Scheduling method and device of unit load, medium, equipment and program product

Abstract

The invention relates to a unit load scheduling method and device, a medium, equipment and a program product, and relates to the technical field of hydraulic engineering, and the method can comprise the steps: determining a plurality of first load distribution schemes corresponding to the unit load of a target unit; determining the fitness corresponding to each first load distribution scheme, wherein the fitness is used for evaluating the advantages and disadvantages of the first load distribution scheme; determining a target load distribution scheme meeting a target constraint condition according to the plurality of first load distribution schemes and the fitness corresponding to each first load distribution scheme; and controlling the target unit to execute the target load distribution scheme. Therefore, the optimal load distribution scheme can be accurately found out from the plurality of first load distribution schemes through a small calculation amount. Meanwhile, under the condition that the target constraint condition is met, reasonable distribution of unit loads can be achieved, and therefore maximum resource utilization of unit operation is achieved, and economic operation is achieved.

Description

Technical Field

[0001] The present disclosure relates to the technical field of water conservancy engineering, and in particular to a method, device, medium, equipment and program product for scheduling unit load. Background Art

[0002] Hydropower stations are a clean, renewable form of energy. Through effective and economical operation, they can ensure the stability and sustainability of electricity supply, meet society's demand for electricity, and reduce dependence on and consumption of mineral resources.

[0003] Optimizing unit load scheduling is of great significance to the economic operation of hydropower stations. However, in actual scenarios, due to the dispersion of hydropower station data and diverse sources, data integration and analysis are difficult, resulting in the lack of intelligence and real-time performance in unit load scheduling methods, and limited equipment monitoring methods, making it difficult to achieve remote real-time monitoring and intelligent maintenance. Summary of the invention

[0004] In order to solve the above technical problems, the present disclosure provides a method, device, medium, equipment and program product for scheduling unit load.

[0005] In a first aspect, the present disclosure provides a method for dispatching unit loads, the method comprising: Determine a plurality of first load distribution schemes corresponding to the unit load of the target unit; Determine a fitness corresponding to each of the first load distribution schemes, where the fitness is used to evaluate the pros and cons of the first load distribution schemes; Determine a target load distribution scheme that satisfies a target constraint condition according to a plurality of first load distribution schemes and the fitness corresponding to each of the first load distribution schemes; The target unit is controlled to execute the target load distribution plan.

[0006] Optionally, determining a target load distribution scheme that satisfies a target constraint condition according to a plurality of first load distribution schemes and the fitness corresponding to each of the first load distribution schemes includes: Determine a plurality of second load distribution schemes whose fitness is greater than or equal to a preset fitness threshold from the plurality of first load distribution schemes; Based on the plurality of second load distribution schemes, determining a plurality of third load distribution schemes; In the case that there is a load distribution scheme satisfying the target constraint condition among the multiple third load distribution schemes, the third load distribution scheme with the greatest fitness is used as the target load distribution scheme.

[0007] Optionally, determining a plurality of third load distribution schemes based on the plurality of second load distribution schemes comprises: Based on the multiple second load distribution schemes, crossover operations and mutation operations are performed in sequence to obtain the multiple third load distribution schemes.

[0008] Optionally, the method further comprises: Obtaining environmental rainfall data corresponding to the target unit in the target area; Constructing a rainfall area map corresponding to the target area according to the environmental rainfall data; The accumulated rainfall area map and the environmental rainfall data are input into a pre-trained water regime prediction model to obtain water regime information corresponding to the target area in the future output by the water regime prediction model.

[0009] Optionally, the water regime prediction model is pre-trained by the following method: Acquire a training sample set, wherein the training sample set includes a plurality of sample inputs and a plurality of sample outputs; The preset training model is trained according to the multiple sample inputs and the multiple sample outputs, and the trained preset training model is used as the water regime prediction model.

[0010] Optionally, the method further comprises: Get a test sample set; The using the trained preset training model as the water regime prediction model comprises: The test sample set is input into the trained preset training model for verification, and according to the verification result, the model parameters in the trained preset training model are updated, and the updated preset training model is used as the water regime prediction model.

[0011] Optionally, the method further comprises: Obtaining unit operating parameters corresponding to the target unit; The unit operating parameters are input into a pre-constructed target state space equation to obtain an operating performance result corresponding to the target unit.

[0012] Optionally, the method further comprises: Obtaining the initial water level, initial gate opening and initial unit load corresponding to the target unit; Inputting the initial water level, the initial gate opening and the initial unit load into a pre-built target water level control model to obtain a target control strategy output by the target water level control model; According to the target control strategy, the gate opening and unit load of the target unit are adjusted.

[0013] Optionally, the target water level control model is pre-constructed in the following manner: Obtaining a gate discharge curve and a water level control target, wherein the gate discharge curve is used to characterize the corresponding relationship between the gate opening and the water level, and the water level control target is used to characterize the expected maintenance range of the water level; The target water level control model is constructed according to the gate discharge curve and the water level control target.

[0014] In a second aspect, the present disclosure provides a device for dispatching unit loads, the device comprising: A first determination module, used to determine a plurality of first load distribution schemes corresponding to the unit load of the target unit; A second determination module, used to determine the fitness corresponding to each of the first load distribution schemes, wherein the fitness is used to evaluate the quality of the first load distribution scheme; A third determination module, configured to determine a target load distribution scheme that satisfies a target constraint condition according to a plurality of first load distribution schemes and a fitness corresponding to each of the first load distribution schemes; A control module is used to control the target unit to execute the target load distribution plan.

