a minimum speed that prevents the pump from deviating from the high-efficiency zone or causing cavitation, and to a speed that is not higher than the rated maximum speed of the motor; the speed change rate constraint is used to limit the impeller acceleration and deceleration amplitude within adjacent adjustment periods to prevent water hammer effect in the pipeline network caused by sudden speed changes.

[0086] A discrete-time pump operation control strategy is constructed based on the sequence of operation command variables, the comprehensive energy consumption accounting model, and constraints including impeller speed constraints. This discrete-time pump operation control strategy serves as the object for subsequent rolling time-domain optimization.

[0087] Furthermore, the changes in unit output in each time period are converted into fuel and coal consumption increments, and the total power consumption of pumps in each time period is converted into coal consumption increments. This includes: obtaining the turbine heat rate of the unit under the current load and operating parameters, and obtaining the fuel consumption increments in each time period based on the changes in unit output in each time period and the turbine heat rate; obtaining the lower heating value and unit price of fuel, and obtaining the fuel and coal consumption increments in each time period based on the fuel consumption increments, lower heating value and unit price in each time period. Obtain the plant's overall power consumption rate. Based on the total pump power consumption and the plant's power consumption rate for each time period, obtain the coal consumption increment corresponding to the total pump power consumption for each time period. Add the fuel coal consumption increment for each time period to the corresponding pump power consumption coal consumption increment to obtain the comprehensive pump operating energy consumption for each time period. Accumulate the comprehensive pump operating energy consumption for all time periods to obtain the total coal consumption cost increment for the entire cycle. Use this as the comprehensive energy consumption accounting model value to solve the pump operation control strategy model in rolling time domain optimization.

[0088] It should be noted that the entire optimization cycle is set to the next 2 hours, discretized into 12 equal-length time periods, each with a duration of... The timeframe is set to 10 minutes (i.e., 0.167 hours). The server uses the fluid distribution mode identifiers that operate within these 12 time periods as optimization decision variables to form a discrete sequence of operating instruction variables.

[0089] For any time period The server retrieves the predicted data and inputs it into the back pressure dynamic response model, then uses the back pressure output correction relationship table to find the change in force. (Unit: kW); Simultaneously, refer to the operating characteristic table to obtain the total power consumption of the pump. (Unit: kW). To achieve uniformity of material and value dimensions in the comprehensive energy consumption accounting model, the server constructed the following calculation method for the incremental total coal consumption cost per time period. The formula: ; The parameter logic is as follows: The plant's power consumption rate is set at 0.05. The turbine heat rate is taken as 7800 kJ / kWh. The actual lower heating value of the fuel is taken as 21000 kJ / kg. The unit price of fuel is 1.0 yuan / kg.

[0090] Suppose that a decision at a certain time period increases the total pump power consumption by 2000kW and the unit output by 1800kW. Substitute these values ​​into the formula to calculate: the cost increment rate caused by the increased pump power consumption is... Yuan / h; Back pressure revenue savings rate Yuan / h. The net coal consumption cost increment is Yuan. The results of 12 time periods are summed to form a comprehensive energy consumption accounting model with the minimum increase in total coal consumption cost.

[0091] Simultaneously, three boundary constraints were set: the limit on the number of pump start-ups and shutdowns was transformed into an upper limit constraint that the total number of state switching times for any single pump within the entire cycle must not exceed one; the minimum operating downtime was transformed into a continuous time period constraint of at least 60 minutes of continuous operation and at least 30 minutes of continuous shutdown; and the upper limit of vacuum was transformed into a cutoff constraint that the predicted condenser back pressure value for all time periods must be ≥4.0 kPa. This combination was used to construct a discrete-time pump operation control strategy.

[0092] 205. Using a continuous sliding time-period regulation method, the initial state of the back pressure dynamic response model is updated with the measured back pressure at each decision time. Combined with the predicted load value, the predicted circulating water inlet temperature value, and the margin value determined by the statistical analysis of historical prediction errors, the pump operation control strategy model is solved to obtain the fluid resource allocation sequence for multiple future time periods. The optimized decision output of the transmission and distribution node for the current time period is output, and at the next decision time, the operation of updating the initial state of the back pressure dynamic response model with the measured back pressure is performed again to solve the problem.

