A power generation plan execution deviation monitoring method and system

Abstract

The application provides a power generation plan execution deviation monitoring method and system, which is suitable for a hydroelectric generating set. The method comprises the following steps: acquiring a power generation plan, a real-time load instruction and actual operation data and checking; dynamically identifying formal modification of the power generation plan or a temporary load instruction of a power grid system; if the modification is a plan modification, monitoring a deviation in a modification period, calculating a deviation power and dynamically correcting a subsequent power generation task; if the modification is a temporary instruction, suspending original plan monitoring during the instruction period, calculating a deviation power after the instruction ends and judging whether the deviation power can be recovered; and optimizing a monitoring strategy by using a prediction model according to historical deviation data. The application can intelligently distinguish instruction types, automatically process plan modification and temporary instructions, dynamically correct subsequent plans, realize accurate monitoring and optimized scheduling of power generation plan execution deviation of the hydroelectric generating set, reduce examination risks, and improve the economic efficiency and safety of the hydroelectric operation.

Description

Technical Field

[0001] This invention relates to the field of power system automated dispatching technology, and in particular to a method and system for monitoring deviations in power generation plan execution. Background Technology

[0002] In power grid systems, grid dispatching agencies typically issue dispatching instructions to power plants based on day-ahead or intraday generation plans, combined with real-time load demand. Monitoring deviations in generation plan execution is a crucial step in ensuring the safe and stable operation of the power grid and achieving a dynamic balance between generation and load.

[0003] After the power grid system issues the power generation plan for cascade power plants in a river basin, the cascade power plants need to determine the start-up and shutdown of their generator units according to the plan and execute it promptly and accurately. When the power grid system modifies the power generation plan, the modified plan should be followed. If there are deviations in the execution of the power generation plan, it is necessary to promptly adjust the start-up and shutdown methods of the generator units or manually adjust the load of the generator units to avoid such situations and correct the deviations in the execution of the power generation plan in a timely manner.

[0004] If the power generation plan of the cascade power plants is not implemented properly, the power generation of each power plant will be lower than the power generation planned, and the power generation will not reach the planned power generation. This will lead to inadequate control of the water level in the reservoirs of the cascade power plants. In particular, when the water level is high, it will affect the arrangement of the power generation operation mode in the later stage. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a method and system for monitoring deviations in power generation plan execution, which can identify instruction types, automatically process plan modifications and temporary instructions, and dynamically correct subsequent plans.

[0006] According to an embodiment of the present invention, a method for monitoring deviations in power generation plan execution is provided, comprising the following steps:

[0007] S1 Data Acquisition and Verification: Acquire power generation plan data, real-time load instructions from the power grid system, and actual operating load data of power plants issued by the power grid system, and verify the validity of the data. The validity verification includes data integrity verification, timestamp consistency verification, and numerical range verification.

[0008] S2 Dynamic Command Recognition: Determines whether the power grid system has officially modified the current power generation plan or issued a temporary load command that is inconsistent with the current power generation plan, and dynamically switches the monitoring mode based on the judgment result;

[0009] S2A Power Generation Plan Modification Processing: If a power generation plan modification is detected, the following sub-steps are executed: A1 During the power generation plan modification period, real-time monitoring of power generation tasks is performed according to the modified power generation plan, and an alarm is issued when the execution deviation exceeds a preset threshold; A2 Calculate the deviation in electricity volume between the modified plan and the original plan during the power generation plan modification period; A3 Dynamically adjust the subsequent power generation tasks of the original power generation plan based on the deviation in electricity volume and the remaining time of the day, and generate load adjustment suggestions for each generator unit; A4 Automatically output subsequent power generation task adjustment schemes and send the load adjustment suggestions to the control systems of each unit for optimized scheduling;

[0010] S2B Temporary Load Command Processing: If determined to be a temporary load command, the following sub-steps are executed: B1 During the period when the temporary load command is in effect, the original power generation plan monitoring is suspended and power generation monitoring and deviation analysis are performed according to the temporary load command; B2 After the temporary load command ends, the deviation between the actual power generation and the original power generation plan in the corresponding time period is calculated; B3 Based on the deviation and the remaining time of the day, it is determined whether the deviation can be made up, and a subsequent power generation task correction plan and load adjustment suggestions for each unit are generated; B4 The load adjustment suggestions are automatically output to the unit control system to achieve execution optimization.

