Hydropower station power generation model platform based on runoff mode

Abstract

The invention discloses a hydropower station power generation model platform based on a runoff mode, and the platform comprises the steps: automatically importing historical operation data, automatically carrying out the screening and arrangement of the historical data under the technical constraint conditions of a reservoir and a generator set, creating a power generation model, and storing the model and parameters; the platform can automatically generate a flow and tail water model, a flow and unit efficiency model and a unit power generation efficiency basic data report, can also provide a power generation plan reference value under the condition of forecasting the reservoir runoff volume, is used for formulating a daily power generation plan, and provides technical support for production, operation and maintenance management of a power station.

Description

Technical Field

[0001] The present invention belongs to the technical field of water conservancy and hydropower modeling, and in particular relates to a hydropower station power generation model platform based on a runoff mode. Background Art

[0002] The basic principle of reservoir scheduling and power generation scheduling of run-of-river hydropower stations is water balance, that is, when the inflow is less than or equal to the maximum water consumption of the unit for power generation, all inflows need to pass through the turbine; when the inflow is greater than the maximum water consumption of the unit for power generation, the unit generates electricity at maximum capacity, and the remaining flow is discharged through the flood discharge gate.

[0003] In order to maximize the power generation benefits, the runoff mode generally requires the reservoir water level to be maintained at a relatively high normal water storage level. During the flood season, the water level should also be maintained at a relatively high level while meeting the flood control water level requirements of the dam. Since the inflow flow of the power station changes in real time, accurately formulating power generation plans and controlling the balance of inflow and outflow flows are one of the main tasks of the runoff power station.

[0004] The traditional way to formulate power generation plans is to query the comprehensive characteristic curve of the turbine and use visual interpolation to obtain the unit efficiency under the corresponding head and unit reference flow, thereby calculating the power generation capacity of the power station. This is cumbersome and inaccurate, and it cannot reflect the change in the unit's power generation head caused by the change in tailwater level after the change in reservoir flow, and it is difficult to compare multiple plans, which has great limitations.

[0005] Therefore, it is necessary to design a hydropower station power generation model platform based on the runoff mode to solve the above problems. Summary of the invention

[0006] The technical problem to be solved by the present invention is to provide a hydropower station power generation model platform based on the runoff mode, and to build the unit power generation model platform by programming in Python language. By automatically importing historical operation data and automatically screening and sorting the historical data under technical constraints, a power generation model is created and the model and parameters are maintained, and a flow and tailwater model, a flow and unit efficiency model, and a basic data report on the unit power generation efficiency are automatically generated. The platform can also provide reference values ​​for power generation plans under predicted inflow runoff, providing a reference for formulating power generation plans.

[0007] In order to achieve the above technical effects, the technical solution adopted by the present invention is: A hydropower station power generation model platform based on the runoff mode, including a data processing module, a unit power generation model building module, a power generation plan making module and a sensor calibration module; The data processing module collects data from the original daily power generation report, including inflow and outflow, upstream and downstream water levels, power generation flow, total output and gross head, and filters them according to the set conditions to obtain a data set that meets the requirements for subsequent analysis and modeling; The unit power generation model construction module is used to construct the flow and tailwater level model and the flow and power generation efficiency model. The flow and tailwater level model is used to clarify the tailwater level change law and determine the unit power generation net head; the flow and power generation efficiency model is used to clarify the relationship between the unit power generation comprehensive efficiency under different operation modes; The power generation plan making module calculates the reference power generation capacity by citing flow, unit head, head loss and power generation efficiency to guide the daily power generation plan making; The sensor calibration module provides calibration functions for upstream and downstream water level sensors and unit power generation flow sensors. By adjusting the offset data of water level and flow, the power generation model is recreated for daily operation and maintenance.

[0008] Preferably, the execution method of the above-mentioned hydropower station power generation model platform based on the runoff mode is as follows: S1, data processing module performs historical data acquisition and recording: Record relevant production data in a fixed format, and read and process data in a fixed format through programming; S2, data processing module performs program data cleaning: Clean up abnormal operating data during production and operation, including: abnormal gate adjustment, abnormal water level control, power plant fault tripping, and human recording errors; S3, the unit power generation model building module determines the head loss Δh; S4, unit power generation model building module computer unit net water head H net : The net water head of each unit is calculated based on the reservoir water level, tail water level, and head loss of each unit, H net =H Gross -Δh; S5, the unit power generation model building module calculates the comprehensive operation efficiency η of the power station; S6, unit power generation model building module fitting inflow flow Q int ~Tailwater level L down Relational model; S7, the unit power generation model building module fits the outflow flow-efficiency model under different numbers of units in operation during the dry season; S8, the power generation plan making module determines the hourly power generation plan; S9, the power generation plan making module generates a daily power generation plan S10, the sensor calibration module performs water level meter and flow meter calibration; S11, according to the change of reservoir water level, appropriately adjust the machine end output or gate opening to adjust the outflow flow.

