Multi-constraint-condition-based intelligent decision-making method and device for startup and shutdown of automatic power generation control system, computer equipment and storage medium

Abstract

The invention relates to an intelligent decision-making method and device for startup and shutdown of an automatic power generation control system based on multiple constraint conditions, computer equipment and a storage medium. The method comprises the following steps: determining a set value variable amplitude according to a current active capacity set value and a historical active capacity set value of the automatic generating capacity control system; under the condition that the current active capacity set value is within the active capacity set range and the variable amplitude of the set value is within the variable amplitude range of the set value, the number of units needing to be started and stopped by the automatic generating capacity control system is determined; under the condition that the number of the units is smaller than or equal to the threshold value of the number of the units, the number of the units is updated, and the updated number of the units is obtained; and generating a unit start-stop instruction of the target unit according to the updated number of the units and the unit priority of the current unit, and performing start-stop processing on the target unit according to the operation mode of the automatic generating capacity control system and the unit start-stop instruction. By adopting the method, the unit start-stop control accuracy can be improved.

Description

Technical Field

[0001] This application relates to the field of power grid technology, and in particular to a method, apparatus, computer equipment, computer-readable storage medium, and computer program product for intelligent decision-making on start-up and shutdown of an automatic power generation control system based on multiple constraints. Background Technology

[0002] In power systems, precise control of generator start-up and shutdown is crucial for maintaining a balance between power supply and demand.

[0003] In traditional technologies, the start-up and shutdown control of generating units is generally handled manually based on experience. However, this manual approach is difficult to handle the dynamic real-time nature of power systems, resulting in low accuracy of the start-up and shutdown control. Summary of the Invention

[0004] Therefore, it is necessary to provide a method, device, computer equipment, computer-readable storage medium, and computer program product for intelligent decision-making on start-up and shutdown of an automatic power generation control system based on multiple constraints, which can improve the accuracy of unit start-up and shutdown control and improve the accuracy of unit start-up and shutdown control.

[0005] In a first aspect, this application provides an intelligent decision-making method for start-up and shutdown of an automatic power generation control system based on multiple constraints, including:

[0006] In response to a unit start-up or shutdown request for the automatic generation control system, the current active power capacity setting value and the historical active power capacity setting value of the automatic generation control system are obtained.

[0007] Based on the current active power capacity setpoint and the historical active power capacity setpoint, the setpoint variation corresponding to the automatic power generation control system is determined;

[0008] When the current active capacity setpoint is within the active capacity setpoint range and the setpoint variation is within the setpoint variation range, the current setpoint interval information corresponding to the current active capacity setpoint is determined, and the current setpoint interval information is input into the trained unit number prediction model to obtain the number of units that the automatic power generation control system needs to start and stop.

[0009] If the number of generating units is less than or equal to the threshold number of generating units, the number of generating units is updated according to the current unit lockout status in the automatic power generation control system to obtain the updated number of generating units.

[0010] Obtain the unit priority of the current unit, and determine the target unit corresponding to the unit start / stop request from the current units based on the updated number of units and the unit priority;

[0011] Generate start-up and shutdown commands for the target unit, and perform corresponding start-up and shutdown processing on the target unit according to the operating mode of the automatic power generation control system and the start-up and shutdown commands.

[0012] In one embodiment, the step of inputting the current set value range information into the trained unit number prediction model to obtain the number of units that the automatic power generation control system needs to start and stop includes:

[0013] Based on the relative position encoding value corresponding to the current set value interval information, the first feature vector of the current set value interval information is obtained;

[0014] The current equipment status information of the automatic power generation control system is obtained, and the sub-equipment status information in the current equipment status information is processed by feature extraction to obtain the feature vector of the sub-equipment status information. The feature vectors of the sub-equipment status information are then concatenated to obtain the second feature vector of the current equipment status information.

[0015] The current environmental information and historical environmental information of the automatic power generation control system are obtained. Based on the current environmental information and the historical environmental information, the environmental change information of the automatic power generation control system is determined. The current environmental information is used as the main data, and the historical environmental information and the environmental change information are used as auxiliary data. The data are input into the feature extraction model to obtain the third feature vector of the current environmental information.

[0016] The first feature vector, the second feature vector, and the third feature vector are input into the trained unit number prediction model to obtain the number of units that the automatic power generation control system needs to start and stop.

[0017] In one embodiment, the trained unit count prediction model includes a first unit count prediction network, a second unit count prediction network, and a third unit count prediction network with different network structures.

[0018] The step of inputting the first feature vector, the second feature vector, and the third feature vector into the trained unit number prediction model to obtain the number of units that the automatic power generation control system needs to start and stop includes:

[0019] The first feature vector is input into the first unit number prediction network to obtain the first predicted number of units to be started and stopped by the automatic power generation control system, and the first predicted probability of the first predicted number of units.

[0020] The second feature vector is input into the second unit number prediction network to obtain the second predicted number of units to be started and stopped by the automatic power generation control system, and the second predicted probability of the second predicted number of units.

[0021] The third feature vector is input into the third unit number prediction network to obtain the third predicted unit number that the automatic power generation control system needs to start and stop, and the third predicted probability of the third predicted unit number.

[0022] The number of generating units that the automatic power generation control system needs to start and stop is determined based on the first predicted number of generating units, the second predicted number of generating units, the third predicted number of generating units, the first predicted probability, the second predicted probability, and the third predicted probability.

[0023] In one embodiment, determining the number of generating units that the automatic power generation control system needs to start and stop based on the first predicted number of generating units, the second predicted number of generating units, the third predicted number of generating units, the first prediction probability, the second prediction probability, and the third prediction probability includes:

[0024] Based on the first prediction probability, the second prediction probability, and the third prediction probability, a first weight corresponding to the number of the first prediction units, a second weight corresponding to the number of the second prediction units, and a third weight corresponding to the number of the third prediction units are determined.

[0025] According to the first weight, the second weight, and the third weight, the number of the first forecasting units, the number of the second forecasting units, and the number of the third forecasting units are fused to obtain the processed number of units.

[0026] Based on the number of generating units after processing, the number of generating units that the automatic power generation control system needs to start and stop is obtained.

[0027] In one embodiment, when the number of generating units is less than or equal to a threshold number of generating units, before updating the number of generating units according to the current unit lockout status in the automatic power generation control system to obtain the updated number of generating units, the method further includes:

[0028] Determine the current outgoing line operating status of the automatic power generation control system;

[0029] Based on the current outgoing line operation status, query the correspondence between the outgoing line operation status and the unit number threshold to obtain the first unit number threshold corresponding to the current outgoing line operation status;

[0030] The current outgoing line operation status is input into the trained unit number threshold prediction model to obtain the second unit number threshold corresponding to the current outgoing line operation status.

[0031] The first unit number threshold and the second unit number threshold are fused together to obtain the unit number threshold.

[0032] In one embodiment, the step of performing corresponding start-up and shutdown processing on the target unit according to the operating mode of the automatic power generation control system and the unit start-up and shutdown command includes:

[0033] When the operating mode is the guidance mode, the unit start-up and shutdown command is sent to the target terminal corresponding to the target unit, so that the target terminal performs start-up and shutdown processing on the target unit according to the unit start-up and shutdown command;

[0034] When the operating mode is automatic, the target unit is started and stopped according to the unit start-up and shutdown instructions.