[0093] Specifically, the operation of solving the pump operation control strategy model and issuing decisions using the continuous sliding time period adjustment method includes: collecting the measured back pressure value at the current decision moment, replacing the initial back pressure value in the back pressure dynamic response model with the measured back pressure value to obtain the model state at the current moment; obtaining the load forecast value and circulating water inlet temperature forecast value for multiple future time periods to form the load forecast sequence and inlet water temperature forecast sequence; determining the back pressure safety margin value and circulating water flow safety margin value based on the historical forecast error statistics; superimposing the back pressure safety margin value onto the threshold of the condenser vacuum upper limit constraint, and superimposing the circulating water flow safety margin value onto the threshold of the circulating water flow lower limit constraint to obtain the adjusted constraint conditions; Starting from the current model state, and combining the load prediction sequence, inlet water temperature prediction sequence, and adjusted constraints, the pump operation control strategy model is solved to obtain the fluid resource allocation sequence for multiple future time periods. The optimization decision of the current time period is extracted from the fluid resource allocation sequence for multiple future time periods, and the optimization decision of the current time period is sent to the circulating water pump for execution. After waiting for the next decision time to arrive, the operation of collecting the measured back pressure value at the current decision time is returned until the optimization cycle ends.

[0094] It should be noted that the current decision time is set to 10:00, the optimization period is the next 2 hours (a total of 12 10-minute intervals), and the server collects the current measured back pressure value of 4.95 kPa. Due to temporary fouling on the inner wall of the pipe causing a slightly lower actual heat transfer efficiency, the server immediately replaced the original initial state calculation benchmark of the back pressure model with 4.95 kPa.

[0095] Obtain forecast data and margins. Meteorological data shows that the circulating water inlet temperature will slowly rise from 25.0℃, and the dispatch plan requires the load to decrease from 450MW to 300MW at 10:30. The server retrieves historical forecast error statistics and determines the back pressure safety margin to be 0.10kPa and the flow rate safety margin to be 1000t / h. These margins are then added to the safety constraint thresholds: the lower limit of back pressure is increased from 4.00kPa to 4.10kPa, and the lower limit of flow rate is increased from 25000t / h to 26000t / h.

[0096] Starting with the updated 4.95 kPa, the server solves the pump operation control strategy model by combining the predicted sequence and adjusted constraints. The algorithm calculates the fluid resource allocation sequence that minimizes the increase in total coal consumption cost. Data from the first four time periods are shown in Table 2 below: Table 2

[0097] Extract the decision for the current first time period (10:00 to 10:10), and issue the instruction to start the A fixed-frequency pump and superimpose the C variable-frequency pump running at 90% speed to the actuator, without issuing subsequent actions.

[0098] After waiting 10 minutes to reach 10:10, the server reads the latest measured back pressure on site and overwrites and updates the initial state of the model; at the same time, the 2-hour window is shifted backward by 10 minutes (i.e., from 10:10 to 12:10), the latest prediction data is read, the solution is recalculated, and the first decision for this period is issued again to achieve closed-loop rolling control.

[0099] 206. Before constructing the pump operation control strategy model, obtain the measured curves of the circulating water flow rate and total pump power consumption over time during the switching of the circulating water pump from one distribution mode state to another pump combination. Extract the flow transition trajectory, power consumption transition trajectory, and the time required for the flow rate to reach the new steady-state value for each adjacent distribution mode state switch, and form a distribution mode state switch transition characteristic table. When constructing the comprehensive energy consumption accounting model, if the pump combination number changes in adjacent time periods, use the distribution mode state switch transition characteristic table to calculate the transition value of the circulating water flow rate and total pump power consumption during the switching process of the adjacent time period, and use the transition value to replace the steady-state value to calculate the change in unit output and total pump power consumption during that time period.