[0011] S3 Optimization and Feedback: Based on historical data of deviations in power generation plan execution, the system uses predictive models to optimize future power generation plan execution monitoring strategies, thereby achieving dynamic deviation compensation and optimized load scheduling.

[0012] Preferably, in S1, the data integrity verification includes checking whether each data source is missing necessary fields and whether the data records are continuous; the timestamp consistency verification includes determining whether the timestamp of each data is synchronized with a unified clock source and the interval conforms to the preset sampling period; the numerical range verification includes checking whether the load value is within the rated capacity range of the generator set and whether it exceeds the grid dispatch limit.

[0013] More preferably, in S2, the dynamic instruction identification adopts the following method: monitoring the message bus of the power grid dispatch automation system and parsing the message type field; if the message type is a plan modification and includes the modification effective period and the modified plan curve, it is determined to be a power generation plan modification; if the message type is a real-time instruction or an automatic power generation control instruction and the absolute value of the difference between the output value at the instruction execution time and the corresponding time value of the current effective plan curve exceeds the preset plan deviation tolerance, it is determined to be a temporary load instruction.

[0014] In a further preferred embodiment, in A3, when dynamically adjusting the subsequent power generation tasks of the original power generation plan based on the deviation power and the remaining time of the day, the deviation power is allocated to the basic planned value of each subsequent time period according to the proportion of the adjustment capacity of each unit, forming a corrected plan curve; the load adjustment suggestion includes the output value that each unit should increase or decrease in each subsequent time period and its adjustment rate requirement.

[0015] In a further preferred embodiment, in S2B, the method for determining whether the deviation can be compensated is as follows: calculate the total length of the remaining adjustable time period from the current time to the end of the day and the upper limit of the remaining adjustable capacity. If the sum of the maximum adjustable output increments of each unit in the remaining time period can cover the deviation power, it is determined to be compensateable, and a compensation plan is generated; otherwise, it is determined to be uncompensable, an adjustment suggestion that minimizes the deviation is generated, and it is marked as requiring scheduling intervention.

[0016] In a further preferred embodiment, in S3, the prediction model outputs the optimal deviation compensation coefficient and monitoring threshold adjustment suggestion for the next period based on the historical deviation sequence, load fluctuation rate, predicted output of new energy sources, grid frequency and tie-line exchange power.

[0017] In a further preferred embodiment, the prediction model collects actual execution deviation data daily at regular intervals and compares it with the prediction output, and fine-tunes the model weights based on the comparison results to achieve online adaptive updating of the model.

[0018] According to an embodiment of the present invention, a power generation plan execution deviation monitoring system is also provided, for the above-described power generation plan execution deviation monitoring method, comprising:

[0019] The data acquisition and verification module is used to acquire power generation plan data, real-time load instructions from the power grid system, and actual operating load data of power plants, and to verify their validity.

[0020] The dynamic command recognition module is used to determine whether the power grid system has made a formal modification to the current power generation plan or has issued a temporary load command that is inconsistent with the current power generation plan, and dynamically switches the monitoring mode.

[0021] The plan modification processing module is used to perform real-time monitoring, deviation alarm, deviation power calculation, subsequent dynamic correction of the plan, and output of load adjustment suggestions during the modification period when it is determined that the power generation plan has been modified.

[0022] The temporary instruction processing module is used to perform monitoring pause and deviation analysis during the execution of the temporary instruction, calculate the deviation power after the instruction ends, determine the recoverability, and output subsequent task correction and load adjustment suggestions when the instruction is determined to be a temporary load instruction.