[0035] Secondly, this application also provides an intelligent decision-making device for start-up and shutdown of an automatic power generation control system based on multiple constraints, comprising:

[0036] The data acquisition module is used to acquire the current active power capacity setpoint and historical active power capacity setpoint of the automatic power generation control system in response to the unit start-up and shutdown request of the automatic power generation control system.

[0037] The amplitude determination module is used to determine the amplitude of the setpoint corresponding to the automatic power generation control system based on the current active power capacity setpoint and the historical active power capacity setpoint.

[0038] The unit number prediction module is used to determine the current set value interval information corresponding to the current active capacity set value when the current active capacity set value is within the active capacity set range and the set value variation is within the set value variation range. The current set value interval information is then input into the trained unit number prediction model to obtain the number of units that the automatic power generation control system needs to start and stop.

[0039] The unit count update module is used to update the unit count according to the unit lockout status of the current unit in the automatic power generation control system when the number of units is less than or equal to the unit count threshold, so as to obtain the updated unit count.

[0040] The unit determination module is used to obtain the unit priority of the current unit, and determine the target unit corresponding to the unit start / stop request from the current unit based on the updated number of units and the unit priority;

[0041] The unit start-stop module is used to generate start-stop commands for the target unit and perform corresponding start-stop processing on the target unit according to the operating mode of the automatic power generation control system and the start-stop commands.

[0042] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0043] In response to a unit start-up or shutdown request for the automatic generation control system, the current active power capacity setting value and the historical active power capacity setting value of the automatic generation control system are obtained.

[0044] Based on the current active power capacity setpoint and the historical active power capacity setpoint, the setpoint variation corresponding to the automatic power generation control system is determined;

[0045] When the current active capacity setpoint is within the active capacity setpoint range and the setpoint variation is within the setpoint variation range, the current setpoint interval information corresponding to the current active capacity setpoint is determined, and the current setpoint interval information is input into the trained unit number prediction model to obtain the number of units that the automatic power generation control system needs to start and stop.

[0046] If the number of generating units is less than or equal to the threshold number of generating units, the number of generating units is updated according to the current unit lockout status in the automatic power generation control system to obtain the updated number of generating units.

[0047] Obtain the unit priority of the current unit, and determine the target unit corresponding to the unit start / stop request from the current units based on the updated number of units and the unit priority;

[0048] Generate start-up and shutdown commands for the target unit, and perform corresponding start-up and shutdown processing on the target unit according to the operating mode of the automatic power generation control system and the start-up and shutdown commands.

[0049] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0050] In response to a unit start-up or shutdown request for the automatic generation control system, the current active power capacity setting value and the historical active power capacity setting value of the automatic generation control system are obtained.

[0051] Based on the current active power capacity setpoint and the historical active power capacity setpoint, the setpoint variation corresponding to the automatic power generation control system is determined;

[0052] When the current active capacity setpoint is within the active capacity setpoint range and the setpoint variation is within the setpoint variation range, the current setpoint interval information corresponding to the current active capacity setpoint is determined, and the current setpoint interval information is input into the trained unit number prediction model to obtain the number of units that the automatic power generation control system needs to start and stop.

[0053] If the number of generating units is less than or equal to the threshold number of generating units, the number of generating units is updated according to the current unit lockout status in the automatic power generation control system to obtain the updated number of generating units.

[0054] Obtain the unit priority of the current unit, and determine the target unit corresponding to the unit start / stop request from the current units based on the updated number of units and the unit priority;

[0055] Generate start-up and shutdown commands for the target unit, and perform corresponding start-up and shutdown processing on the target unit according to the operating mode of the automatic power generation control system and the start-up and shutdown commands.

[0056] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:

[0057] In response to a unit start-up or shutdown request for the automatic generation control system, the current active power capacity setting value and the historical active power capacity setting value of the automatic generation control system are obtained.

[0058] Based on the current active power capacity setpoint and the historical active power capacity setpoint, the setpoint variation corresponding to the automatic power generation control system is determined;

[0059] When the current active capacity setpoint is within the active capacity setpoint range and the setpoint variation is within the setpoint variation range, the current setpoint interval information corresponding to the current active capacity setpoint is determined, and the current setpoint interval information is input into the trained unit number prediction model to obtain the number of units that the automatic power generation control system needs to start and stop.

[0060] If the number of generating units is less than or equal to the threshold number of generating units, the number of generating units is updated according to the current unit lockout status in the automatic power generation control system to obtain the updated number of generating units.

[0061] Obtain the unit priority of the current unit, and determine the target unit corresponding to the unit start / stop request from the current units based on the updated number of units and the unit priority;

[0062] Generate start-up and shutdown commands for the target unit, and perform corresponding start-up and shutdown processing on the target unit according to the operating mode of the automatic power generation control system and the start-up and shutdown commands.

[0063] The aforementioned intelligent decision-making method, device, computer equipment, storage medium, and computer program product for start-up and shutdown of an automatic generation control system based on multiple constraints first responds to unit start-up and shutdown requests for the automatic generation control system by acquiring the current active power capacity setpoint and historical active power capacity setpoint of the automatic generation control system. Then, based on the current and historical active power capacity setpoints, it determines the corresponding setpoint variation range for the automatic generation control system. Next, if the current active power capacity setpoint is within the active power capacity setting range and the setpoint variation range is within the setpoint variation range, it determines the current setpoint interval information corresponding to the current active power capacity setpoint and sets the current setpoint interval accordingly. The setpoint range information is input into the trained unit count prediction model to obtain the number of units required to start or stop the automatic power generation control system. Then, if the number of units is less than or equal to the unit count threshold, the unit count is updated according to the unit interlocking status of the current units in the automatic power generation control system to obtain the updated unit count. The unit priority of the current units is also obtained. Based on the updated unit count and unit priority, the target unit corresponding to the unit start / stop request is determined from the current units. Finally, the unit start / stop command of the target unit is generated, and the corresponding start / stop processing of the target unit is performed according to the operating mode of the automatic power generation control system and the unit start / stop command. In this way, when controlling the start-up and shutdown of generating units, multiple constraints, such as the current active power capacity setpoint and the requirement that the amplitude of the change be within a safe range, ensure compliance of the decision-making premise. The trained unit number prediction model replaces manual experience in calculating the required number of units to start and stop. The number of units is dynamically updated in combination with the unit lockout status to exclude unavailable units. Then, the target units are accurately selected based on unit priority. Finally, the instructions are executed according to the operating mode. This allows for precise start-up and shutdown processing of the target units, which helps improve the accuracy of unit start-up and shutdown control. Moreover, the entire process does not require manual intervention, avoiding the shortcomings of manual experience-based methods, which are unable to cope with the dynamic real-time nature of the power system and result in low accuracy of unit start-up and shutdown control. This further improves the accuracy of unit start-up and shutdown control. Attached Figure Description

[0064] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0065] Figure 1 This is a flowchart illustrating an intelligent decision-making method for start-up and shutdown of an automatic power generation control system based on multiple constraints in one embodiment.