[0015] Optionally, the third determination module is used to determine multiple second load distribution schemes whose fitness is greater than or equal to a preset fitness threshold from the multiple first load distribution schemes; determine multiple third load distribution schemes based on the multiple second load distribution schemes; and if there is a load distribution scheme that satisfies the target constraint condition among the multiple third load distribution schemes, the third load distribution scheme with the largest fitness is used as the target load distribution scheme.

[0016] Optionally, the third determination module is used to perform a crossover operation and a mutation operation in sequence based on the multiple second load distribution schemes to obtain the multiple third load distribution schemes.

[0017] Optionally, the device further comprises: The first acquisition module is used to obtain the environmental rainfall data corresponding to the target unit in the target area; A construction module, used for constructing a rainfall area map corresponding to the target area according to the environmental rainfall data; The prediction module is used to input the accumulated rainfall area map and the environmental rainfall data into a pre-trained water situation prediction model to obtain the water situation information corresponding to the target area in the future output by the water situation prediction model.

[0018] Optionally, the water regime prediction model is pre-trained by the following method: Acquire a training sample set, wherein the training sample set includes a plurality of sample inputs and a plurality of sample outputs; The preset training model is trained according to the multiple sample inputs and the multiple sample outputs, and the trained preset training model is used as the water regime prediction model.

[0019] Optionally, the method further comprises: Get a test sample set; The using the trained preset training model as the water regime prediction model comprises: The test sample set is input into the trained preset training model for verification, and according to the verification result, the model parameters in the trained preset training model are updated, and the updated preset training model is used as the water regime prediction model.

[0020] Optionally, the device further comprises: A second acquisition module is used to acquire the unit operating parameters corresponding to the target unit; The first simulation module is used to input the unit operation parameters into a pre-constructed target state space equation to obtain the operation performance result corresponding to the target unit.

[0021] Optionally, the device further comprises: A third acquisition module is used to obtain the initial water level, initial gate opening and initial unit load corresponding to the target unit; A second simulation module is used to input the initial water level, the initial gate opening and the initial unit load into a pre-built target water level control model to obtain a target control strategy output by the target water level control model; The regulating module is used to regulate the gate opening and unit load of the target unit according to the target control strategy.

[0022] Optionally, the target water level control model is pre-constructed in the following manner: Obtaining a gate discharge curve and a water level control target, wherein the gate discharge curve is used to characterize the corresponding relationship between the gate opening and the water level, and the water level control target is used to characterize the expected maintenance range of the water level; The target water level control model is constructed according to the gate discharge curve and the water level control target.

[0023] In a third aspect, the present disclosure provides a non-temporary computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the unit load scheduling method provided in the first aspect of the present disclosure.

[0024] In a fourth aspect, the present disclosure provides an electronic device, comprising: a memory on which a computer program is stored; and a processor for executing the computer program in the memory to implement the steps of the unit load scheduling method provided in the first aspect of the present disclosure.

[0025] In a fifth aspect, the present disclosure provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of the unit load scheduling method provided in the first aspect of the present disclosure.

[0026] Through the above technical solution, first, multiple first load distribution schemes corresponding to the unit load of the target unit are determined. After that, the fitness corresponding to each of the first load distribution schemes is determined, and the fitness is used to evaluate the pros and cons of the first load distribution scheme. Then, according to the fitness corresponding to the multiple first load distribution schemes and each of the first load distribution schemes, the target load distribution scheme that meets the target constraint condition is determined, and the target constraint condition is used to characterize the operation target corresponding to the target unit. Finally, the target unit is controlled to execute the target load distribution scheme. By determining the fitness of each first load distribution scheme, according to the fitness corresponding to each first load distribution scheme, the target load distribution scheme that meets the target constraint condition is determined from the multiple first load distribution schemes, and the optimal load distribution scheme can be accurately found from the multiple first load distribution schemes through a small amount of calculation. At the same time, it is possible to achieve a reasonable distribution of the unit load while meeting the target constraint condition, thereby maximizing the resource utilization of the unit operation and achieving economic operation.

[0027] Other features and advantages of the present disclosure will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The accompanying drawings are used to provide a further understanding of the present disclosure and constitute a part of the specification. Together with the following specific embodiments, they are used to explain the present disclosure but do not constitute a limitation of the present disclosure. In the accompanying drawings: Figure 1 The present invention is a flow chart of a method for dispatching unit load according to an exemplary embodiment.

[0029] Figure 2 The present invention is a flow chart showing another method for dispatching unit load according to an exemplary embodiment.

[0030] Figure 3 The present invention is a flow chart showing another method for dispatching unit load according to an exemplary embodiment.

[0031] Figure 4 The present invention is a flow chart showing another method for dispatching unit load according to an exemplary embodiment.

[0032] Figure 5 It is a block diagram of a device for dispatching unit load according to an exemplary embodiment.

[0033] Figure 6 It is a block diagram of another device for dispatching unit load according to an exemplary embodiment.

[0034] Figure 7 It is a block diagram of another device for dispatching unit load according to an exemplary embodiment.

[0035] Figure 8 It is a block diagram of another device for dispatching unit load according to an exemplary embodiment.

[0036] Fig. 9 It is a block diagram of an electronic device according to an exemplary embodiment. DETAILED DESCRIPTION

[0037] The specific implementation of the present disclosure is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described herein is only used to illustrate and explain the present disclosure, and is not used to limit the present disclosure.