[0009] Preferably, in step S2, the data screening condition is: a. Filter out non-numeric data; b. Power generation flow Q power Satisfy n*Q min-one ≤Q power ≤n*Q max-one , Q min-one Q is the power generation reference flow rate of a single unit in a power plant under the minimum load corresponding to the operation avoiding the vibration zone. max-one is the maximum power generation reference flow of the unit, and n is the number of installed units in the power station; c. Total active power of the whole plant P total Satisfied in n*P min-one ≤P total ≤n*P max-one , P min-one It is the minimum load corresponding to the single unit running to avoid the vibration zone, P max-one is the maximum allowable active power of a single machine, and n is the number of installed units in the power station; d. The overall operating efficiency η satisfies η min ≤η<η max , when the reservoir is operating normally, ηmin should not be less than 0.8, ηmax=1, that is, the comprehensive operating efficiency of the unit is less than 1; e. Tailwater level of downstream powerhouse L down Meet L down-min ≤L down ≤L down-max , L down-min It is the tailwater level or the designed minimum tailwater level when all units of the power station are shut down. down-max It is the tailwater level value under the design flood of the power station; f. Power station gross water head H Gross Satisfy H Gross-min ≤H Gross ≤ H Gross-max ;H Gross-min The minimum operating head of the unit design, H Gross-max It is the maximum design operating head of the power station. The actual water level of the runoff power station is often far from the design water level during actual operation. The range can be adjusted according to the actual situation. g. Upstream reservoir water level L up Meet L up-min ≤L up ≤ L up-max , L up-min The upstream water level when the minimum head is designed is usually the dead water level of the reservoir. up-max Design flood levels for reservoirs; h. Upstream reservoir water level L up - Downstream water level L down -Rough water head H Gross The data with an absolute value exceeding 0.1m should normally be 0. Considering the water level fluctuation, the absolute value difference should be less than 0.1m to avoid data transcription errors. i. Outbound flow Q out -Inbound flow Q int Absolute value ≤ Q max-one Avoid selecting data during temporary adjustments to unit load or gate opening, and avoid selecting data during periods of unbalanced inflow and outflow; j. Power generation flow Q power -Outbound flow Q out ≤Q max-one / 2, avoid selecting data during load adjustment; k. Other screening conditions will be adjusted according to actual conditions.

[0010] Preferably, in step S3, the head loss Δh of the hydropower station includes the local head loss of the water inlet and the trash rack and the loss along the water diversion penstock. The power generation reference flow Q corresponding to the minimum load under the operation of a single unit avoiding the vibration zone is obtained by linear interpolation. min-one Maximum application flow Q of a single machine max-one Perform linear interpolation, specifically: Through the Python built-in keyword def, combined with the specific values ​​given in the reference file, a head loss function is defined to calculate the head loss target through the unit reference flow.

[0011] Preferably, in step S5, formula P is used total =ρgηH net Q power , that is, using the total active power of the power station P total , the gravity acceleration g in the area where the point is located, and the water density ρ of the river where it is located. These two data can be obtained from the feasibility study report; the power generation flow Q power , the comprehensive operation efficiency η of the power station is calculated; the comprehensive operation efficiency of the power station, including the efficiency of the turbine and the generator, can more intuitively display the efficiency of the hydropower station in converting the potential energy of water into electrical energy compared to the comprehensive output coefficient K value of the hydropower station.

[0012] Preferably, in step S6, the original data is divided into the dry season and the flood season according to whether the inflow is less than or equal to the full flow of the unit, or greater than the full flow of the unit. In the flood season, when the inflow is greater than the maximum reference flow of the unit, the power station should generate electricity according to the total number of installed units n and the rated installed capacity; at this time, the main factor affecting the unit efficiency is the unit head. Under the premise that the upstream water level remains unchanged by default, the downstream tailwater level is jointly affected by the power generation flow and the flood discharge flow: a. Determine the dry season Q int ~ L down Relationship model: In the dry season, the inflow, outflow and power generation flow are equal. If the power plant unit is equipped with an ultrasonic flowmeter, the power generation flow measured by the ultrasonic flowmeter can be directly used as the outflow flow. b. Determine the flood season Q int ~ L down Relational Model: During the flood season, since the opening of flood discharge structures is somewhat selective, it is necessary to consider whether the application of different flood discharge facilities has the same impact on the tailwater level after the project design and construction is completed. If there is a significant difference, it is necessary to group the pen tube flood discharge facilities when recording historical production data and consider the model relationship under different circumstances; Polynomial regression fitting is performed using the polyld function in the numpy library, and the highest order is set to 2. After the first fitting result, three times the standard deviation is used as the boundary to remove outliers far from the mean to ensure that the values ​​in the data set are within a reasonable range, thereby improving the quality and reliability of the data. Then, the data after the removal is selected for a second fitting to obtain the final curve result for subsequent analysis and calculation.