[0066] Figure 2 This is a flowchart illustrating the steps for obtaining the number of generating units to be started and stopped by the automatic power generation control system in one embodiment.

[0067] Figure 3 This is a flowchart illustrating the intelligent decision-making method for start-up and shutdown of an automatic power generation control system based on multiple constraints in another embodiment.

[0068] Figure 4 This is a structural block diagram of an intelligent decision-making device for start-up and shutdown of an automatic power generation control system based on multiple constraints, as shown in one embodiment.

[0069] Figure 5 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0070] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0071] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0072] In one exemplary embodiment, such as Figure 1 As shown, an intelligent decision-making method for start-up and shutdown of an automatic power generation control system based on multiple constraints is provided. This embodiment illustrates the method by applying it to a server; it is understood that the method can also be applied to a terminal, or to a system including a terminal and a server, and is implemented through interaction between the terminal and the server. The terminal can be, but is not limited to, various personal computers, laptops, smartphones, and tablets; the server can be a standalone server or a server cluster composed of multiple servers. In this embodiment, the method includes the following steps:

[0073] Step S101: In response to a unit start-up / shutdown request for the automatic generation control system, obtain the current active power capacity setpoint and historical active power capacity setpoint of the automatic generation control system.

[0074] The automatic generation control system refers to the core control system used to automatically adjust the output of power plant units and control the start and stop of the units. It is also known as the AGC (Automatic Generation Control) system.

[0075] Among them, the unit start-up / shutdown request refers to the request that triggers the automatic power generation control system to perform unit start-up or shutdown operations.

[0076] The current active capacity setpoint, also known as the current active power setpoint, refers to the target active power value that the automatic power generation control system needs to execute at the present time.

[0077] Among them, the historical active capacity setpoint, also known as the historical active power setpoint, refers to the active power setpoint executed by the automatic power generation control system at a certain time in the past (such as 10 minutes ago or 1 hour ago).

[0078] For example, the server receives a generator start / stop request sent by the terminal for the automatic generation control system; then, in response to the generator start / stop request for the automatic generation control system, the server calculates the current active power capacity setpoint of the automatic generation control system using a PID (Proportional-Integral-Derivative) regulation algorithm or a load sharing model based on the grid frequency deviation of the automatic generation control system and the difference between the current total power output and the target total power output; then, the server retrieves the historical active power capacity setpoint of the automatic generation control system from the database based on the system identifier of the automatic generation control system.

[0079] Step S102: Determine the setpoint variation of the automatic power generation control system based on the current active power capacity setpoint and the historical active power capacity setpoint.

[0080] The setpoint variation is used to represent the difference between the current active capacity setpoint and the historical active capacity setpoint.

[0081] For example, the server uses the difference between the current active power capacity setpoint and the historical active power capacity setpoint as the setpoint variation of the automatic power generation control system.

[0082] Step S103: If the current active capacity setpoint is within the active capacity setpoint range and the setpoint variation is within the setpoint variation range, determine the current setpoint interval information corresponding to the current active capacity setpoint, input the current setpoint interval information into the trained unit number prediction model, and obtain the number of units that the automatic power generation control system needs to start and stop.

[0083] The active power capacity setting range refers to the upper and lower limits corresponding to the target active power preset by the automatic power generation control system.

[0084] The setpoint range refers to the maximum fluctuation range allowed by the setpoint range preset by the automatic power generation control system.

[0085] Among them, the current set value range information is used to indicate the range in which the current active power capacity set value is located (e.g., when generating electricity, 180 < capacity ≤ 480 corresponds to 1 unit).

[0086] Among them, the unit number prediction model refers to a network model that can use the information of the current set value range to obtain the number of units that the automatic power generation control system needs to start and stop, such as a convolutional neural network model.

[0087] The number of generating units refers to the total number of generating units that need to be started or stopped.

[0088] For example, the server determines the current active capacity setpoint and setpoint variation based on the active capacity set range and setpoint variation range, respectively. If the current active capacity setpoint is within the active capacity set range and the setpoint variation is within the setpoint variation range, the server determines the current setpoint interval information corresponding to the current active capacity setpoint according to the preset interval division rules. Then, the server performs feature extraction processing on the current setpoint interval information to obtain the feature vector of the current setpoint interval information, and inputs the feature vector of the current setpoint interval information into the trained unit number prediction model to obtain the number of units that the automatic power generation control system needs to start and stop.

[0089] Step S104: If the number of generating units is less than or equal to the threshold number of generating units, update the number of generating units according to the current unit lockout status in the automatic power generation control system to obtain the updated number of generating units.

[0090] The unit number threshold refers to the maximum number of units preset by the automatic power generation control system.

[0091] The term "current unit" refers to the set of all units within the monitoring range of the automatic power generation control system.

[0092] The unit lockout status indicates that the unit cannot perform start-up or shutdown operations, also known as the BLOCK status ("generator BLOCK" or "pump BLOCK").

[0093] The updated number of generating units refers to the number of generating units after the update.

[0094] For example, the server determines the number of generating units based on a unit number threshold. If the number of generating units is less than or equal to the unit number threshold, it queries the correspondence between the unit lockout status and the adjustment coefficient based on the current unit lockout status in the automatic power generation control system to obtain the adjustment coefficient corresponding to the number of generating units. Then, it adjusts the number of generating units according to the adjustment coefficient corresponding to the number of generating units to obtain the updated number of generating units.

[0095] Step S105: Obtain the unit priority of the current unit, and determine the target unit corresponding to the unit start / stop request from the current units based on the updated number of units and unit priority.

[0096] Among them, the unit priority is used to indicate the priority corresponding to the current unit's start-up and shutdown.

[0097] The target unit refers to the unit that will ultimately perform start-up and shutdown operations.

[0098] For example, the server obtains the unit efficiency information (e.g., prioritizing the start-up and shutdown of units with higher power generation efficiency and lower energy consumption), health status information (e.g., prioritizing the start-up and shutdown of units with good health to avoid frequent start-up and shutdown of units in critical states), operation and maintenance plan information (e.g., lowering the priority of units that need to be inspected in the near future to avoid shutdown immediately after startup), and response speed information (e.g., pumped storage units have a fast start-up speed and higher priority in peak shaving scenarios). The server then inputs the unit efficiency information, health status information, operation and maintenance plan information, and response speed information of the current unit into the trained unit priority prediction model to obtain the unit priority of the current unit. Next, the server determines the unit corresponding to the updated number of units and unit priority from the current units as the target unit for the unit start-up and shutdown request.

[0099] Step S106: Generate the start-up and shutdown command for the target unit, and perform corresponding start-up and shutdown processing on the target unit according to the operating mode of the automatic power generation control system and the start-up and shutdown command.

[0100] Among them, the unit start-up and shutdown command refers to the command to start up or shut down the target unit.

[0101] The operating modes include guided mode and automatic mode (automatic mode executes directly, while guided mode executes after manual confirmation).