[0038] The terms "first", "second", etc. in the specification and claims of this application and the above drawings are used to distinguish similar objects and do not have to be understood as a specific order or sequence. In addition, in the description with reference to the drawings, the same symbols in different drawings represent the same elements.

[0039] In the description of the present disclosure, unless otherwise specified, "multiple" means two or more than two, and other quantifiers are similar thereto; "at least one item", "one item or multiple items" or similar expressions refer to any combination of these items, including any combination of singular or plural items. For example, at least one item a can represent any number of a; for another example, one item or multiple items among a, b and c can represent: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple; "and / or" is a kind of description of the association relationship of associated objects, indicating that there can be three kinds of relationships, for example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone, where A and B can be singular or plural. The character " / " indicates that the associated objects before and after are in an "or" relationship.

[0040] Before introducing the unit load scheduling method, apparatus, medium, equipment and program product provided by the present disclosure, the application scenarios involved in each embodiment of the present disclosure are first introduced.

[0041] The economic operation of hydropower station units is mainly affected by factors such as water inflow, climate change, equipment performance, market demand and policy regulation. The economic operation of hydropower stations is a systematic project involving resource integration, technology integration, business management, policies and regulations. Although the economic operation of hydropower stations has made some progress at this stage, it still faces many challenges in practical applications: (1) Water conditions are affected by climate change, river morphology, geological structure, etc., which increases uncertainty and affects the economic operation of hydropower stations. Some hydropower stations have serious aging equipment, untimely maintenance, and low efficiency, which affects power generation efficiency and economic benefits.

[0042] (2) Traditional load dispatching relies on manual experience, which makes load dispatching difficult, with low power generation efficiency and difficulty in adapting to the complex and changing market and technical environment. The automation level of some hydropower stations is not high, and it is difficult to adjust the load and discharge parameters of the units, resulting in delayed response and insufficient regulation accuracy.

[0043] (3) Some hydropower plants still use traditional operation and management models, have not adopted advanced information technology, and lack systematic support. Hydropower stations have information asymmetry, and the collection and monitoring parameters are not comprehensive and accurate, which lacks the comprehensiveness and accuracy of business decisions.

[0044] In order to solve the above technical problems, the present invention provides a method, device, medium, equipment and program product for dispatching unit loads, by determining the fitness of each first load distribution scheme, and according to the fitness corresponding to each first load distribution scheme, determining a target load distribution scheme that meets the target constraint conditions from multiple first load distribution schemes, and accurately finding the optimal load distribution scheme from multiple first load distribution schemes through a small amount of calculation. At the same time, it is possible to achieve reasonable distribution of unit loads while meeting the target constraint conditions, thereby maximizing resource utilization of unit operation and achieving economic operation.

[0045] The specific implementation modes of the present invention are described in detail below with reference to the accompanying drawings.

[0046] Figure 1 is a flow chart of a method for dispatching unit load according to an exemplary embodiment. Figure 1 As shown, the method may include the following steps: In step S101, a plurality of first load distribution schemes corresponding to the unit load of the target unit are determined.

[0047] The target unit may be, for example, a unit corresponding to a hydropower station.

[0048] In this step, the unit load parameters of the target unit within a period of time can be obtained, and then the unit load parameters are divided into multiple discrete time periods. Afterwards, an initial load distribution plan is randomly generated for the unit load parameters within each discrete time period, thereby obtaining multiple first load distribution plans. Among them, the load distribution plan is usually used to indicate the unit load participating in the distribution, the working time of the unit load, and the unsafe area of ​​the unit (that is, the load cannot be distributed in this area). For example, the load distribution plan may include a 24-hour distribution plan for the unit load, such as the total load for 24 hours, the load of the unit participating in the load distribution, the current load, and the unsafe area of ​​the unit.

[0049] In this embodiment, the scheduling of the unit load can be optimized by a genetic algorithm. Therefore, when a plurality of first load distribution schemes are obtained, the plurality of first load distribution schemes can be used as individuals of a population.

[0050] In step S102, the fitness corresponding to each of the first load distribution schemes is determined.

[0051] The fitness is used to evaluate the quality of the first load distribution scheme.

[0052] In genetic algorithms, fitness is usually used to describe individual performance. According to the size of fitness, individuals are selected to determine the survival of the fittest. Fitness is the driving force of genetic algorithms. From a biological perspective, fitness is equivalent to the biological survival ability of "survival of the fittest" and is of great significance in the genetic process.

[0053] In this step, the total load corresponding to each first load distribution scheme is the same, but the total energy consumption corresponding to each first load distribution scheme is not necessarily the same. In the hydropower station scenario, different total energy consumption means different power generation costs and power generation efficiencies. In this embodiment, in order to maximize power generation efficiency and minimize power generation costs, the fitness function in the genetic algorithm can be used to calculate the fitness corresponding to each first load distribution scheme, thereby evaluating the total energy consumption corresponding to each first load distribution scheme.

[0054] In some embodiments, the load distribution target of the hydropower station can be predetermined, and relevant constraints can be introduced. For example, the load distribution target can include that the units participating in the distribution have the maximum output, maintain the highest water level, and abandon the least water. Constraints can include reducing abandoned water, maximizing resources, avoiding unsafe operation areas of the units, and the most stable operation of the units. Then, according to the load distribution target and the constraints, a corresponding objective function is established. By establishing a mapping relationship between the objective function of the optimization problem and the fitness of the individual, the optimization of the objective function can be achieved in the process of group evolution. The fitness function, also known as the evaluation function, is a standard for distinguishing the good and bad individuals in the group determined according to the objective function. According to the fitness function, the fitness of each first load distribution scheme (that is, the individual) is determined, that is, the optimal result that satisfies the objective function is solved.