[0100] Furthermore, measured curves of circulating water flow rate and total pump power consumption over time are obtained during the switching of circulating water pumps from one distribution mode state to another pump combination. The flow transition trajectory, power consumption transition trajectory, and time required for the flow rate to reach a new steady-state value are extracted for each adjacent distribution mode state switch, forming a distribution mode state switch transition characteristic table. This includes: under the condition of stable unit load and stable circulating water inlet temperature, switching from each distribution mode state to each adjacent pump combination in sequence according to a pre-set distribution mode state switch test order. During each switch, the circulating water flow rate and total pump power consumption are recorded at a fixed sampling interval to obtain the measured flow response curve and power consumption response curve for each switch. Taking the start time of the switch action as the time zero reference, the measured flow response curves obtained from multiple switches from the same starting pump combination to the same ending pump combination are averaged point by point to obtain the average flow transition trajectory of the adjacent pump combination. The measured power consumption response curves obtained from multiple switching operations are averaged point by point to obtain the average power consumption transition trajectory of the adjacent pump combination. The shortest time required for the flow rate to reach and remain within the allowable deviation range of the new steady-state value is determined from the average flow transition trajectory and taken as the flow stabilization time of the adjacent pump combination. The starting pump combination number, ending pump combination number, average flow transition trajectory, average power consumption transition trajectory, and flow stabilization time of each adjacent pump combination are compiled into a table entry. All entries are summarized to obtain the transmission and distribution mode state switching transition characteristic table. During the construction of the pump operation control strategy model, when the pump combination numbers of adjacent time periods... At the same time, using the starting pump combination number and ending pump combination number of the adjacent time period as indexes, the corresponding average flow transition trajectory, average power consumption transition trajectory, and flow stabilization time are retrieved from the transmission and distribution mode state switching transition characteristic table; the time interval of the adjacent time period is aligned with the retrieved average flow transition trajectory and average power consumption transition trajectory in time, and the transition trajectory segment located within the adjacent time period is extracted; the average circulating water flow rate of the extracted transition trajectory segment is calculated by averaging the circulating water flow rate value of the segment, and the average total pump power consumption of the segment is calculated by averaging the total pump power consumption value of the segment. The average circulating water flow rate and average total pump power consumption are used to replace the original steady-state circulating water flow rate and steady-state total pump power consumption during this period, respectively, to calculate the change in unit output and total pump power consumption during this period.

[0101] It should be noted that the transmission and distribution mode switching test was carried out at a load of 450MW and a water temperature of 25.0℃.

[0102] The initial state is set as Combination 1 (single A fixed-frequency pump operating), with a circulating water flow rate of 40,000 t / h in this steady state. Based on the general principle of the positive correlation between flow rate and power consumption in fluid mechanics, the work load of a single A pump at a high flow rate of 40,000 t / h is higher than the load at the aforementioned 30,000 t / h shared flow rate. The measured total power consumption of the pump in this steady state is 2800 kW. The operator issues a command to switch to Combination 2 (A fixed-frequency pump superimposed with C variable-frequency pump operating at 90% speed), as in step 202. The total flow rate in this new steady state is 55,000 t / h, and the total power consumption is 3900 kW.

[0103] The server recorded data at 2-second intervals and repeated the test 5 times, averaging the data point by point. The server determined that the shortest time required for the flow rate to reach the allowable deviation range of 55,000 t / h was 180 seconds; the time required for the power consumption to reach 3,900 kW was 120 seconds. The average flow rate during the 180-second flow transition period was 48,000 t / h; the average power consumption during the 120-second power consumption transition period was 3,400 kW.

[0104] Repeat the above operations to obtain Table 3, which shows the transition characteristics of the power transmission and distribution mode state switching. Table 3

[0105] When constructing the optimization problem, the time step is set to 10 minutes (600 seconds). Assume that in a certain adjacent time period, the power output changes from combination 1 to combination 2. Directly assigning 65000 t / h and 4900 kW based on steady-state logic would result in severe distortion.