[0013] Preferably, in step S7, the generator set is divided into two stages: dry season and flood season, based on whether the generator set reaches full power. During the dry season, all inflows should flow through the generator set, so the outflow is equal to the power generation flow. Since the outflows are different for different numbers of running units, the water heads of the units are different, and the corresponding comprehensive operating efficiencies are also different, which need to be treated differently. Therefore, the actual number of running units needs to be recorded during the production operation data recording process so that it can be used as a variable for analysis when creating the model. When the installed capacity of the power station is n units, it is necessary to consider data screening again, and the screening condition is the actual number of units in operation n real , and the active power P in the power station total Should satisfy n real *P min-one ≤P total ≤n real *P max-one ; Get n respectively real After the data of the unit operation is obtained, the flow rate Q is stored according to the specified independent variable int , and the dependent variable comprehensive operating efficiency η is used for regression fitting.

[0014] Preferably, in step S8, the specific method for determining the hourly power generation plan is: a. Obtain the inflow flow, which can be obtained by measuring the inflow flow station or by forecasting; b. Obtaining the expected tailwater level: Based on the inflow data, in the runoff mode, the expected tailwater level can be calculated according to the flow-tailwater level relationship model in step 7 above; c. Obtaining the expected unit net water head: In the net flow mode, the upstream water level remains stable, and the unit's expected head loss can be determined by the quoted flow of a single unit. The unit net water head is calculated according to steps S3 and S4; d. Obtaining the expected comprehensive efficiency of the power plant: Determine the expected comprehensive power generation efficiency based on the inflow flow data and the flow-efficiency model obtained in step S6; e. Calculate the expected active power reference value; according to the formula P in step S5 总 =ρgH net Q power η, the expected total active power of the reference power station is calculated.

[0015] Preferably, in step S9, by importing the 24-hour inflow forecast value within a day, and automatically generating a daily reference power generation plan according to the current operation mode, i.e., the upstream water level and the power load of the plant, after obtaining the whole plant output plan, it is saved as an Excel file for the user to export.

[0016] Preferably, in step S10, the water level sensor and the power generation flow sensor offset calibration function is performed: In the actual production and operation of the power station, the tailwater river channel changes, unit efficiency changes, water level and flow sensor analog drift may cause inaccurate historical data. A certain offset is added to the historical data of different periods for correction; mainly including: Added data offset of upstream and downstream water level meters and generator flow meters; According to the characteristics of data in different periods, data can be processed and saved in different situations and batches; The batch-processed data is re-imported into the platform to automatically regenerate new power generation models and unit efficiency tables.

[0017] The beneficial effects of the present invention are as follows: 1. The model realizes the rolling fitting of data, and the original power generation data can be freely selected for model creation. When creating the model, the runoff mode in the dry season, the different runoff modes in the flood season, and the operation conditions of different numbers of units under the runoff mode in the dry season are distinguished, achieving a better fitting effect and ensuring the accuracy of the model.

[0018] 2. The model platform can correct, filter and maintain historical data, make full use of historical production data, and flexibly create model parameters for different periods. It can also automatically derive the tailwater level relationship curve and the power station comprehensive efficiency curve, which intuitively shows the historical hydrology and power generation of the power station, can determine the maximum power generation efficiency of the power station, and improve the power generation efficiency of the power station; at the same time, by comparing the saved historical models, it can give prompts for changes in unit efficiency and tailwater level changes, providing a reference for unit maintenance and river dredging; the model can realize the one-piece production of power generation plans, which has improved the level of automated management of hydropower stations to a certain extent.

[0019] 3. The model fully considers the particularity of run-of-river hydropower stations that need to maintain a balance between inflow and outflow, and proposes a two-dimensional model of inflow flow and unit efficiency. The parameters can be adjusted according to the actual characteristics of similar hydropower stations, which determines its wide potential for promotion and application. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a schematic diagram of the power generation model platform software interface in an embodiment of the present invention; Figure 2 Schematic diagram of a flow-tailwater level regression model in an embodiment of the present invention; Figure 3 Schematic diagram of a flow-unit efficiency regression model in an embodiment of the present invention; Figure 4 is a schematic diagram of a sensor calibration function in an embodiment of the present invention; Figure 5 A schematic diagram of a flow chart for creating a power generation model under a runoff mode in an embodiment of the present invention; Figure 6 It is a schematic diagram of a flow chart of power generation plan calculation under the runoff mode in an embodiment of the present invention. DETAILED DESCRIPTION