[0055] In step S103, a target load distribution scheme that meets the target constraint condition is determined according to a plurality of first load distribution schemes and the fitness corresponding to each of the first load distribution schemes.

[0056] In this step, first, according to the fitness corresponding to each of the first load distribution schemes, multiple second load distribution schemes whose fitness is greater than or equal to the preset fitness threshold can be determined from the multiple first load distribution schemes. That is, individuals in the population are selected according to the fitness, and individuals with higher fitness are selected as parent individuals for breeding the next generation (that is, selection operation).

[0057] Secondly, based on the plurality of second load distribution schemes, a plurality of third load distribution schemes are determined.

[0058] For example, based on the multiple second load distribution schemes, a crossover operation and a mutation operation are performed in sequence to obtain the multiple third load distribution schemes.

[0059] Among them, the meaning of the crossover operation in the genetic algorithm is to exchange part of the chromosomes of the parent individuals through the crossover operation to generate new individuals and retain excellent genes. The meaning of the mutation operation is to perform a mutation operation on the newly generated individuals to introduce randomness to increase the diversity of the search. In some embodiments, multiple second load distribution schemes can be expanded through crossover operations and mutation operations, and the solved results can be automatically verified in combination with the unit operation rules or experience, and the solution that meets the requirements of the objective function is selected as the parent of the iteration, and is screened again and verified again. Afterwards, by replacing the individuals with lower fitness in the original population with the newly generated individuals, a new generation of populations is formed (i.e., replacement operation), and the final multiple third load distribution schemes are obtained.

[0060] Finally, when there is a load distribution scheme satisfying the target constraint condition among the multiple third load distribution schemes, the third load distribution scheme satisfying the target constraint condition with the greatest fitness is used as the target load distribution scheme.

[0061] The target constraint condition may include one or more of reaching a maximum number of iterations, population similarity satisfying a preset similarity condition, and external factors (eg, user intervention stop, etc.).

[0062] That is to say, by repeatedly performing selection operations, crossover operations, mutation operations and replacement operations on multiple second load distribution schemes until the target constraints are met, the third load distribution scheme with the greatest fitness can be used as the target load distribution scheme, that is, the optimal load distribution scheme for the hydropower station units.

[0063] In step S104, the target unit is controlled to execute the target load distribution plan.

[0064] In this way, the optimal target load distribution plan is determined by the above method, and the genetic algorithm is used to optimize the load scheduling of the hydropower station units to achieve the goal of minimizing the power generation cost or maximizing the power generation efficiency, ensuring the stable and efficient operation of the units.

[0065] By adopting the above method, by determining the fitness of each first load distribution scheme, according to the fitness corresponding to each first load distribution scheme, a target load distribution scheme that meets the target constraint condition is determined from multiple first load distribution schemes, and the optimal load distribution scheme can be accurately found from multiple first load distribution schemes through a small amount of calculation. At the same time, it is possible to achieve a reasonable distribution of unit loads while meeting the target constraint condition, thereby maximizing the resource utilization of the unit operation and achieving economic operation.

[0066] Considering that in some scenarios, the water conditions of hydropower stations are easily affected by climate change, river morphology, geological structure, etc., the uncertainty factor is increased, which affects the economic operation of the hydropower station. In order to solve the above problems and realize the prediction of water conditions changes, in some embodiments, such as Figure 2 As shown, the method may also include the following steps: In step S105, the corresponding environmental rainfall data of the target unit in the target area is obtained.

[0067] The target area may be the area where the target unit is located.

[0068] In this step, environmental rainfall data in the target area can be collected. The environmental rainfall data may include, for example but not limited to, rainfall period, rainfall intensity, longitude and latitude, rainfall, rainfall area distribution, rainfall spatial distribution, water level, etc. It may also include parameters such as the watershed soil type and vegetation coverage rate corresponding to the target area.

[0069] In step S106, a rainfall area map corresponding to the target area is constructed according to the environmental rainfall data.

[0070] In this step, the environmental rainfall data can be processed including but not limited to cleaning, conversion, selection, construction, dimensionality reduction, etc., to obtain the corresponding rainfall characteristic values. Among them, the cleaning process refers to deleting duplicate values ​​in the environmental rainfall data, and interpolating the missing values ​​using the mean, median, mode, etc. The conversion process refers to normalizing or standardizing the environmental rainfall data to achieve the conversion and scaling of the data format. The selection process refers to selecting the characteristic value most relevant to the target variable, wherein the target variable includes, for example, flood volume, rainfall, etc. The construction process refers to creating new features based on the environmental rainfall data, such as the interaction between two or more variables, such as multiplication, division, and summation. The dimensionality reduction process refers to reducing the dimension of the environmental rainfall data while retaining the information of the original data as much as possible.

[0071] In some embodiments, a rainfall area map corresponding to the target area is constructed based on the environmental rainfall data acquired within a certain period of time, wherein the rainfall area map is used to represent the accumulated rainfall in the target area within a certain period of time.

[0072] In step S107, the accumulated rainfall area map and the environmental rainfall data are input into a pre-trained water regime prediction model to obtain water regime information corresponding to the target area in the future output by the water regime prediction model.

[0073] In this step, the accumulated rainfall area map and the environmental rainfall data can be used as input parameters and input into the pre-trained water regime prediction model to obtain the water regime information corresponding to the target area in the future output by the water regime prediction model. The water regime information may include, for example, the rainfall, water level, water flow and other results at a certain point in the future or in the target area.