[0102] For example, the server obtains the target unit's start-up and shutdown command by encapsulating the target unit's unit identifier and operating parameters into instructions; then, the server performs corresponding start-up and shutdown processing on the target unit according to the operating mode of the automatic power generation control system and the unit start-up and shutdown command.

[0103] In the aforementioned intelligent decision-making method for start-up and shutdown of an automatic generation control system based on multiple constraints, the method first responds to unit start-up and shutdown requests to the automatic generation control system by acquiring the current active power capacity setpoint and historical active power capacity setpoint. Then, based on the current and historical active power capacity setpoints, the corresponding setpoint variation for the automatic generation control system is determined. Next, if the current active power capacity setpoint is within the active power capacity setting range and the setpoint variation is within the setpoint variation range, the current setpoint interval information corresponding to the current active power capacity setpoint is determined, and this current setpoint interval information is input to... The trained unit count prediction model obtains the number of units required to be started or stopped by the automatic generation control system. Then, when the number of units is less than or equal to the unit count threshold, the unit count is updated according to the unit interlocking status of the current units in the automatic generation control system to obtain the updated unit count. The unit priority of the current units is also obtained. Based on the updated unit count and unit priority, the target unit corresponding to the unit start / stop request is determined from the current units. Finally, the unit start / stop command of the target unit is generated, and the corresponding start / stop processing of the target unit is performed according to the operating mode of the automatic generation control system and the unit start / stop command. In this way, when controlling the start-up and shutdown of generating units, multiple constraints, such as the current active power capacity setpoint and the requirement that the amplitude of the change be within a safe range, ensure compliance of the decision-making premise. The trained unit number prediction model replaces manual experience in calculating the required number of units to start and stop. The number of units is dynamically updated in combination with the unit lockout status to exclude unavailable units. Then, the target units are accurately selected based on unit priority. Finally, the instructions are executed according to the operating mode. This allows for precise start-up and shutdown processing of the target units, which helps improve the accuracy of unit start-up and shutdown control. Moreover, the entire process does not require manual intervention, avoiding the shortcomings of manual experience-based methods, which are unable to cope with the dynamic real-time nature of the power system and result in low accuracy of unit start-up and shutdown control. This further improves the accuracy of unit start-up and shutdown control.

[0104] In one exemplary embodiment, such as Figure 2 As shown, the current setpoint range information is input into the trained unit number prediction model to obtain the number of units that the automatic power generation control system needs to start and stop. The specific steps include the following:

[0105] Step S201: Based on the relative position encoding value corresponding to the current set value interval information, obtain the first feature vector of the current set value interval information.

[0106] Step S202: Obtain the current equipment status information of the automatic power generation control system, perform feature extraction processing on the sub-equipment status information in the current equipment status information to obtain the feature vector of the sub-equipment status information, and concatenate the feature vectors of the sub-equipment status information to obtain the second feature vector of the current equipment status information.

[0107] Step S203: Obtain the current environmental information and historical environmental information of the automatic power generation control system. Based on the current environmental information and historical environmental information, determine the environmental change information of the automatic power generation control system. Use the current environmental information as the main data and the historical environmental information and environmental change information as auxiliary data, and input them into the feature extraction model to obtain the third feature vector of the current environmental information.

[0108] Step S204: Input the first feature vector, the second feature vector, and the third feature vector into the trained unit number prediction model to obtain the number of units that the automatic power generation control system needs to start and stop.

[0109] The relative position code value is used to represent the code value corresponding to the relative position of the current active capacity setting value within its interval.

[0110] The first feature vector refers to the feature vector corresponding to the information in the current set value range.

[0111] The current equipment status information is used to represent all data related to the operating status of the generator unit and associated equipment in the automatic power generation control system. This includes data such as the generator unit's vibration level, temperature, speed, output feedback, and the voltage, current, and transmission power of the outgoing lines.

[0112] Among them, sub-equipment status information refers to the detailed status data of the current equipment status information according to the functional units of the unit.

[0113] The second feature vector refers to the feature vector corresponding to the current device status information.

[0114] The current environmental information refers to the external environmental information of the automatic power generation control system at the current time. For example, the head and flow rate of a hydropower station, the calorific value of fuel in a thermal power plant, and data such as temperature and humidity in the area where the power station is located.

[0115] Historical environmental information refers to the external environmental information of the automatic power generation control system over a period of time.

[0116] Among them, environmental change information is used to represent the difference between current environmental information and historical environmental information. For example, if the current water head is 80m and the historical average water head for the same period is 85m, the environmental change information would be "water head decreased by 5m".

[0117] Among them, feature extraction models refer to network models that can extract feature vectors, such as the transformer model.

[0118] The third feature vector refers to the feature vector corresponding to the current environmental information.

[0119] For example, the server performs relative position encoding processing on the relative position of the current active capacity setpoint within the active capacity setpoint range according to the relative position encoding model, and obtains the relative position encoding value corresponding to the current setpoint interval information; then, the server uses the relative position encoding value corresponding to the current setpoint interval information as the first feature vector of the current setpoint interval information; then, the server collects the current equipment status information of the automatic power generation control system through the status sensor associated with the automatic power generation control system, performs feature extraction processing on the sub-equipment status information in the current equipment status information, obtains the feature vector of the sub-equipment status information, and splices the feature vector of the sub-equipment status information according to the preset splicing order to obtain the second feature vector of the current equipment status information; Next, the server collects the current environmental information of the automatic power generation control system through environmental sensors associated with the automatic power generation control system, and retrieves the historical environmental information of the automatic power generation control system from the database. The difference between the current environmental information and the historical environmental information is used as the environmental change information of the automatic power generation control system. The current environmental information is used as the primary data, and the historical environmental information and environmental change information are used as auxiliary data. This data is then input into the feature extraction model to obtain the third feature vector of the current environmental information. Then, the server fuses the first, second, and third feature vectors to obtain a fused feature vector. This fused feature vector is then input into the trained unit number prediction model to obtain the number of units that the automatic power generation control system needs to start and stop.

[0120] In this embodiment, by extracting the relative position features of the current set value range, the comprehensive features of the equipment status, and the dynamic features of environmental factors, a comprehensive coverage of the key influencing dimensions of unit start-up and shutdown decisions is achieved. At the same time, the multi-dimensional features are integrated and input into the unit number prediction model, avoiding decision-making bias caused by one-dimensional information and improving the accuracy of unit number prediction.

[0121] In one exemplary embodiment, the trained unit count prediction model includes a first unit count prediction network, a second unit count prediction network, and a third unit count prediction network with different network structures.

[0122] Then, step S204 above, which inputs the first feature vector, the second feature vector, and the third feature vector into the trained unit number prediction model to obtain the number of units that the automatic power generation control system needs to start and stop, specifically includes the following: inputting the first feature vector into the first unit number prediction network to obtain the first predicted number of units that the automatic power generation control system needs to start and stop, and the first predicted probability of the first predicted number of units; inputting the second feature vector into the second unit number prediction network to obtain the second predicted number of units that the automatic power generation control system needs to start and stop, and the second predicted probability of the second predicted number of units; inputting the third feature vector into the third unit number prediction network to obtain the third predicted number of units that the automatic power generation control system needs to start and stop, and the third predicted probability of the third predicted number of units; and determining the number of units that the automatic power generation control system needs to start and stop based on the first predicted number of units, the second predicted number of units, the third predicted number of units, the first predicted probability, the second predicted probability, and the third predicted probability.