[0106] After introducing this mechanism, the server retrieves the transition characteristics. This 600-second period is then segmented and aligned. A time-weighted averaging method is used: for flow rate, 180 seconds is multiplied by the average of 48000 t / h, plus the remaining 420 seconds multiplied by the steady-state value of 55000 t / h, and the sum is divided by 600 seconds to obtain the average circulating water flow rate of 52900 t / h for this period. For power consumption, 120 seconds is multiplied by the average of 3400 kW, plus the remaining 480 seconds multiplied by the steady-state value of 3900 kW, and the sum is divided by 600 seconds to obtain the average total pump power consumption of 3800 kW.

[0107] The server used the calculated 52900t / h to replace the steady-state value of 55000t / h in the back pressure model to calculate the output force change; it used 3800kW to replace the steady-state value of 3900kW to calculate the power consumption increment, accurately restoring the water hammer effect in the pipeline network and the hydraulic delay caused by the large inertia of the fluid at the moment of pump start-up and shutdown.

[0108] Figure 5 This is a schematic diagram of the structure of a circulating water pump optimization device for a coal-fired power generation unit according to an embodiment of the present invention. The device 300 may include: a processor 301, a receiver 302, a transmitter 303, and a memory 304. The receiver 302, transmitter 303, and memory 304 are respectively connected to the processor 301 via a bus. It should be noted that in some possible implementations, the processor 301 and the memory 304 may be integrated together.

[0109] The processor 301 includes one or more processing cores. The processor 301 executes the methods performed by the base station in the random access method provided in this application embodiment by running software programs and modules. The memory 304 can be used to store software programs and modules. Specifically, the memory 304 can store an operating system 3041 and at least one application module 3042 required for a function. The receiver 302 is used to receive communication data sent by other devices, and the transmitter 303 is used to send communication data to other devices.

[0110] The present invention also provides a circulating water pump optimization device for coal-fired power generation units. The circulating water pump optimization device for coal-fired power generation units includes a memory and a processor. The memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, the processor performs the steps of the circulating water pump optimization method for coal-fired power generation units in the above embodiments.

[0111] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when the instructions are executed on a computer, cause the computer to perform the steps of the coal-fired generator set circulating water pump optimization method.

[0112] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0113] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0114] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for optimizing the circulating water pump of a coal-fired power generation unit, characterized in that, include: Acquire unit test data and establish a dynamic response model of condenser back pressure to changes in circulating water flow rate; Based on pump performance data and pipeline characteristic data, determine the total circulating water flow and total pump power consumption under each cooling water delivery configuration mode, and obtain the pump operating hydraulic characteristic parameters. Based on the unit operating characteristic data, determine the change in unit output caused by back pressure changes under different loads, and obtain the back pressure output correction relationship; Using the back pressure dynamic response model, the pump operating hydraulic characteristic parameters and the back pressure output correction relationship, a pump operation control strategy is constructed with the goal of optimizing the overall operating cost of the pump and system and with the mechanical and hydraulic operating boundaries of the equipment as constraints. The pump operation control strategy includes a pump impeller speed control strategy. By combining pipeline network prediction data and measured status parameters, dynamic parameter matching is performed on the pump operation control strategy to obtain a fluid resource allocation sequence, and the optimization decision of the transmission and distribution nodes for the current period is output.

2. The method for optimizing the circulating water pump of a coal-fired power generation unit according to claim 1, characterized in that, include: Extract the unit output data corresponding to different condenser back pressures from the unit operating characteristic data; Determine the change in unit output caused by back pressure changes at multiple typical load points, and obtain the back pressure correction coefficients corresponding to each of the multiple typical load points; The back pressure correction coefficients corresponding to each typical load point are organized according to the load range to form the back pressure output correction relationship.

3. The method for optimizing the circulating water pump of a coal-fired power generation unit according to claim 1, characterized in that, include: The operation and regulation cycle is discretized into multiple time periods, and the fluid transport and distribution mode identifier in each time period is used as the optimization decision variable; For each time period, the change in unit output during that time period is converted into fuel and coal consumption increment, and the total power consumption of the pump during that time period is converted into coal consumption increment. Based on the fuel and coal consumption increments and the coal consumption increments for all time periods, a comprehensive energy consumption accounting model is constructed with the minimum comprehensive energy consumption of the pump throughout the entire cycle as the comprehensive accounting index. The mechanical and hydraulic operating boundaries of the equipment are transformed into constraints for corresponding operating command variables. The mechanical and hydraulic operating boundaries of the equipment include: frequency limit for switching state of transmission and distribution resources, minimum state dwell time limit, and upper limit of condenser vacuum.