[0074] In some embodiments, the water regime prediction model is pre-trained by the following method: First, obtain the training sample set.

[0075] The training sample set includes multiple sample inputs and multiple sample outputs. The sample input may include environmental rainfall data collected in a historical period, and the sample output is a real water condition result corresponding to each rainfall environmental data.

[0076] Then, the preset training model is trained according to the multiple sample inputs and the multiple sample outputs, and the trained preset training model is used as the water regime prediction model.

[0077] The preset training model may include, for example, a support vector machine (SVM). The preset training model is trained according to the multiple sample inputs and the multiple sample outputs, and the trained preset training model is used as the water regime prediction model. The model training method in the relevant technology can be referred to, and will not be elaborated here. By training the preset training model, the model parameters are adjusted to obtain the best prediction performance, and the corresponding water regime prediction model is obtained.

[0078] In other embodiments, a part of the environmental rainfall data collected in the historical time period can be extracted as a test sample set. After the preset training model is trained by the training sample set, the trained preset training model can be verified by the test sample set to evaluate the prediction ability of the model. And according to the verification results, the trained preset training model can be further optimized by adjusting the model parameters or changing the input parameters to improve the prediction accuracy of the model. According to the verification results, the updated preset training model is used as the water situation prediction model.

[0079] In this way, water situation prediction is carried out through the water situation prediction model to realize real-time monitoring of the water situation of the hydropower station and predict future changes in water situation, so as to adjust the current control strategy in time and ensure the safe and stable operation of the hydropower station units.

[0080] After obtaining the above-mentioned water situation prediction results (i.e. water situation information), if it is determined that the water level of the hydropower station will rise significantly in the future, in order to ensure the safe operation of the unit, it is necessary to adjust the gates of the hydropower station accordingly to control the water level and flow changes of the hydropower station in advance and ensure the safe and reliable operation of the hydropower station. Specifically, Figure 3 As shown, the method may further include the following steps: In step S108, the initial water level, initial gate opening and initial unit load corresponding to the target unit are obtained.

[0081] Among them, the initial water level, initial gate opening and initial unit load can be understood as the water level, gate opening and unit load corresponding to the target unit at the current moment.

[0082] In step S109, the initial water level, the initial gate opening and the initial unit load are input into a pre-constructed target water level control model to obtain a target control strategy output by the target water level control model.

[0083] The target control strategy may include, for example, a target gate opening and a target unit load.

[0084] In one application scenario, when it is determined based on water conditions that the water level of a hydropower station will rise significantly in the future, the initial water level, the initial gate opening and the initial unit load can be input into a pre-built target water level control model to obtain a target control strategy output by the target water level control model.

[0085] In some embodiments, the target water level control model is pre-constructed in the following manner: First, the gate discharge curve and water level control target can be obtained.

[0086] Among them, the gate discharge curve is used to characterize the corresponding relationship between the gate opening and the water level, and the water level control target is used to characterize the expected water level maintenance range, that is, the expected water level is maintained within an appropriate range.

[0087] Specifically, a relationship model between gate opening and water level can be established in advance, and the gate discharge curve can be dynamically fitted through a deep learning model. Specifically, the relationship between gate opening and water level can be modeled through some basic hydraulic principles, based on the common empirical methods of energy conservation and continuity equations.

[0088] For example, for an actual gate, its discharge flow is usually affected by factors such as gate opening and water level. Therefore, the discharge flow of the target unit can be determined by the following formula: Discharge flow (Q) calculation formula, that is, Q=C·A·v, where: C is the flow coefficient, which depends on the gate design and the dynamic characteristics of the fluid, A is the opening area of ​​the gate, which is directly related to the gate opening, and v is the outflow velocity. Then, the outflow velocity v can be determined according to the Torricelli formula, that is, v=√(2gh), where: v is the outflow velocity (m / s), g is the acceleration of gravity (about 9.81 m / s²), and h is the vertical distance from the head or fluid surface to the outflow hole (m). That is, the discharge flow Q=C·A·√(2gh) can be obtained.

[0089] After the discharge flow is obtained, a gate discharge curve fitting model can be established according to the discharge flow. In this way, after the gate opening and water level are obtained, the corresponding discharge flow can be obtained through the gate discharge curve fitting model.

[0090] Then, the target water level control model is constructed according to the gate discharge curve and the water level control target.

[0091] Specifically, the target water level control model can be constructed in the following ways: Define water level control objectives: Determine the water level control objectives that the system needs to achieve, such as maintaining the water level within a certain range, minimizing water level fluctuations, etc.

[0092] System identification: Collect the input data and output data of the system. The input data may include water level, gate opening and unit load, while the output data includes gate opening and unit load.

[0093] The change of water level can be regarded as the accumulation and discharge process of water. Based on this, the dynamic equation is established: frac{dV}{dt}=Q_{in}-Q_{out} Where V is the amount of water in the container, Q_{in} is the amount of water flowing in, and Q_{out} is the amount of water flowing out (depending on overflow, drain port design, etc.).

[0094] Parameterization: Determine the parameters in the model, such as the geometry of the container (used to convert the water volume V to the water level h), the outflow coefficient, etc.

[0095] Control strategy design: Design a feedback control strategy such as PID controller to adjust the outflow (Q_{out}) to maintain the water level h within the desired range.

[0096] Based on the above method, the target water level control model can be gradually obtained.