[0023] The optimization feedback module is used to optimize the monitoring strategy for future power generation plan execution based on historical data of deviations in power generation plan execution, thereby achieving dynamic deviation compensation and optimized load scheduling.

[0024] Preferably, the dynamic instruction recognition module makes the judgment by monitoring the message bus of the power grid dispatch automation system or by parsing the instruction marker field in the data acquisition and monitoring control system; the system also includes an interface adaptation unit to be compatible with the data formats and communication protocols of different dispatch master station systems.

[0025] Compared with the prior art, the present invention has the following beneficial effects:

[0026] When a plan is modified or a temporary instruction is issued, the system not only records the immediate deviation but also automatically calculates the deviation in electricity volume. Based on the remaining adjustment capacity, it dynamically adjusts the subsequent tasks of the original power generation plan, generates clear load adjustment suggestions, and issues them for execution. This mechanism proactively eliminates accumulated deviations, ensuring that the total daily power generation more closely matches dispatch requirements and reduces grid performance risks.

[0027] By identifying dynamic commands, the system can accurately distinguish between formal modifications to the power generation plan and temporary load commands, avoiding misreporting temporary commands as execution deviations, significantly reducing invalid alarms, and improving the reliability of the monitoring system and the work efficiency of operators. Detailed Implementation

[0028] The technical solutions of the present invention will be further described below with reference to the embodiments.

[0029] Example 1

[0030] This embodiment provides a method for monitoring deviations in the execution of power generation plans for hydro turbine units:

[0031] S1 Data Acquisition and Verification

[0032] The system acquires the following four types of data from the dispatch master station and the internal system of the hydropower station at a set sampling period: power generation plan data, real-time load command of the power grid, actual operating load data of the power plant, and hydrological data;

[0033] The above data is validated for validity, including: integrity check, checking whether each data source is missing, and issuing an alarm if more than three consecutive sampling points are missing; timestamp consistency check, synchronizing with the BeiDou or Global Positioning System clock, and correcting for deviations exceeding 200 milliseconds; and numerical range check, verifying whether the output is within the set range and checking the expected output based on the current head.

[0034] S2 Dynamic Command Recognition

[0035] The system monitors the message bus of the dispatch master station and the instructions of the automatic power generation control system inside the hydropower station.

[0036] If the message type is "plan modification" and includes the effective period and the modified curve, it is determined to be a power generation plan modification, and a plan modification flag is set.

[0037] If the message type is an automatic generation control command or a real-time load command, extract the command value and compare it with the current planned value. If the absolute value of the difference exceeds the preset tolerance threshold, it is determined to be a temporary load command.

[0038] S2A power generation plan modification processing

[0039] When the planned modification is in effect, perform the following sub-steps:

[0040] A1 Real-time monitoring and alarm: The deviation between the actual output of the computer unit and the modified planned value. If the actual output falls into the vibration zone while the planned value is not in the vibration zone, an alarm will be issued even if the deviation is less than the threshold. When the deviation exceeds the basic threshold and there are more than three consecutive sampling points, an alarm message will be generated, including the time, unit identification, deviation value, current head and suggested measures.

[0041] A2 calculates the deviation in electricity generation during the modification period. The modified planned value minus the original planned value is integrated over the modification period. A positive value indicates that more electricity needs to be generated, and a negative value indicates that less electricity needs to be generated.

[0042] A3 dynamically adjusts subsequent power generation tasks, setting the current time as the end time of the modification period, and the remaining time period as until 24:00 on the same day;

[0043] Obtain the original planned value for the remaining period. The revised planned value equals the original planned value plus the revision amount. The revision amount is allocated according to the load forecast coefficient for the remaining period.

[0044] When generating load adjustment suggestions for each unit, the focus is on avoiding vibration zones and optimizing efficiency. A dynamic programming algorithm is used to minimize the total water consumption of the entire plant while satisfying the vibration zone constraints of each unit. If the total power demand cannot completely avoid the vibration zone, the optimal approximate solution is output and a prompt is given.