4. The method for optimizing the circulating water pump of a coal-fired power generation unit according to claim 1, characterized in that, include: A continuous sliding time-period adjustment method is adopted, and the measured back pressure value is collected at the current decision moment as the measured state parameter to update the initial state of the back pressure dynamic response model; The predicted load values ​​and circulating water inlet temperature values ​​for multiple future time periods are obtained as the predicted data, and the constraints are corrected in combination with the preset safety margin value. After updating the initial state and correcting the constraints, the pump operation control strategy is solved to obtain the fluid resource allocation sequence, and the optimal distribution node optimization decision output for the current time period is extracted.

5. The method for optimizing the circulating water pump of a coal-fired power generation unit according to claim 3, characterized in that, Also includes: Acquire the transitional characteristic data of circulating water flow rate and total pump power consumption over time during the switching of adjacent distribution mode states of the circulating water pump; When constructing the pump operation control strategy, if the transmission and distribution mode state switch occurs between adjacent time periods, the average transition value in the transition characteristic data is used to replace the steady-state value, and the change in unit output and the total power consumption of the pump during the corresponding time period are calculated.

6. The method for optimizing the circulating water pump of a coal-fired power generation unit according to claim 2, characterized in that, The process involves determining the change in unit output caused by back pressure variations at multiple typical load points, obtaining the back pressure correction coefficients corresponding to each of the multiple typical load points, and organizing the back pressure correction coefficients corresponding to each typical load point according to load intervals to form the back pressure output correction relationship, including: Linear regression is performed on the unit output data extracted under each typical load point to obtain the back pressure correction coefficient for the corresponding typical load point; Using the adjacent typical load points as interval endpoints, linear interpolation is used to obtain the corresponding back pressure correction coefficient for any load value between adjacent endpoints, thus forming the back pressure output correction relationship.

7. The method for optimizing the circulating water pump of a coal-fired power generation unit according to claim 3, characterized in that, The step of converting the change in unit output during the period into an increase in fuel consumption, and converting the total power consumption of the pumps during the period into an increase in coal consumption, includes: The turbine heat rate, the lower heating value of fuel, and the unit price are obtained. Based on the change in the unit output in each time period, the fuel consumption increment in each time period is calculated. Obtain the total plant power consumption rate and calculate the corresponding coal consumption increment based on the total power consumption of the pumps in each time period; The fuel consumption increment for each time period is added to the coal consumption increment for the corresponding time period to obtain the comprehensive operating energy consumption of the water pump for each time period.

8. The method for optimizing the circulating water pump of a coal-fired power generation unit according to claim 7, characterized in that, The incremental fuel consumption for each time period is added to the incremental coal consumption for the corresponding time period to obtain the comprehensive operating energy consumption of the water pump for each time period. The incremental total coal consumption cost for a single time period is calculated using the following formula. : ； In the formula, k is the time period number; Let k be the total power consumption of the pump during time period k; Let k be the change in unit output during time period k; This represents the plant's total power consumption rate. This refers to the turbine heat rate; The lower heating value of fuel; Price per unit; The duration of a single time period.

9. An optimization device for circulating water pumps in a coal-fired power generation unit, characterized in that, The coal-fired power generator set circulating water pump optimization device includes: a memory and at least one processor, wherein the memory stores instructions; The at least one processor invokes the instructions in the memory to cause the coal-fired generator set circulating water pump optimization device to perform the coal-fired generator set circulating water pump optimization method as described in any one of claims 1-8.

10. A computer-readable storage medium storing instructions thereon, characterized in that, When the instruction is executed by the processor, it implements the method for optimizing the circulating water pump of a coal-fired power generation unit as described in any one of claims 1-8.

Images