[0021] Embodiment 1: A hydropower station power generation model platform based on the runoff mode, including a data processing module, a unit power generation model building module, a power generation plan making module and a sensor calibration module; The data processing module collects data from the original daily power generation report, including inflow and outflow, upstream and downstream water levels, power generation flow, total output and gross head, and filters them according to the set conditions to obtain a data set that meets the requirements for subsequent analysis and modeling; The unit power generation model construction module is used to construct the flow and tailwater level model and the flow and power generation efficiency model. The flow and tailwater level model is used to clarify the tailwater level change law and determine the unit power generation net head; the flow and power generation efficiency model is used to clarify the relationship between the unit power generation comprehensive efficiency under different operation modes; The power generation plan making module calculates the reference power generation capacity by citing flow, unit head, head loss and power generation efficiency to guide the daily power generation plan making; The sensor calibration module provides calibration functions for upstream and downstream water level sensors and unit power generation flow sensors. By adjusting the offset data of water level and flow, the power generation model is recreated for daily operation and maintenance.

[0022] Preferably, the execution method of the above-mentioned hydropower station power generation model platform based on the runoff mode is as follows: S1, data processing module performs historical data acquisition and recording: Record relevant production data in a fixed format, and read and process data in a fixed format through programming; S2, data processing module performs program data cleaning: Clean up abnormal operating data during production and operation, including: abnormal gate adjustment, abnormal water level control, power plant fault tripping, and human recording errors; S3, the unit power generation model building module determines the head loss Δh; S4, unit power generation model building module computer unit net water head H net : The net water head of each unit is calculated based on the reservoir water level, tail water level, and head loss of each unit, H net =H Gross -Δh; S5, the unit power generation model building module calculates the comprehensive operation efficiency η of the power station; S6, unit power generation model building module fitting inflow flow Q int ~Tailwater level L down Relational model; S7, the unit power generation model building module fits the outflow flow-efficiency model under different numbers of units in operation during the dry season; S8, the power generation plan making module determines the hourly power generation plan; S9, the power generation plan making module generates a daily power generation plan S10, the sensor calibration module performs water level meter and flow meter calibration; S11, according to the change of reservoir water level, appropriately adjust the machine end output or gate opening to adjust the outflow flow.

[0023] Preferably, in step S2, the data screening condition is: a. Filter out non-numeric data; b. Power generation flow Q power Satisfy n*Q min-one ≤Q power ≤n*Q max-one , Qmin-one Q is the power generation reference flow rate of a single unit in a power plant under the minimum load corresponding to the operation avoiding the vibration zone. max-one is the maximum power generation reference flow of the unit, and n is the number of installed units in the power station; c. Total active power of the whole plant P total Satisfied in n*P min-one ≤P total ≤n*P max-one , P min-one It is the minimum load corresponding to the single unit running to avoid the vibration zone, P max-one is the maximum allowable active power of a single machine, and n is the number of installed units in the power station; d. The overall operating efficiency η satisfies η min ≤η<η max , when the reservoir is operating normally, ηmin should not be less than 0.8, ηmax=1, that is, the comprehensive operating efficiency of the unit is less than 1; e. Tailwater level of downstream powerhouse L down Meet L down-min ≤L down ≤L down-max , L down-min It is the tailwater level or the designed minimum tailwater level when all units of the power station are shut down. down-max It is the tailwater level value under the design flood of the power station; f. Power station gross water head H Gross Satisfy H Gross-min ≤H Gross ≤ H Gross-max ;H Gross-min The minimum operating head of the unit design, H Gross-max It is the maximum design operating head of the power station. The actual water level of the runoff power station is often far from the design water level during actual operation. The range can be adjusted according to the actual situation. g. Upstream reservoir water level L up Meet L up-min ≤L up ≤ L up-max , L up-min The upstream water level when the minimum head is designed is usually the dead water level of the reservoir. up-max Design flood levels for reservoirs; h. Upstream reservoir water level L up - Downstream water level L down -Rough water head H Gross The data with an absolute value exceeding 0.1m should normally be 0. Considering the water level fluctuation, the absolute value difference should be less than 0.1m to avoid data transcription errors. i. Outbound flow Q out -Inbound flow Q int Absolute value ≤ Q max-oneAvoid selecting data during temporary adjustments to unit load or gate opening, and avoid selecting data during periods of unbalanced inflow and outflow; j. Power generation flow Q power -Outbound flow Q out ≤Q max-one / 2, avoid selecting data during load adjustment; k. Other screening conditions will be adjusted according to actual conditions.

[0024] Preferably, in step S3, the head loss Δh of the hydropower station includes the local head loss of the water inlet and the trash rack and the loss along the water diversion penstock. The power generation reference flow Q corresponding to the minimum load under the operation of a single unit avoiding the vibration zone is obtained by linear interpolation. min-one Maximum application flow Q of a single machine max-one Perform linear interpolation, specifically: Through the Python built-in keyword def, combined with the specific values ​​given in the reference file, a head loss function is defined to calculate the head loss target through the unit reference flow.