[0045] A4 output and distribution generates files containing the revised planning curves and water level control recommendations, and distributes load adjustment recommendations to the local control units of each unit;

[0046] S2B Temporary Load Command Processing

[0047] When a temporary load instruction is determined, the following sub-steps are executed:

[0048] During the B1 temporary command period, monitor, set the temporary command effective flag, record the start time, command value and expected end time, suspend the original monitoring during the effective period, and use the temporary command value as the expected output. Before the hydropower unit executes a large temporary command, estimate the pressure change. If the pressure rise rate exceeds the set range threshold, issue a risk warning and suggest segmented adjustment.

[0049] After the B2 instruction ends, the deviation in electricity is calculated. After the temporary instruction ends, the integral deviation between the actual power generation and the original power generation plan in the corresponding time period is calculated.

[0050] B3 recoverability assessment and correction scheme, statistical analysis of the comprehensive regulation capacity of each unit in the remaining period, is subject to three constraints: upper and lower limits of output, expected upper limit of output under the current head, available water volume, and also needs to meet the rigid constraints of ecological flow.

[0051] If the deviation is positive, power generation needs to be reduced. Determine whether the sum of the maximum adjustable power and the remaining time is greater than or equal to the deviation.

[0052] If it is negative, power generation needs to be increased. Determine whether the sum of the maximum adjustable power and the remaining time is greater than or equal to the absolute value of the deviation power, and whether the water volume is sufficient.

[0053] If the power can be recovered, a correction plan is generated; if it cannot be recovered, the maximum recoverable power is calculated, and a prompt indicating that scheduling intervention is required is output.

[0054] B4 output load adjustment suggestions are provided with adjustment rate parameters to prevent problems caused by excessive speed control.

[0055] S3 Optimization and Feedback

[0056] The system performs an offline optimization calculation once a day at a set time, and the optimization calculation is based on historical deviation data;

[0057] The system reads the following data sequences from the historical database for the past thirty days: the actual deviation value of each sampling point, the occurrence time and modification range of planned modification events, the occurrence time, instruction range and actual tracking error of temporary instruction events, the head change sequence, the inflow sequence, the number of times the unit crosses the vibration zone, the degree of deviation of the unit efficiency curve, and the grid frequency deviation value.

[0058] These data are constructed into a time series sample set. In this embodiment, a gated recurrent unit is used to build a prediction model. Its structure is more concise than that of a long short-term memory network and is suitable for the computing resource conditions of hydropower stations.

[0059] The model structure is as follows: the input layer has a time window of 60 sampling points, representing historical data from the past minute; the gated recurrent unit layer contains 56 units, employing the hyperbolic tangent activation function with a dropout rate of 20%; followed by a fully connected layer containing 28 neurons, using a linear rectified function as the activation function; the output layer contains three neurons, outputting the deviation compensation coefficient for the next time period, the monitoring threshold adjustment, and the vibration zone crossing warning index, respectively, to indicate whether there is a risk of crossing the vibration zone in the future.

[0060] The model's loss function uses mean squared error plus a regularization term, and the optimizer employs an adaptive moment estimation algorithm. After model training, the system runs in inference mode: every ten seconds, the model inputs the most recent historical data and outputs compensation coefficients, threshold adjustment amounts, and early warning indices; the system obtains the feedforward compensated planned value based on the planned values ​​for subsequent moments of the original power generation plan. Simultaneously, the deviation alarm threshold is obtained based on the original threshold and the threshold adjustment amount.

[0061] When the warning index exceeds the set range, the system will prominently display a vibration zone risk warning on the monitoring interface, suggesting that the plan be proactively adjusted to avoid the vibration zone.

[0062] Example 2

[0063] This embodiment provides a power generation plan execution deviation monitoring system, which includes the following components:

[0064] The data acquisition and verification module connects the dispatch master station and the remote terminal unit of the power plant through a standardized communication interface. This module is responsible for performing all the data acquisition and verification functions described in the first step of the aforementioned embodiment.