[0123] Among them, the first unit number prediction network can refer to a lightweight architecture network adapted to one-dimensional numerical features, such as a 1D-CNN (1-Dimensional Convolutional Neural Network) network.

[0124] The second unit number prediction network can refer to a hybrid architecture network, such as a CNN (Convolutional Neural Network) - fully connected fusion network.

[0125] Among them, the third unit number prediction network can refer to a CNN network adapted to time series data, such as a 2D-CNN (2-Dimensional Convolutional Neural Network) network.

[0126] The first forecast unit number refers to the number of forecast units that the automatic power generation control system output by the first unit number forecast network needs to start and stop.

[0127] The first prediction probability refers to the prediction probability corresponding to the first predicted number of generating units output by the first generating unit number prediction network.

[0128] The second forecast unit number refers to the number of forecast units that the automatic power generation control system output by the second unit number forecast network needs to start and stop.

[0129] The second prediction probability refers to the prediction probability corresponding to the second prediction unit number output by the second unit number prediction network.

[0130] Among them, the number of third forecasting units refers to the number of forecasting units that the automatic power generation control system output by the third unit forecasting network needs to start and stop.

[0131] The third prediction probability refers to the prediction probability corresponding to the number of third prediction units output by the third prediction unit number prediction network.

[0132] For example, the server inputs a first feature vector into a first unit number prediction network, and uses the first unit number prediction network to predict the first predicted number of units that the automatic power generation control system needs to start or stop, as well as a first predicted probability of the first predicted number of units. The server then inputs a second feature vector into a second unit number prediction network, and uses the second unit number prediction network to predict the second predicted number of units that the automatic power generation control system needs to start or stop, as well as a second predicted probability of the second predicted number of units. Finally, the server inputs a third feature vector into a third unit number prediction network, and uses the third unit number prediction network to predict the third predicted number of units that the automatic power generation control system needs to start or stop, as well as a third predicted probability of the third predicted number of units. Then, the server determines the number of units that the automatic power generation control system needs to start or stop based on the first predicted number of units, the second predicted number of units, the third predicted number of units, the first predicted probability, the second predicted probability, and the third predicted probability.

[0133] In this embodiment, by allowing three branch networks to independently predict the number of generating units and output confidence scores, the adaptation limitations of a single network when processing multi-source features are avoided. Furthermore, the reliability of each prediction result can be quantified by relying on the probability value of each prediction result, effectively reducing the impact of single-dimensional bias on the final result, which is conducive to improving the accuracy and robustness of generating unit prediction.

[0134] In an exemplary embodiment, the number of generating units that the automatic power generation control system needs to start and stop is determined based on the first predicted number of generating units, the second predicted number of generating units, the third predicted number of generating units, the first predicted probability, the second predicted probability, and the third predicted probability. Specifically, this includes: determining a first weight corresponding to the first predicted number of generating units, a second weight corresponding to the second predicted number of generating units, and a third weight corresponding to the third predicted number of generating units based on the first predicted probability, the second predicted probability, and the third predicted probability; performing a fusion process on the first predicted number of generating units, the second predicted number of generating units, and the third predicted number of generating units according to the first weight, the second weight, and the third weight to obtain a processed number of generating units; and obtaining the number of generating units that the automatic power generation control system needs to start and stop based on the processed number of generating units.

[0135] The first weight refers to the weight coefficient corresponding to the number of the first forecasting units.

[0136] The second weight refers to the weight coefficient corresponding to the number of the second forecasting units.

[0137] The third weight refers to the weight coefficient corresponding to the number of the third forecasting units.

[0138] The number of units after processing refers to the number of units obtained by merging the number of units predicted in the first forecast, the number of units predicted in the second forecast, and the number of units predicted in the third forecast.

[0139] For example, the server uses the first prediction probability as the first initial weight corresponding to the number of first prediction units, the second prediction probability as the second initial weight corresponding to the number of second prediction units, and the third prediction probability as the third initial weight corresponding to the number of third prediction units. The server then normalizes the first, second, and third initial weights to obtain the first weight corresponding to the number of first prediction units, the second weight corresponding to the number of second prediction units, and the third weight corresponding to the number of third prediction units. Next, the server applies the first, second, and third weights to the number of first prediction units, the number of second prediction units, and the number of third prediction units. The number of generating units and the number of units predicted by the third forecast are summed to obtain the processed number of generating units. Then, the server determines the rounding method corresponding to the processed number of generating units based on the current health status information of the generating units. For example, if the standby generating unit to be started has a critical state such as excessive vibration, temperature close to the threshold, or the outgoing current has reached 90% of the transmission capacity (restarting the unit may overload), it is necessary to abandon rounding up and instead round down to ensure that the equipment does not exceed the limit. Then, the server rounds the processed number of generating units according to the rounding method corresponding to the processed number of generating units to obtain the number of generating units that the automatic power generation control system needs to start and stop.

[0140] In this embodiment, by converting the prediction probabilities of each branch prediction result into corresponding weights, dynamic weighted fusion of multi-dimensional prediction results is achieved, avoiding low-reliability prediction from interfering with decision-making. This not only overcomes the one-sidedness of single-dimensional prediction, but also reflects the actual influence of different factors through weight allocation, ensuring that the output results meet the power regulation requirements.

[0141] In an exemplary embodiment, before updating the number of generating units according to the unit interlocking status of the current generating units in the automatic power generation control system when the number of generating units is less than or equal to the unit number threshold, step S104 specifically includes the following: determining the current outgoing line operation status of the automatic power generation control system; querying the correspondence between the outgoing line operation status and the unit number threshold based on the current outgoing line operation status to obtain the first unit number threshold corresponding to the current outgoing line operation status; inputting the current outgoing line operation status into the trained unit number threshold prediction model to obtain the second unit number threshold corresponding to the current outgoing line operation status; and fusing the first unit number threshold and the second unit number threshold to obtain the unit number threshold.

[0142] The current outgoing line operating status can refer to the outgoing line operating condition of the automatic power generation control system at the current time (such as line maintenance).

[0143] Among them, the first unit number threshold refers to the unit number threshold corresponding to the current outgoing line operation status obtained based on the correspondence relationship.

[0144] Among them, the unit number threshold prediction model refers to a network model that can obtain the unit number threshold corresponding to the current outgoing line operation status using the current outgoing line operation status, such as a gradient boosting tree.

[0145] Among them, the second unit number threshold refers to the unit number threshold corresponding to the current outgoing line operation status output by the unit number threshold prediction model.