[0025] Preferably, in step S5, formula P is used total =ρgηH net Q power , that is, using the total active power of the power station P total , the gravity acceleration g in the area where the point is located, and the water density ρ of the river where it is located. These two data can be obtained from the feasibility study report; the power generation flow Q power , the comprehensive operation efficiency η of the power station is calculated; the comprehensive operation efficiency of the power station, including the efficiency of the turbine and the generator, can more intuitively display the efficiency of the hydropower station in converting the potential energy of water into electrical energy compared to the comprehensive output coefficient K value of the hydropower station.

[0026] Preferably, in step S6, the original data is divided into the dry season and the flood season according to whether the inflow is less than or equal to the full flow of the unit, or greater than the full flow of the unit. In the flood season, when the inflow is greater than the maximum reference flow of the unit, the power station should generate electricity according to the total number of installed units n and the rated installed capacity; at this time, the main factor affecting the unit efficiency is the unit head. Under the premise that the upstream water level remains unchanged by default, the downstream tailwater level is jointly affected by the power generation flow and the flood discharge flow: a. Determine the dry season Q int ~ L down Relationship model: In the dry season, the inflow, outflow and power generation flow are equal. If the power plant unit is equipped with an ultrasonic flowmeter, the power generation flow measured by the ultrasonic flowmeter can be directly used as the outflow flow. b. Determine the flood season Q int ~ L down Relational Model: During the flood season, since the opening of flood discharge structures is somewhat selective, it is necessary to consider whether the application of different flood discharge facilities has the same impact on the tailwater level after the project design and construction is completed. If there is a significant difference, it is necessary to group the pen tube flood discharge facilities when recording historical production data and consider the model relationship under different circumstances; Polynomial regression fitting is performed using the polyld function in the numpy library, and the highest order is set to 2. After the first fitting result, three times the standard deviation is used as the boundary to remove outliers far from the mean to ensure that the values ​​in the data set are within a reasonable range, thereby improving the quality and reliability of the data. Then, the data after the removal is selected for a second fitting to obtain the final curve result for subsequent analysis and calculation.

[0027] Preferably, in step S7, the generator set is divided into two stages: dry season and flood season, based on whether the generator set reaches full power. During the dry season, all inflows should flow through the generator set, so the outflow is equal to the power generation flow. Since the outflows are different for different numbers of running units, the water heads of the units are different, and the corresponding comprehensive operating efficiencies are also different, which need to be treated differently. Therefore, the actual number of running units needs to be recorded during the production operation data recording process so that it can be used as a variable for analysis when creating the model. When the installed capacity of the power station is n units, it is necessary to consider data screening again, and the screening condition is the actual number of units in operation n real , and the active power P in the power station total Should satisfy n real *P min-one ≤P total ≤n real *P max-one ; Get n respectively real After the data of the unit operation is obtained, the flow rate Q is stored according to the specified independent variable int , and the dependent variable comprehensive operating efficiency η is used for regression fitting.

[0028] Preferably, in step S8, the specific method for determining the hourly power generation plan is: a. Obtain the inflow flow, which can be obtained by measuring the inflow flow station or by forecasting; b. Obtaining the expected tailwater level: Based on the inflow data, in the runoff mode, the expected tailwater level can be calculated according to the flow-tailwater level relationship model in step 7 above; c. Obtaining the expected unit net water head: In the net flow mode, the upstream water level remains stable, and the unit's expected head loss can be determined by the quoted flow of a single unit. The unit net water head is calculated according to steps S3 and S4; d. Obtaining the expected comprehensive efficiency of the power plant: Determine the expected comprehensive power generation efficiency based on the inflow flow data and the flow-efficiency model obtained in step S6; e. Calculate the expected active power reference value; according to the formula P in step S5 总 =ρgH net Q power η, the expected total active power of the reference power station is calculated.

[0029] Preferably, in step S9, by importing the 24-hour inflow forecast value within a day, and automatically generating a daily reference power generation plan according to the current operation mode, i.e., the upstream water level and the power load of the plant, after obtaining the whole plant output plan, it is saved as an Excel file for the user to export.

[0030] Preferably, in step S10, the water level sensor and the power generation flow sensor offset calibration function is performed: In the actual production and operation of the power station, the tailwater river channel changes, unit efficiency changes, water level and flow sensor analog drift may cause inaccurate historical data. A certain offset is added to the historical data of different periods for correction; mainly including: Added data offset of upstream and downstream water level meters and generator flow meters; According to the characteristics of data in different periods, data can be processed and saved in different situations and batches; The batch-processed data is re-imported into the platform to automatically regenerate new power generation models and unit efficiency tables.