[0097] After obtaining the target water level control model, the model can be simulated and tested, using computer simulation to test the performance of the control strategy under different circumstances, and sensitivity analysis can be performed to understand the impact of key parameters. At the same time, scenarios can also be analyzed, such as simulating extreme weather, different seasons, and different demand scenarios to ensure the robustness of the control strategy. After that, the operation data after implementation can be collected and its effect can be monitored. Control parameters and strategies can be adjusted as needed. And in the subsequent use process, the model operation results are continuously monitored to ensure the achievement of control objectives. Regularly evaluate and upgrade the control system to cope with environmental and technological changes.

[0098] In step S110, the gate opening and unit load of the target unit are adjusted according to the target control strategy.

[0099] Accordingly, the gate opening of the target unit can be adjusted according to the target gate opening, and the unit load of the target unit can be adjusted according to the target unit load to optimize the control of the water level, which can achieve more accurate and stable water level control and improve the efficiency and safety of water conservancy projects.

[0100] In some scenarios, in order to maximize the performance of the target unit and avoid risks such as adverse operating conditions and vibration zones, the power end of the unit can also be simulated to fully understand and evaluate the unit performance and stability. Figure 4 As shown, the method may further include the following steps: In step S111, the unit operating parameters corresponding to the target unit are obtained.

[0101] The unit operation parameters may include mechanical operation parameters, electrical operation parameters and hydraulic operation parameters. Mechanical operation parameters may include inertia, speed, vibration, efficiency, damping coefficient, etc. Electrical operation parameters may include voltage, current, load characteristics and control parameters of various sensors, controllers and adjustment devices, which are used to monitor and adjust the operating status and performance of the system to ensure that the parameter settings are in line with the actual situation. Hydraulic operation parameters may include water flow, water level, water quality, etc.

[0102] In step S112, the unit operating parameters are input into a pre-constructed target state space equation to obtain an operating performance result corresponding to the target unit.

[0103] The operating performance results may include the unit operating status and / or unit operating conditions. The unit operating status may include water flow power, turbine performance, generator electrical characteristics, control system response, stability and failure, efficiency and environment, etc. The unit operating conditions may include normal operation, start and stop, load change, frequency regulation and peak regulation, extreme conditions, emergency situations, system joint adjustment, etc. In practical applications, the operating performance results of different states or conditions of the simulation model can be selected according to needs.

[0104] The application of the target state space equation is described in detail below: (1) System modeling: The power system of the unit is modeled as a mathematical model, including the mechanical part, electrical part (including the control system) and hydraulic part, etc. The state space equation of the power system is established by combining traditional physical modeling methods with modern control theory.

[0105] Specifically, the traditional state space equation can be modeled as: [ dot{x} = Ax + Bu ] [ y = Cx ] Among them, x is the state vector of the system, A is the state transfer matrix, B is the input control matrix, u is the control input vector, y is the output vector, and C is the output matrix.

[0106] In this embodiment, a deep learning model can be introduced to learn the nonlinear characteristics and unknown dynamics of the power system, which can be achieved by using models such as recurrent neural networks (RNNs) or long short-term memory networks (LSTMs). Specifically, a neural network model can be used to approximate the state transfer function or output function of the power system, thereby establishing a state space equation based on deep learning, that is, a target state space equation.

[0107] [ dot{x} = f(x, u, theta) ] [ y = g(x, theta) ] Among them, x is the state vector of the system, u is the control input vector, f and g represent the state transition and output functions respectively, theta is the parameter of the neural network model, and y is the output vector.

[0108] In this embodiment, by using a neural network to approximate the state transfer function and output function of the power system, complex dynamic characteristics such as nonlinearity and time variation can be better handled, and learning and modeling of unknown dynamics can be achieved.

[0109] (2) Simulation environment selection: Select appropriate simulation tools and environments, such as Matlab / Simulink, LabVIEW, etc., to build a simulation platform for the power system.

[0110] (3) Parameter setting: Using the above-mentioned mechanical operating parameters, electrical operating parameters and hydraulic operating parameters as the input of the equation can more realistically simulate the operating behavior of the unit under different water conditions, providing an important reference for actual production operation.

[0111] (4) Initial state setting: Setting the initial state and parameters of each component of the power system, including but not limited to turbine, generator characteristic parameters, water conditions, governor, reservoir and waterway system, auxiliary equipment and other components.

[0112] In the simulation model of the power section of the hydro-generator set, the initial states and parameters of each component include but are not limited to the following aspects: Characteristic parameters of turbines and generators: including speed characteristic curves, efficiency curves, moment of inertia and other parameters of turbines and generators, which directly affect the power performance of the unit.

[0113] Water condition parameters: including water level, water flow and other water condition parameters, which have a direct impact on the output power and operating status of the unit.

[0114] Parameters of the turbine speed control system: including the characteristic curve of the speed governor, the response time of the speed control system and other parameters. These parameters affect the response speed of the unit to changes in water flow and water level.

[0115] Parameters of the reservoir and waterway system of a hydroelectric power station: including parameters such as reservoir volume, reservoir water level, waterway water level and water flow rate. These parameters have a direct impact on the head and water inlet flow of the unit.

[0116] Parameters of other auxiliary equipment: including parameters of auxiliary equipment such as pressure regulating valves, water pipes, cooling systems, etc. These parameters will also affect the operating status and performance of the unit.

[0117] (5) Simulation operation: Run the power system simulation model to analyze the unit's power response, performance, stability, health status, etc.