[0065] The dynamic instruction recognition module is built into the message bus listener. It continuously listens to messages sent or sent by the scheduling master station, parses the instruction type field, and sets a plan modification flag or temporary instruction flag based on the parsing results for use by subsequent modules.

[0066] The plan modification processing module further comprises four sub-units: a real-time monitoring unit, used to calculate the deviation during the plan modification period and compare it with a threshold to generate an alarm; a deviation power calculation unit, used to integrate the difference between the modified plan and the original plan during the modification period; a subsequent plan correction unit, used to correct the subsequent plan based on the deviation power and remaining regulation capacity and generate unit load adjustment suggestions; and an instruction issuance unit, used to send the adjustment plan to the unit control system through the communication interface.

[0067] The temporary instruction processing module also contains four sub-units: a temporary instruction marking unit, used to record the start time, instruction value, and end time of temporary instructions, and to suspend the original plan monitoring during this period; a deviation power calculation unit, used to calculate the integral deviation between the actual power generation and the original plan after the temporary instruction ends; a recoverability judgment unit, used to assess whether the deviation can be recovered through adjustments within the remaining time period and to generate a corresponding correction plan; and an instruction issuance unit, whose function is similar to the issuance unit in the plan modification processing module.

[0068] The optimized feedback module includes a long short-term memory network model trainer, a model inference engine, and a threshold adaptive adjustment unit. The model trainer periodically reads data from the historical database for offline training; the model inference engine receives real-time feature data and outputs compensation coefficients and threshold adjustment amounts; the threshold adaptive adjustment unit dynamically modifies the deviation alarm threshold based on the output values.

[0069] Finally, it should be noted that the above 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 preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for monitoring deviations in power generation plan execution, characterized in that, Includes the following steps: S1 Data Acquisition and Verification: Acquire power generation plan data, real-time load instructions from the power grid system, and actual operating load data of power plants issued by the power grid system, and verify the validity of the data. The validity verification includes data integrity verification, timestamp consistency verification, and numerical range verification. S2 Dynamic Command Recognition: Determines whether the power grid system has officially modified the current power generation plan or issued a temporary load command that is inconsistent with the current power generation plan, and dynamically switches the monitoring mode based on the judgment result; S2A Power Generation Plan Modification Processing: If a power generation plan modification is detected, the following sub-steps are executed: A1 During the power generation plan modification period, real-time monitoring of power generation tasks is performed according to the modified power generation plan, and an alarm is issued when the execution deviation exceeds a preset threshold; A2 Calculate the deviation in electricity volume between the modified plan and the original plan during the power generation plan modification period; A3 Dynamically adjust the subsequent power generation tasks of the original power generation plan based on the deviation in electricity volume and the remaining time of the day, and generate load adjustment suggestions for each generator unit; A4 Automatically output subsequent power generation task adjustment schemes and send the load adjustment suggestions to the control systems of each unit for optimized scheduling; S2B Temporary Load Command Processing: If determined to be a temporary load command, the following sub-steps are executed: B1 During the period when the temporary load command is in effect, the original power generation plan monitoring is suspended and power generation monitoring and deviation analysis are performed according to the temporary load command; B2 After the temporary load command ends, the deviation between the actual power generation and the original power generation plan in the corresponding time period is calculated; B3 Based on the deviation and the remaining time of the day, it is determined whether the deviation can be made up, and a subsequent power generation task correction plan and load adjustment suggestions for each unit are generated; B4 The load adjustment suggestions are automatically output to the unit control system to achieve execution optimization. S3 Optimization and Feedback: Based on historical data of deviations in power generation plan execution, the system uses predictive models to optimize future power generation plan execution monitoring strategies, thereby achieving dynamic deviation compensation and optimized load scheduling.