[0031] Embodiment 2: Table 1 below shows the daily production report of the power plant, which is used to record hourly production data, mainly including time, upstream water level, downstream water level, inflow, outflow, power generation flow, total plant active power, operating unit number identification, etc. This data can be adjusted according to the power plant situation to achieve model creation under different operating modes.

[0032]

[0033] Table 1: Reference format of daily production report Figure 1 It is the software interface of the power generation model platform. After the model is created, the software can automatically calculate the downstream tailwater level, unit head, head loss, unit efficiency, unit power generation flow, and unit active power under the corresponding relationship by filling in the inflow flow, reservoir water level and the number of operating units, which can be used as reference values ​​for power generation plan making.

[0034] Model creation: pass Figure 1Select Create Model from the menu, import the historical production data report, and the model parameters will be automatically generated and the sorted historical data of the model will be automatically saved. At the same time, by selecting the historical model, you can also load the models created in different periods, which is convenient for comparing models in different periods.

[0035] Sensor calibration: In the daily production process, the data of upstream and downstream water level sensors or power generation flow sensors may be offset and inaccurate, and need to be readjusted. Therefore, a sensor offset function is added to correct historical data and recreate the power generation model.

[0036] Model and historical data export: pass Figure 1 After creating the model, the software can automatically export the sorted operation history data with time tags, which can be used for data comparison. It can also export the flow-tailwater level regression model, such as Figure 2 As shown; and the flow-generation efficiency regression model, as Figure 3 shown.

[0037] It is also possible to export a basic efficiency table based on an Excel table made according to different flow intervals, as shown in Table 2, for actual production use.

[0038]

[0039] Table 2: Efficiency basic data table Daily power generation plan formulation; through Figure 1 The software can automatically generate a daily power generation plan reference report by importing the pre-filled forecast flow data into the program, which can be used for daily power generation plan submission and adjustment.

[0040] Figure 2 The flow-tailwater level regression model shown can be used to study the relationship between tailwater levels under different outflow flows. By comparing with the design model and models of different periods, the actual state of the tailwater level can be found, providing a reference for downstream river cleaning. At the same time, during the flood season, the impact of the flow state of the flood discharge gate on the tailwater level of the powerhouse under different flood discharge flows can also be studied to provide technical support for production operation and maintenance. Figure 3 The flow-unit efficiency regression model shown in the figure shows the relationship between the comprehensive efficiency of the power plant under different operating unit conditions. Through this model, the optimal operating mode in the power plant production process can be sought to improve the power generation efficiency.

[0041] Table 2 shows the basic efficiency table under different flow ranges, which can be used to make power generation plans based on Excel, making it easier for production dispatchers to formulate and adjust power generation plans. It can also provide power grid companies (power purchasers) with a reference for power generation capacity under the capacity electricity price and runoff mode mechanism, and has realized the verification of the reliability indicators of power stations by power purchasers.

[0042] Figure 4 The figure shows the sensor calibration function. Considering that the sensors of upstream and downstream water levels, power generation flow and other sensors may have been readjusted due to installation and failure in different operating periods, it is necessary to recalibrate and organize the historical data. The program can automatically organize and merge data from different periods, and then regenerate the power generation model through the calibrated data.

[0043] Table 3 shows the power generation plan report. This program can automatically import the forecast inflow flow for a certain period of time, automatically calculate and generate the power generation plan report for the corresponding period of time for reference by production and operation personnel.

[0044]

[0045] Table 3: Power Generation Plan Report Figure 5 The figure shows the flow chart of power generation model creation under runoff mode. According to the production data recorded in the historical production data report, mainly including inflow, outflow, power generation flow, upstream water level, downstream water level and number of units in operation, the data is screened through programming, the relationship between outflow and downstream water level is fitted, and the relationship between power generation flow and efficiency under different numbers of units in operation is fitted, so as to establish the power generation model.

[0046] Figure 6 The figure shows the flow chart of power generation plan calculation under runoff mode. By predicting the inflow, real-time upstream water level and the actual number of units in operation, the power generation model created above is used to automatically calculate the power generation plan under the current operating conditions.

Claims

1. A hydropower station power generation model platform based on runoff mode, characterized in that: It includes data processing module, unit power generation model building module, power generation plan making module and sensor calibration module; The data processing module collects data from the original daily power generation report, including inflow and outflow, upstream and downstream water levels, power generation flow, total output and gross head, and filters them according to the set conditions to obtain a data set that meets the requirements for subsequent analysis and modeling; The unit power generation model construction module is used to construct the flow and tailwater level model and the flow and power generation efficiency model. The flow and tailwater level model is used to clarify the tailwater level change law and determine the unit power generation net head; the flow and power generation efficiency model is used to clarify the relationship between the unit power generation comprehensive efficiency under different operation modes; The power generation plan making module calculates the reference power generation capacity by citing the flow rate, unit head, head loss and power generation efficiency to guide the daily power generation plan making; The sensor calibration module provides calibration functions for upstream and downstream water level sensors and unit power generation flow sensors. By adjusting the offset data of water level and flow, the power generation model is recreated for daily operation and maintenance.