[0118] Among them, for the simulation of health status, the degradation trend analysis of the key performance indicators of the equipment can be performed based on the physical design parameters of the equipment, so as to quantify the impact of faults on equipment performance and other methods to simulate equipment performance degradation and equipment failure, and provide equipment operating limit and performance threshold warnings. The simulated data is compared with the real-time collected data, and the health status of the equipment is predicted through deviation analysis. System simulation predicts the problems that the equipment may encounter in actual operation during the design phase to reduce future maintenance costs and potential downtime.

[0119] In addition, after simulating the operating performance results of the target unit through the target state space equation, the parameters can be optimized and adjusted according to the simulation results, and the simulation results can be analyzed and the performance evaluated to make the simulation results of the model closer to the actual situation. The simulation results can also be analyzed to evaluate the efficiency, performance and stability of the unit, predict the health status of the equipment, find potential problems and make improvements.

[0120] Through the above steps, the target state space equation can be used to simulate the power section of the target unit, evaluate the performance and stability of the unit, and provide an important reference for the design and commissioning of the hydropower station unit. It improves computing power and has higher reliability and accuracy. It is highly coupled with the physical scene and has strong practicality.

[0121] Figure 5 is a block diagram of a scheduling device for unit load according to an exemplary embodiment. Figure 5 As shown, the device 200 includes: A first determination module 201 is used to determine a plurality of first load distribution schemes corresponding to the unit load of the target unit; A second determination module 202 is used to determine the fitness corresponding to each of the first load distribution schemes, where the fitness is used to evaluate the quality of the first load distribution scheme; A third determination module 203 is used to determine a target load distribution scheme that meets the target constraint condition according to the plurality of first load distribution schemes and the fitness corresponding to each of the first load distribution schemes; The control module 204 is used to control the target unit to execute the target load distribution plan.

[0122] Optionally, the third determination module 203 is used to determine, from the multiple first load distribution schemes, multiple second load distribution schemes whose fitness is greater than or equal to a preset fitness threshold; based on the multiple second load distribution schemes, determine multiple third load distribution schemes; if there is a load distribution scheme that satisfies the target constraint condition among the multiple third load distribution schemes, the third load distribution scheme with the largest fitness is used as the target load distribution scheme.

[0123] Optionally, the third determination module 203 is used to perform a crossover operation and a mutation operation in sequence based on the multiple second load distribution schemes to obtain the multiple third load distribution schemes.

[0124] Alternatively, if Figure 6 As shown, the device 200 also includes: The first acquisition module 205 is used to obtain the corresponding environmental rainfall data of the target unit in the target area; A construction module 206 is used to construct a rainfall area map corresponding to the target area according to the environmental rainfall data; The prediction module 207 is used to input the accumulated rainfall area map and the environmental rainfall data into a pre-trained water regime prediction model to obtain the water regime information corresponding to the target area in the future output by the water regime prediction model.

[0125] Optionally, the water regime prediction model is pre-trained by the following method: Obtaining a training sample set, the training sample set comprising a plurality of sample inputs and a plurality of sample outputs; The preset training model is trained according to the multiple sample inputs and the multiple sample outputs, and the trained preset training model is used as the water regime prediction model.

[0126] Optionally, the method further comprises: Get a test sample set; The method of using the trained preset training model as the water regime prediction model includes: The test sample set is input into the trained preset training model for verification, and according to the verification result, the model parameters in the trained preset training model are updated, and the updated preset training model is used as the water regime prediction model.

[0127] Alternatively, if Figure 7 As shown, the device 200 also includes: The second acquisition module 208 is used to acquire the unit operation parameters corresponding to the target unit; The first simulation module 209 is used to input the unit operation parameters into a pre-constructed target state space equation to obtain the operation performance result corresponding to the target unit.

[0128] Alternatively, if Figure 8 As shown, the device 200 also includes: The third acquisition module 210 is used to obtain the initial water level, initial gate opening and initial unit load corresponding to the target unit; The second simulation module 211 is used to input the initial water level, the initial gate opening and the initial unit load into a pre-built target water level control model to obtain a target control strategy output by the target water level control model; The adjustment module 212 is used to adjust the gate opening and unit load of the target unit according to the target control strategy.

[0129] Optionally, the target water level control model is pre-constructed in the following manner: Obtaining a gate discharge curve and a water level control target, wherein the gate discharge curve is used to characterize the corresponding relationship between the gate opening and the water level, and the water level control target is used to characterize the expected maintenance range of the water level; According to the gate discharge curve and the water level control target, the target water level control model is constructed.

[0130] By adopting the above device, by determining the fitness of each first load distribution scheme, according to the fitness corresponding to each first load distribution scheme, a target load distribution scheme that meets the target constraint condition is determined from multiple first load distribution schemes, and the optimal load distribution scheme can be accurately found from multiple first load distribution schemes through a small amount of calculation. At the same time, it is possible to achieve a reasonable distribution of unit loads while meeting the target constraint condition, thereby maximizing the resource utilization of the unit operation and achieving economic operation.

[0131] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0132] Fig. 9 is a block diagram of an electronic device 300 according to an exemplary embodiment. For example, the electronic device 300 may be provided as a server. Fig. 9 The electronic device 300 includes a processor 322, which may be one or more, and a memory 332 for storing a computer program executable by the processor 322. The computer program stored in the memory 332 may include one or more modules, each corresponding to a set of instructions. In addition, the processor 322 may be configured to execute the computer program to execute the above-mentioned unit load scheduling method.