[0146] For example, the server acquires electrical parameter data (such as real-time current, voltage, active power, reactive power, power factor, etc.), physical state data (such as outgoing conductor temperature, insulator insulation resistance, etc.), and fault data (such as whether overcurrent or overvoltage has been triggered) of the automatic power generation control system. Based on the electrical parameter data, physical state data, and fault data, the server determines the current outgoing line operating state of the automatic power generation control system. Next, based on the current outgoing line operating state, the server queries the correspondence between the outgoing line operating state and the unit number threshold to obtain the unit number threshold corresponding to the current outgoing line operating state, which is used as the first unit number threshold. Then, the server performs feature extraction processing on the current outgoing line operating state to obtain the feature vector of the current outgoing line operating state, and inputs the feature vector of the current outgoing line operating state into the trained unit number threshold prediction model to obtain the second unit number threshold corresponding to the current outgoing line operating state. Finally, the server sums the first unit number threshold and the second unit number threshold according to their respective weights to obtain the unit number threshold.

[0147] In this embodiment, the method of obtaining the unit number threshold through a dual path of preset rule query and model dynamic prediction can ensure that the output unit number threshold not only meets the standardized safety constraints, but also fits the dynamic requirements of the real-time operation scenario. This effectively avoids the risks caused by misjudgment of a single rule or extreme values ​​of a single model, and helps to improve the accuracy and safety of the outgoing line carrying unit number constraint.

[0148] In an exemplary embodiment, step S106 above, which involves performing corresponding start-up and shutdown processing on the target unit according to the operating mode of the automatic power generation control system and the unit start-up and shutdown command, specifically includes the following: when the operating mode is the guidance mode, the unit start-up and shutdown command is sent to the target terminal corresponding to the target unit, so that the target terminal performs start-up and shutdown processing on the target unit according to the unit start-up and shutdown command; when the operating mode is the automatic mode, the target unit performs start-up and shutdown processing on the target unit according to the unit start-up and shutdown command.

[0149] The guidance mode refers to the operating mode that requires manual or on-site terminal execution of instructions.

[0150] Among them, the target terminal refers to the field control terminal equipment (such as the unit's local control cabinet, handheld operating terminal for maintenance personnel, workshop monitoring terminal, etc.) that is directly associated with the target unit.

[0151] Automatic mode refers to an operating mode that does not require manual intervention and directly controls the start and stop of the target unit according to the start and stop commands.

[0152] For example, the server determines the operating mode; when the operating mode is guided mode, the server sends the unit start / stop command to the target terminal corresponding to the target unit through the network path between the server and the target terminal, so that the target terminal performs the start / stop process of the target unit according to the unit start / stop command; when the operating mode is automatic mode, the server directly performs the start / stop process of the target unit according to the unit start / stop command.

[0153] In this embodiment, by distinguishing between the guidance mode and the automatic mode, flexible adaptation of unit start-up and shutdown control can be achieved. This not only meets the differentiated needs of different operating scenarios for safety confirmation and control efficiency, but also ensures consistent control standards through unified instruction logic, ultimately achieving the optimal balance between reliability and efficiency in unit start-up and shutdown operations.

[0154] In one exemplary embodiment, such as Figure 3 As shown, another intelligent decision-making method for start-up and shutdown of an automatic power generation control system based on multiple constraints is provided. Taking the application of this method to a server as an example, the specific steps include:

[0155] Step S301: In response to the unit start-up / shutdown request for the automatic generation control system, obtain the current active power capacity setpoint and historical active power capacity setpoint of the automatic generation control system.

[0156] Step S302: Determine the setpoint variation of the automatic power generation control system based on the current active power capacity setpoint and the historical active power capacity setpoint.

[0157] Step S303: When the current active capacity setting value is within the active capacity setting range and the setting value variation is within the setting value variation range, the first feature vector of the current setting value interval information is obtained based on the relative position encoding value corresponding to the current setting value interval information.

[0158] Step S304: Obtain the current equipment status information of the automatic power generation control system, perform feature extraction processing on the sub-equipment status information in the current equipment status information to obtain the feature vector of the sub-equipment status information, and concatenate the feature vectors of the sub-equipment status information to obtain the second feature vector of the current equipment status information.

[0159] Step S305: Obtain the current environmental information and historical environmental information of the automatic power generation control system. Based on the current environmental information and historical environmental information, determine the environmental change information of the automatic power generation control system. Use the current environmental information as the main data and the historical environmental information and environmental change information as auxiliary data, and input them into the feature extraction model to obtain the third feature vector of the current environmental information.

[0160] Step S306: Input the first feature vector, the second feature vector, and the third feature vector into the trained unit number prediction model to obtain the number of units that the automatic power generation control system needs to start and stop.

[0161] Step S307: Determine the current outgoing line operating status of the automatic power generation control system.

[0162] Step S308: Based on the current outgoing line operation status, query the correspondence between the outgoing line operation status and the unit number threshold to obtain the first unit number threshold corresponding to the current outgoing line operation status.

[0163] Step S309: Input the current outgoing line operation status into the trained unit number threshold prediction model to obtain the second unit number threshold corresponding to the current outgoing line operation status.

[0164] Step S310: The first unit number threshold and the second unit number threshold are fused to obtain the unit number threshold.

[0165] Step S311: If the number of generating units is less than or equal to the threshold number of generating units, update the number of generating units according to the current unit lockout status in the automatic power generation control system to obtain the updated number of generating units.

[0166] Step S312: Obtain the unit priority of the current unit, and determine the target unit corresponding to the unit start / stop request from the current units based on the updated number of units and unit priority.

[0167] Step S313: Generate the start-up and shutdown command for the target unit, and perform corresponding start-up and shutdown processing on the target unit according to the operating mode of the automatic power generation control system and the start-up and shutdown command.

[0168] In the aforementioned intelligent decision-making method for start-up and shutdown of automatic power generation control systems based on multiple constraints, the decision premise is ensured to be compliant during unit start-up and shutdown control by verifying multiple constraints such as the current active power capacity setpoint and the amplitude of the change being within a safe range. A trained unit count prediction model is used to replace manual experience in calculating the required number of units to start or stop. The number of units is dynamically updated based on the unit lockout status to exclude unavailable units. Target units are then accurately selected based on unit priority. Finally, instructions are executed according to the appropriate operating mode. This allows for precise start-up and shutdown processing of target units, improving the accuracy of unit start-up and shutdown control. Furthermore, the entire process requires no manual intervention, avoiding the shortcomings of manual experience-based methods that struggle to cope with the dynamic real-time nature of the power system, leading to lower accuracy in unit start-up and shutdown control. This further improves the accuracy of unit start-up and shutdown control.

[0169] In one exemplary embodiment, to more clearly illustrate the intelligent start-up and shutdown decision-making method for an automatic power generation control system based on multiple constraints provided in this application, the following specific embodiment will be used to describe this method in detail. In one embodiment, this application also provides yet another intelligent start-up and shutdown decision-making method for an automatic power generation control system based on multiple constraints. Specifically, it includes the following:

[0170] 1. Receive start-up and shutdown instructions: Obtain the plant's total start-up capacity setting value issued by the power station / dispatch center, and determine the direction (power generation / pumping).

[0171] 2. Verify capacity validity: Check whether the set value is within the "allowable active capacity setting range" and whether the fluctuation does not exceed the capacity adjustment step size. If it exceeds the range, an alarm will be triggered and execution will be refused.