2. A hydropower station power generation model platform based on runoff mode according to claim 1, characterized in that: The execution method is as follows: S1, data processing module performs historical data acquisition and recording: Record relevant production data in a fixed format, and read and process data in a fixed format through programming; S2, data processing module performs program data cleaning: Clean up abnormal operating data during production and operation, including: abnormal gate adjustment, abnormal water level control, power plant fault tripping, and human recording errors; S3, the unit power generation model building module determines the head loss Δh; S4, unit power generation model building module computer unit net water head H net : The net water head of each unit is calculated based on the reservoir water level, tail water level, and head loss of each unit, H net =H Gross -Δh; S5, the unit power generation model building module calculates the comprehensive operation efficiency η of the power station; S6, unit power generation model building module fitting inflow flow Q int ~Tailwater level L down Relational model; S7, the unit power generation model building module fits the outflow flow-efficiency model under different numbers of units in operation during the dry season; S8, the power generation plan making module determines the hourly power generation plan; S9, the power generation plan making module generates a daily power generation plan S10, the sensor calibration module performs water level meter and flow meter calibration; S11, according to the change of reservoir water level, appropriately adjust the machine end output or gate opening to adjust the outflow flow.

3. A hydropower station power generation model platform based on runoff mode according to claim 2, characterized in that: In step S2, the data screening conditions are: a. Filter out non-numeric data; b. Power generation flow Q power Satisfy n*Q min-one ≤Q power ≤n*Q max-one ,Q min-one Q is the power generation reference flow rate of a single unit in a power plant under the minimum load corresponding to the operation avoiding the vibration zone. max-one is the maximum power generation reference flow of the unit, and n is the number of installed units in the power station; c. Total active power of the whole plant P total Satisfied in n*P min-one ≤P total ≤n*P max-one , P min-one It is the minimum load corresponding to the single unit running to avoid the vibration zone, P max-one is the maximum allowable active power of a single machine, and n is the number of installed units in the power station; d. The overall operating efficiency η satisfies η min ≤η<η max , when the reservoir is operating normally, ηmin should not be less than 0.8, ηmax=1, that is, the comprehensive operating efficiency of the unit is less than 1; e. Tailwater level of downstream powerhouse L down Meet L down-min ≤L down ≤L down-max , L down-min It is the tailwater level or the designed minimum tailwater level when all units of the power station are shut down. down-max It is the tailwater level value under the design flood of the power station; f. Power station gross water head H Gross Satisfy H Gross-min ≤H Gross ≤ H Gross-max ;H Gross-min The minimum operating head of the unit design, H Gross-max It is the maximum design operating head of the power station. The actual water level of the runoff power station is often far from the design water level during actual operation. The range can be adjusted according to the actual situation. g. Upstream reservoir water level L up Meet L up-min ≤L up ≤ L up-max , L up-min The upstream water level when the minimum head is designed is usually the dead water level of the reservoir. up-max Design flood levels for reservoirs; h. Upstream reservoir water level L up - Downstream water level L down -Rough water head H Gross The data with an absolute value exceeding 0.1m should normally be 0. Considering the water level fluctuation, the absolute value difference should be less than 0.1m to avoid data transcription errors. i. Outbound flow Q out -Inbound flow Q int Absolute value ≤ Q max-one Avoid selecting data during temporary adjustments to unit load or gate opening, and avoid selecting data during periods of unbalanced inflow and outflow; j. Power generation flow Q power -Outbound flow Q out ≤Q max-one / 2, avoid selecting data during load adjustment; k. Other screening conditions will be adjusted according to actual conditions.

4. A hydropower station power generation model platform based on runoff mode according to claim 3, characterized in that: In step S3, the head loss Δh of the hydropower station includes the local head loss of the water inlet and the trash rack and the loss along the water diversion penstock. The power generation reference flow Q corresponding to the minimum load under the operation of a single unit to avoid the vibration zone is obtained by linear interpolation. min-one Maximum application flow Q of a single machine max-one Perform linear interpolation, specifically: Through the Python built-in keyword def, combined with the specific values ​​given in the reference file, a head loss function is defined to calculate the head loss target through the unit reference flow.