[0133] In addition, the electronic device 300 may further include a power supply component 326 and a communication component 350, wherein the power supply component 326 may be configured to perform power management of the electronic device 300, and the communication component 350 may be configured to implement communication of the electronic device 300, for example, wired or wireless communication. In addition, the electronic device 300 may further include an input / output (I / O) interface 358. The electronic device 300 may operate based on an operating system stored in the memory 332.

[0134] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, and when the program instructions are executed by a processor, the steps of the above-mentioned unit load scheduling method are implemented. For example, the non-transitory computer-readable storage medium can be the above-mentioned memory 332 including program instructions, and the above-mentioned program instructions can be executed by the processor 322 of the electronic device 300 to complete the above-mentioned unit load scheduling method.

[0135] In another exemplary embodiment, a computer program product is also provided. The computer program product includes a computer program executable by a programmable device, and the computer program has a code portion for executing the above-mentioned unit load scheduling method when executed by the programmable device.

[0136] The preferred embodiments of the present disclosure are described in detail above in conjunction with the accompanying drawings; however, the present disclosure is not limited to the specific details in the above embodiments. Within the technical concept of the present disclosure, a variety of simple modifications can be made to the technical solution of the present disclosure, and these simple modifications all fall within the protection scope of the present disclosure.

[0137] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the present disclosure will not further describe various possible combinations.

[0138] In addition, various embodiments of the present disclosure may be arbitrarily combined, and as long as they do not violate the concept of the present disclosure, they should also be regarded as the contents disclosed by the present disclosure.

Claims

1. A method for dispatching unit load, characterized in that: The method comprises: Determine a plurality of first load distribution schemes corresponding to the unit load of the target unit; Determine a fitness corresponding to each of the first load distribution schemes, where the fitness is used to evaluate the pros and cons of the first load distribution schemes; Determine a target load distribution scheme that satisfies a target constraint condition according to a plurality of first load distribution schemes and the fitness corresponding to each of the first load distribution schemes; The target unit is controlled to execute the target load distribution plan.

2. The method according to claim 1, characterized in that The determining, according to the plurality of first load distribution schemes and the fitness corresponding to each of the first load distribution schemes, of a target load distribution scheme satisfying the target constraint condition comprises: Determine a plurality of second load distribution schemes whose fitness is greater than or equal to a preset fitness threshold from the plurality of first load distribution schemes; Based on the plurality of second load distribution schemes, determining a plurality of third load distribution schemes; In the case that there is a load distribution scheme satisfying the target constraint condition among the multiple third load distribution schemes, the third load distribution scheme with the greatest fitness is used as the target load distribution scheme.

3. The method according to claim 2, characterized in that The determining a plurality of third load distribution schemes based on the plurality of second load distribution schemes comprises: Based on the multiple second load distribution schemes, crossover operations and mutation operations are performed in sequence to obtain the multiple third load distribution schemes.

4. The method according to claim 1, characterized in that: The method further comprises: Obtaining environmental rainfall data corresponding to the target unit in the target area; Constructing a rainfall area map corresponding to the target area according to the environmental rainfall data; The accumulated rainfall area map and the environmental rainfall data are input into a pre-trained water regime prediction model to obtain water regime information corresponding to the target area in the future output by the water regime prediction model.

5. The method according to claim 4, characterized in that The water regime prediction model is pre-trained by the following method: Acquire a training sample set, wherein the training sample set includes a plurality of sample inputs and a plurality of sample outputs; The preset training model is trained according to the multiple sample inputs and the multiple sample outputs, and the trained preset training model is used as the water regime prediction model.

6. The method according to claim 5, characterized in that The method further comprises: Get a test sample set; The using the trained preset training model as the water regime prediction model comprises: The test sample set is input into the trained preset training model for verification, and according to the verification result, the model parameters in the trained preset training model are updated, and the updated preset training model is used as the water regime prediction model.

7. The method according to claim 1, characterized in that The method further comprises: Obtaining unit operating parameters corresponding to the target unit; The unit operating parameters are input into a pre-constructed target state space equation to obtain an operating performance result corresponding to the target unit.

8. The method according to any one of claims 1 to 7, characterized in that The method further comprises: Obtaining the initial water level, initial gate opening and initial unit load corresponding to the target unit; Inputting the initial water level, the initial gate opening and the initial unit load into a pre-built target water level control model to obtain a target control strategy output by the target water level control model; According to the target control strategy, the gate opening and unit load of the target unit are adjusted.

9. The method according to claim 8, characterized in that The target water level control model is pre-constructed in the following way: Obtaining a gate discharge curve and a water level control target, wherein the gate discharge curve is used to characterize the corresponding relationship between the gate opening and the water level, and the water level control target is used to characterize the expected maintenance range of the water level; The target water level control model is constructed according to the gate discharge curve and the water level control target.

10. A dispatching device for unit load, characterized in that: The device comprises: A first determination module, used to determine a plurality of first load distribution schemes corresponding to the unit load of the target unit; A second determination module, used to determine the fitness corresponding to each of the first load distribution schemes, wherein the fitness is used to evaluate the quality of the first load distribution scheme; A third determination module, configured to determine a target load distribution scheme that satisfies a target constraint condition according to a plurality of first load distribution schemes and a fitness corresponding to each of the first load distribution schemes; A control module is used to control the target unit to execute the target load distribution plan.

11. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 9 are implemented.

12. An electronic device, characterized in that: include: a memory having a computer program stored thereon; A processor, configured to execute the computer program in the memory to implement the steps of the method according to any one of claims 1 to 9.

13. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 9 are implemented.

Images