[0172] 3. Calculate the theoretical number of units to start / stop: Based on the range of the set value (e.g., 1 unit corresponds to 180 < capacity ≤ 480 during power generation), determine the theoretical number of units that need to be started / stopped.

[0173] 4. Constraint filtering:

[0174] Considering outgoing line capacity limitations: Calculate the maximum number of units currently allowed to be started based on the outgoing line operating conditions (such as line maintenance). If the theoretical number of units is exceeded, the system will be locked and an alarm will be triggered.

[0175] Check the unit's BLOCK status: If the unit's "Power Generation BLOCK" or "Pumping BLOCK" is engaged, deduct its corresponding capacity and recalculate the actual number of units that can be operated.

[0176] 5. Determine the target unit: Based on the preset unit priority (e.g., 4321 represents that unit 4 has priority to start), sort the units that meet the conditions by priority and select the unit that needs to be started / stopped.

[0177] 6. Execute start-up and shutdown control: Based on the "Plant-wide AGC start-up and shutdown function" mode (guided / automatic), issue start-up and shutdown commands (automatic mode executes directly, guided mode executes after manual confirmation).

[0178] In the above embodiments, when performing unit start-up and shutdown control, multiple constraints, such as the current active power capacity setpoint and the requirement that the amplitude of the change be within a safe range, are used to ensure compliance of the decision-making premise. A trained unit number prediction model is used to replace manual experience in calculating the required number of units to start and stop. The number of units is dynamically updated in combination with the unit lockout status to exclude unavailable units. Then, the target units are accurately selected according to the unit priority. Finally, the instructions are executed according to the operating mode. This allows for precise start-up and shutdown processing of the target units, which helps improve the accuracy of unit start-up and shutdown control. Moreover, the entire process does not require manual intervention, avoiding the shortcomings of manual experience-based processing, which is difficult to cope with the dynamic real-time nature of the power system and leads to low accuracy of unit start-up and shutdown control. This further improves the accuracy of unit start-up and shutdown control.

[0179] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0180] Based on the same inventive concept, this application also provides a device for implementing the above-described intelligent decision-making method for starting and stopping an automatic power generation control system based on multiple constraints. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations of one or more embodiments of the intelligent decision-making device for starting and stopping an automatic power generation control system based on multiple constraints provided below can be found in the limitations of the intelligent decision-making method for starting and stopping an automatic power generation control system based on multiple constraints described above, and will not be repeated here.

[0181] In one exemplary embodiment, such as Figure 4As shown, an intelligent decision-making device for start-up and shutdown of an automatic power generation control system based on multiple constraints is provided, including: a data acquisition module 401, a variable amplitude determination module 402, a unit count prediction module 403, a unit count update module 404, a unit determination module 405, and a unit start-up and shutdown module 406, wherein:

[0182] The data acquisition module 401 is used to acquire the current active power capacity setpoint and historical active power capacity setpoint of the automatic power generation control system in response to the unit start-up and shutdown request for the automatic power generation control system.

[0183] The amplitude determination module 402 is used to determine the amplitude of the setpoint corresponding to the automatic power generation control system based on the current active power capacity setpoint and the historical active power capacity setpoint.

[0184] The unit count prediction module 403 is used to determine the current setpoint interval information corresponding to the current active capacity setpoint when the current active capacity setpoint is within the active capacity setpoint range and the setpoint variation is within the setpoint variation range. The current setpoint interval information is then input into the trained unit count prediction model to obtain the number of units that the automatic power generation control system needs to start and stop.

[0185] The unit count update module 404 is used to update the unit count based on the unit lockout status of the current unit in the automatic power generation control system when the unit count is less than or equal to the unit count threshold, so as to obtain the updated unit count.

[0186] The unit determination module 405 is used to obtain the unit priority of the current unit and determine the target unit corresponding to the unit start / stop request from the current units based on the updated number of units and unit priority.

[0187] The unit start-stop module 406 is used to generate start-stop commands for the target unit and perform corresponding start-stop processing on the target unit according to the operating mode of the automatic power generation control system and the start-stop commands.

[0188] In an exemplary embodiment, the unit count prediction module 403 is further configured to: obtain a first feature vector of the current set value interval information based on the relative position encoding value corresponding to the current set value interval information; acquire the current equipment status information of the automatic power generation control system; perform feature extraction processing on the sub-equipment status information in the current equipment status information to obtain the feature vector of the sub-equipment status information; and concatenate the feature vectors of the sub-equipment status information to obtain a second feature vector of the current equipment status information; acquire the current environmental information and historical environmental information of the automatic power generation control system; determine the environmental change information of the automatic power generation control system based on the current environmental information and historical environmental information; and input the current environmental information as the main data and the historical environmental information and environmental change information as auxiliary data into the feature extraction model to obtain a third feature vector of the current environmental information; and input the first feature vector, the second feature vector, and the third feature vector into the trained unit count prediction model to obtain the number of units that the automatic power generation control system needs to start and stop.

[0189] In an exemplary embodiment, the unit count prediction module 403 is further configured to input a first feature vector into a first unit count prediction network to obtain a first predicted number of units to be started or stopped by the automatic power generation control system, and a first predicted probability of the first predicted number of units; input a second feature vector into a second unit count prediction network to obtain a second predicted number of units to be started or stopped by the automatic power generation control system, and a second predicted probability of the second predicted number of units; input a third feature vector into a third unit count prediction network to obtain a third predicted number of units to be started or stopped by the automatic power generation control system, and a third predicted probability of the third predicted number of units; and determine the number of units to be started or stopped by the automatic power generation control system based on the first predicted number of units, the second predicted number of units, the third predicted number of units, the first predicted probability, the second predicted probability, and the third predicted probability.

[0190] In an exemplary embodiment, the unit number prediction module 403 is further configured to determine, based on the first prediction probability, the second prediction probability, and the third prediction probability, a first weight corresponding to the first predicted number of generating units, a second weight corresponding to the second predicted number of generating units, and a third weight corresponding to the third predicted number of generating units; perform fusion processing on the first predicted number of generating units, the second predicted number of generating units, and the third predicted number of generating units according to the first weight, the second weight, and the third weight to obtain the processed number of generating units; and obtain the number of generating units that the automatic power generation control system needs to start and stop based on the processed number of generating units.

[0191] In an exemplary embodiment, the intelligent decision-making device for start-up and shutdown of the automatic power generation control system based on multiple constraints further includes a threshold determination module, used to determine the current outgoing line operation status of the automatic power generation control system; based on the current outgoing line operation status, query the correspondence between the outgoing line operation status and the unit number threshold to obtain the first unit number threshold corresponding to the current outgoing line operation status; input the current outgoing line operation status into the trained unit number threshold prediction model to obtain the second unit number threshold corresponding to the current outgoing line operation status; and fuse the first unit number threshold and the second unit number threshold to obtain the unit number threshold.

[0192] In an exemplary embodiment, the unit start-stop module 406 is further configured to send a unit start-stop command to the target terminal corresponding to the target unit when the operating mode is guided mode, so that the target terminal performs start-stop processing on the target unit according to the unit start-stop command; and to perform start-stop processing on the target unit according to the unit start-stop command when the operating mode is automatic mode.