2. The method for monitoring deviations in power generation plan execution according to claim 1, characterized in that, In S1, the data integrity verification includes checking whether each data source is missing necessary fields and whether the data records are continuous; the timestamp consistency verification includes determining whether the timestamp of each data is synchronized with a unified clock source and whether the interval conforms to the preset sampling period; the numerical range verification includes checking whether the load value is within the rated capacity range of the generator set and whether it exceeds the grid dispatch limit.

3. The method for monitoring deviations in power generation plan execution according to claim 1, characterized in that, In S2, the dynamic instruction identification is performed in the following way: the message bus of the power grid dispatch automation system is monitored and the message type field is parsed; if the message type is a plan modification and includes the effective period of the modification and the modified plan curve, it is determined to be a power generation plan modification; if the message type is a real-time instruction or an automatic power generation control instruction and the absolute value of the difference between the output value at the time of instruction execution and the corresponding time value of the current effective plan curve exceeds the preset plan deviation tolerance, it is determined to be a temporary load instruction.

4. The method for monitoring deviations in power generation plan execution according to claim 1, characterized in that, In A3, when dynamically adjusting the subsequent power generation tasks of the original power generation plan based on the deviation power and the remaining time of the day, the deviation power is allocated to the basic planned value of each subsequent time period according to the proportion of the adjustment capacity of each unit, forming the corrected plan curve; the load adjustment suggestion includes the output value that each unit should increase or decrease in each subsequent time period and its adjustment rate requirement.

5. The method for monitoring deviations in power generation plan execution according to claim 1, characterized in that, In S2B, the method for determining whether the deviation can be compensated is as follows: calculate the total length of the remaining adjustable time period from the current time to the end of the day and the upper limit of the remaining adjustable capacity. If the sum of the maximum adjustable output increment of each unit in the remaining time period can cover the deviation power, it is determined to be compensateable and a compensation plan is generated; otherwise, it is determined to be uncompensable, an adjustment suggestion that minimizes the deviation is generated and marked as requiring scheduling intervention.

6. The method for monitoring deviations in power generation plan execution according to claim 1, characterized in that, In S3, the prediction model outputs the optimal deviation compensation coefficient and monitoring threshold adjustment suggestion for the next time period based on historical data.

7. The method for monitoring deviations in power generation plan execution according to claim 6, characterized in that, The prediction model collects actual execution deviation data daily and compares it with the prediction output. Based on the comparison results, it fine-tunes the model weights to achieve online adaptive updates of the model.

8. A power generation plan execution deviation monitoring system, characterized in that, The method for monitoring deviations in power generation plan execution as described in any one of claims 1-7 includes: The data acquisition and verification module is used to acquire power generation plan data, real-time load instructions from the power grid system, and actual operating load data of power plants, and to verify their validity. The dynamic command recognition module is used to determine whether the power grid system has made a formal modification to the current power generation plan or has issued a temporary load command that is inconsistent with the current power generation plan, and dynamically switches the monitoring mode. The plan modification processing module is used to perform real-time monitoring, deviation alarm, deviation power calculation, subsequent dynamic correction of the plan, and output of load adjustment suggestions during the modification period when it is determined that the power generation plan has been modified. The temporary instruction processing module is used to perform monitoring pause and deviation analysis during the execution of the temporary instruction, calculate the deviation power after the instruction ends, determine the recoverability, and output subsequent task correction and load adjustment suggestions when the instruction is determined to be a temporary load instruction. The optimization feedback module is used to optimize the monitoring strategy for future power generation plan execution based on historical data of deviations in power generation plan execution, thereby achieving dynamic deviation compensation and optimized load scheduling.

9. A power generation plan execution deviation monitoring system according to claim 8, characterized in that, The dynamic command recognition module makes judgments by monitoring the message bus of the power grid dispatch automation system or by parsing the command tag field in the data acquisition and monitoring control system; the system also includes an interface adaptation unit to be compatible with the data formats and communication protocols of different dispatch master station systems.