5. A hydropower station power generation model platform based on runoff mode according to claim 4, characterized in that: In step S5, using formula P total =ρgηH net Q power , that is, using the total active power of the power station P total , the gravity acceleration g in the area where the point is located, and the water density ρ of the river where it is located. These two data can be obtained from the feasibility study report; the power generation flow Q power , the comprehensive operation efficiency η of the power station is calculated; the comprehensive operation efficiency of the power station, including the efficiency of the turbine and the generator, can more intuitively display the efficiency of the hydropower station in converting the potential energy of water into electrical energy compared to the comprehensive output coefficient K value of the hydropower station.

6. A hydropower station power generation model platform based on runoff mode according to claim 5, characterized in that: In step S6, the original data is divided into dry season and flood season according to whether the inflow is less than or equal to the full flow of the unit, or greater than the full flow of the unit. In the flood season, when the inflow is greater than the maximum reference flow of the unit, the power station should generate electricity according to the total number of installed units n and the rated installed capacity; at this time, the main factor affecting the unit efficiency is the unit head. Under the premise that the upstream water level remains unchanged by default, the downstream tailwater level is jointly affected by the power generation flow and the flood discharge flow: a. Determine the dry season Q int ~ L down Relationship model: In the dry season, the inflow, outflow and power generation flow are equal. If the power plant unit is equipped with an ultrasonic flowmeter, the power generation flow measured by the ultrasonic flowmeter can be directly used as the outflow flow. b. Determine the flood season Q int ~ L down Relational Model: During the flood season, since the opening of flood discharge structures is somewhat selective, it is necessary to consider whether the application of different flood discharge facilities has the same degree of impact on the tailwater level after the project design and construction is completed. If there is a significant difference, it is necessary to group the pen tube flood discharge facilities when recording historical production data and consider the model relationship under different circumstances; Polynomial regression fitting is performed using the polyld function in the numpy library, and the highest order is set to 2. After the first fitting result, three times the standard deviation is used as the boundary to remove outliers far from the mean to ensure that the values ​​in the data set are within a reasonable range, thereby improving the quality and reliability of the data. Then, the data after the removal is selected for a second fitting to obtain the final curve result for subsequent analysis and calculation.

7. A hydropower station power generation model platform based on runoff mode according to claim 6, characterized in that: In step S7, the generator is divided into two stages: dry season and flood season, depending on whether the generator has reached full power. During the dry season, all inflows should flow through the generator set, so the outflow is equal to the power generation flow. Different outflows correspond to different numbers of running units, and the heads of the units are different, so the corresponding comprehensive operating efficiencies are also different and need to be treated differently. Therefore, the actual number of running units needs to be recorded during the production operation data recording process so that it can be used as a variable for analysis when creating the model. When the installed capacity of the power station is n units, it is necessary to consider data screening again, and the screening condition is the actual number of units in operation n real , and the active power P in the power station total Should satisfy n real *P min-one ≤P total ≤n real *P max-one ; Get n respectively real After the data of the unit operation is obtained, the flow rate Q is stored according to the specified independent variable int , and the dependent variable comprehensive operating efficiency η is used for regression fitting.

8. The hydropower station power generation model platform based on the runoff mode according to claim 7 is characterized in that: In step S8, the specific method for determining the hourly power generation plan is: a. Obtain the inflow flow, which can be obtained by measuring the inflow flow station or by forecasting; b. Obtaining the expected tailwater level: Based on the inflow data, in the runoff mode, the expected tailwater level can be calculated according to the flow-tailwater level relationship model in step 7 above; c. Obtaining the expected unit net water head: In the net flow mode, the upstream water level remains stable, and the unit's expected head loss can be determined by the quoted flow of a single unit. The unit net water head is calculated according to steps S3 and S4; d. Obtaining the expected comprehensive efficiency of the power plant: Determine the expected comprehensive power generation efficiency based on the inflow flow data and the flow-efficiency model obtained in step S6; e. Calculate the expected active power reference value; according to the formula P in step S5 总 =ρgH net Q power η, the expected total active power of the reference power station is calculated.

9. A hydropower station power generation model platform based on runoff mode according to claim 8, characterized in that: In step S9, the daily reference power generation plan is automatically generated by importing the predicted value of the 24-hour inflow flow within a day and according to the current operation mode, that is, the upstream water level and the power load of the plant. After obtaining the power output plan of the whole plant, it is saved as an Excel file for users to export.

10. A hydropower station power generation model platform based on runoff mode according to claim 9, characterized in that: In step S10, the water level sensor and power generation flow sensor offset calibration function is performed: In the actual production and operation of the power station, the tailwater river channel changes, unit efficiency changes, water level and flow sensor analog drift may cause inaccurate historical data. A certain offset is added to the historical data of different periods for correction; mainly including: Added data offset of upstream and downstream water level meters and generator flow meters; According to the characteristics of data in different periods, data can be processed and saved in different situations and batches; The batch-processed data is re-imported into the platform to automatically regenerate new power generation models and unit efficiency tables.

Images