[0193] The modules in the aforementioned intelligent decision-making device for start-up and shutdown of the automatic power generation control system based on multiple constraints can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0194] In one exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 5 As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs stored in the non-volatile storage media. The database stores data such as current active power capacity setpoints and historical active power capacity setpoints. The I / O interfaces are used for information exchange between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements an intelligent decision-making method for start-up and shutdown of an automatic power generation control system based on multiple constraints.

[0195] Those skilled in the art will understand that Figure 5 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0196] In one exemplary embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0197] In one exemplary embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above-described method embodiments.

[0198] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0199] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0200] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0201] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for intelligent decision-making on start-up and shutdown of an automatic power generation control system based on multiple constraints, characterized in that, The method includes: In response to a unit start-up or shutdown request for the automatic generation control system, the current active power capacity setting value and the historical active power capacity setting value of the automatic generation control system are obtained. Based on the current active power capacity setpoint and the historical active power capacity setpoint, the setpoint variation corresponding to the automatic power generation control system is determined; When the current active capacity setpoint is within the active capacity setpoint range and the setpoint variation is within the setpoint variation range, the current setpoint interval information corresponding to the current active capacity setpoint is determined, and the current setpoint interval information is input into the trained unit number prediction model to obtain the number of units that the automatic power generation control system needs to start and stop. If the number of generating units is less than or equal to the threshold number of generating units, the number of generating units is updated according to the current unit lockout status in the automatic power generation control system to obtain the updated number of generating units. Obtain the unit priority of the current unit, and determine the target unit corresponding to the unit start / stop request from the current units based on the updated number of units and the unit priority; Generate start-up and shutdown commands for the target unit, and perform corresponding start-up and shutdown processing on the target unit according to the operating mode of the automatic power generation control system and the start-up and shutdown commands.

2. The method according to claim 1, characterized in that, The step of inputting the current set value range information into the trained unit number prediction model to obtain the number of units that the automatic power generation control system needs to start and stop includes: Based on the relative position encoding value corresponding to the current set value interval information, the first feature vector of the current set value interval information is obtained; The current equipment status information of the automatic power generation control system is obtained, and the sub-equipment status information in the current equipment status information is processed by feature extraction to obtain the feature vector of the sub-equipment status information. The feature vectors of the sub-equipment status information are then concatenated to obtain the second feature vector of the current equipment status information. The current environmental information and historical environmental information of the automatic power generation control system are obtained. Based on the current environmental information and the historical environmental information, the environmental change information of the automatic power generation control system is determined. The current environmental information is used as the main data, and the historical environmental information and the environmental change information are used as auxiliary data. The data are input into the feature extraction model to obtain the third feature vector of the current environmental information. The first feature vector, the second feature vector, and the third feature vector are input into the trained unit number prediction model to obtain the number of units that the automatic power generation control system needs to start and stop.

3. The method according to claim 2, characterized in that, The trained unit count prediction model includes a first unit count prediction network, a second unit count prediction network, and a third unit count prediction network with different network structures. The step of inputting the first feature vector, the second feature vector, and the third feature vector into the trained unit number prediction model to obtain the number of units that the automatic power generation control system needs to start and stop includes: The first feature vector is input into the first unit number prediction network to obtain the first predicted number of units to be started and stopped by the automatic power generation control system, and the first predicted probability of the first predicted number of units. The second feature vector is input into the second unit number prediction network to obtain the second predicted number of units to be started and stopped by the automatic power generation control system, and the second predicted probability of the second predicted number of units. The third feature vector is input into the third unit number prediction network to obtain the third predicted unit number that the automatic power generation control system needs to start and stop, and the third predicted probability of the third predicted unit number. The number of generating units that the automatic power generation control system needs to start and stop is determined based on the first predicted number of generating units, the second predicted number of generating units, the third predicted number of generating units, the first predicted probability, the second predicted probability, and the third predicted probability.

4. The method according to claim 3, characterized in that, The step of determining the number of generating units that the automatic power generation control system needs to start and stop based on the first predicted number of generating units, the second predicted number of generating units, the third predicted number of generating units, the first prediction probability, the second prediction probability, and the third prediction probability includes: Based on the first prediction probability, the second prediction probability, and the third prediction probability, a first weight corresponding to the number of the first prediction units, a second weight corresponding to the number of the second prediction units, and a third weight corresponding to the number of the third prediction units are determined. According to the first weight, the second weight, and the third weight, the number of the first forecasting units, the number of the second forecasting units, and the number of the third forecasting units are fused to obtain the processed number of units. Based on the number of generating units after processing, the number of generating units that the automatic power generation control system needs to start and stop is obtained.

5. The method according to claim 1, characterized in that, If the number of generating units is less than or equal to a threshold number of generating units, the number of generating units is updated according to the current unit interlocking status in the automatic power generation control system. Before obtaining the updated number of generating units, the process further includes: Determine the current outgoing line operating status of the automatic power generation control system; Based on the current outgoing line operation status, query the correspondence between the outgoing line operation status and the unit number threshold to obtain the first unit number threshold corresponding to the current outgoing line operation status; The current outgoing line operation status is input into the trained unit number threshold prediction model to obtain the second unit number threshold corresponding to the current outgoing line operation status. The first unit number threshold and the second unit number threshold are fused together to obtain the unit number threshold.

6. The method according to any one of claims 1 to 5, characterized in that, The step of performing corresponding start-up and shutdown processing on the target unit according to the operating mode of the automatic power generation control system and the unit start-up and shutdown command includes: When the operating mode is the guidance mode, the unit start-up and shutdown command is sent to the target terminal corresponding to the target unit, so that the target terminal performs start-up and shutdown processing on the target unit according to the unit start-up and shutdown command; When the operating mode is automatic, the target unit is started and stopped according to the unit start-up and shutdown instructions.

7. A smart decision-making device for start-up and shutdown of an automatic power generation control system based on multiple constraints, characterized in that, The device includes: The data acquisition module is used to acquire the current active power capacity setpoint and historical active power capacity setpoint of the automatic power generation control system in response to the unit start-up and shutdown request of the automatic power generation control system. The amplitude determination module is used to determine the amplitude of the setpoint corresponding to the automatic power generation control system based on the current active power capacity setpoint and the historical active power capacity setpoint. The unit number prediction module is used to determine the current set value interval information corresponding to the current active capacity set value when the current active capacity set value is within the active capacity set range and the set value variation is within the set value variation range. The current set value interval information is then input into the trained unit number prediction model to obtain the number of units that the automatic power generation control system needs to start and stop. The unit count update module is used to update the unit count according to the unit lockout status of the current unit in the automatic power generation control system when the number of units is less than or equal to the unit count threshold, so as to obtain the updated unit count. The unit determination module is used to obtain the unit priority of the current unit, and determine the target unit corresponding to the unit start / stop request from the current unit based on the updated number of units and the unit priority; The unit start-stop module is used to generate start-stop commands for the target unit and perform corresponding start-stop processing on the target unit according to the operating mode of the automatic power generation control system and the start-stop commands.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images