Unit prediction state verification method and device, electronic equipment and storage medium

Abstract

The invention provides a unit prediction state verification method and device, electronic equipment and a storage medium, and belongs to the technical field of power systems. The unit prediction state of each power unit in a power unit combination in each prediction period is predicted through a machine learning model; the method comprises the following steps: acquiring an initial unit state of each power unit in a power unit combination in an initial time period, updating a unit prediction state of each power unit in a first prediction time period according to the initial unit state to obtain a unit feasible state of the first prediction time period, and taking a second prediction time period as a starting point to obtain a unit feasible state of the first prediction time period; and updating the unit prediction state of the power unit in the current prediction time period according to the unit feasible state of the power unit in the previous prediction time period to obtain the unit feasible state of the power unit in the current prediction time period until the unit feasible state of the power unit in the last prediction time period is obtained. The feasibility of the unit prediction state can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of power systems, and in particular to a method, device, electronic equipment, and storage medium for verifying predicted status of a unit. Background Art

[0002] In the related art, a machine learning model is used to predict the unit state of a power unit combination to achieve a rapid solution to the unit combination problem. However, the machine learning model often generates unit state prediction values ​​that violate unit constraints, resulting in a reduction in the feasibility of the unit state prediction values. Summary of the invention

[0003] The main purpose of the embodiments of the present application is to propose a unit prediction state verification method, device, electronic device, and storage medium, aiming to improve the feasibility of the unit state prediction value.

[0004] To achieve the above-mentioned purpose, a first aspect of an embodiment of the present application proposes a unit prediction state verification method, the method comprising:

[0005] Acquire a power unit model of the power unit combination; the power unit model includes a unit operation cost function and unit operation constraint conditions of the power unit combination;

[0006] Using the unit operation cost function and the unit operation constraint condition through a machine learning model, predicting the unit prediction state of each power unit in the power unit combination in each prediction period;

[0007] Acquiring an initial unit state of each power unit in the power unit combination in an initial period;

[0008] For each power unit, updating the unit prediction state of the power unit in the first prediction period according to the corresponding initial unit state to obtain the unit feasible state of the power unit in the first prediction period;

[0009] For each power unit, taking the second prediction period as the starting point, the unit prediction state of the power unit in the current prediction period is updated according to the unit feasible state of the power unit in the previous prediction period, and the unit feasible state of the power unit in the current prediction period is obtained, until the unit feasible state of the power unit in the last prediction period is obtained.

[0010] In some embodiments, the initial unit state includes an initial unit operating state and an initial unit output, the unit prediction state includes a predicted unit operating state and a predicted unit output, and the updating of the unit prediction state of the power unit in the first prediction period according to the corresponding initial unit state to obtain the unit feasible state of the power unit in the first prediction period includes:

[0011] The predicted unit operating state of the power unit in the first prediction period is updated according to the initial unit output to obtain the feasible unit operating state of the power unit in the first prediction period;

[0012] Creating a first intermediate unit output and a second intermediate unit output according to the initial unit operating state, the initial unit output, and the unit feasible operating state of the power unit in the first prediction period;

[0013] updating the predicted unit output of the power unit in the first prediction period according to the first intermediate unit output and the second intermediate unit output to obtain the unit feasible output of the power unit in the first prediction period;

[0014] The feasible state of the power unit in the first prediction period is determined according to the feasible operating state of the power unit in the first prediction period and the feasible output of the power unit in the first prediction period.

[0015] In some embodiments, updating the predicted unit operating state of the power unit in the first prediction period according to the initial unit output to obtain the feasible unit operating state of the power unit in the first prediction period includes:

[0016] comparing the initial unit output with the preset unit output;

[0017] If the initial unit output is greater than the preset unit output, the predicted unit operating state of the power unit in the first prediction period is updated to a first preset state, and the feasible unit operating state of the power unit in the first prediction period is obtained; the first preset state is used to indicate that the power unit is in an operating state;

[0018] If the initial unit output is less than or equal to the preset unit output, the unit continuous time state is determined according to the initial unit operating state and the predicted unit operating state of the power unit in the first prediction period; the predicted unit operating state of the power unit in the first prediction period is updated according to the unit continuous time state to obtain the feasible unit operating state of the power unit in the first prediction period.

[0019] In some embodiments, if the initial unit output is less than or equal to the preset unit output, determining the unit continuous time state according to the initial unit operating state and the predicted unit operating state of the power unit in the first prediction period includes:

[0020] If the initial unit output is less than or equal to the preset unit output, obtaining the difference between the predicted unit operating state of the power unit in the first prediction period and the initial unit operating state to obtain unit difference data;

[0021] If the unit difference data is equal to the first preset difference value, it is determined that the unit continuous time state is a second preset state; the second preset state is used to indicate that the power unit is in an offline state;

[0022] If the unit difference data is equal to a second preset difference value, it is determined that the unit continuous time state is a continuous online state; wherein the second preset difference value is greater than the first preset difference value;

[0023] If the unit difference data is equal to a third preset difference value, it is determined that the unit continuous time state is a continuous offline state; wherein the third preset difference value is smaller than the first preset difference value.

[0024] In some embodiments, updating the predicted unit operating state of the power unit in the first prediction period according to the unit continuous time state to obtain the unit feasible operating state of the power unit in the first prediction period includes:

[0025] If the continuous time state of the unit is the second preset state, rounding the predicted unit operation state of the power unit in the first prediction period is performed to obtain the feasible unit operation state of the power unit in the first prediction period;

[0026] If the continuous time state of the unit is a continuous online state, the predicted unit operation state of the power unit in the first prediction period is updated to a first preset state, and the feasible unit operation state of the power unit in the first prediction period is obtained;

[0027] If the continuous time state of the unit is a continuous offline state, the predicted unit operating state of the power unit in the first prediction period is updated to a second preset state to obtain the feasible unit operating state of the power unit in the first prediction period.

[0028] In some embodiments, creating a first intermediate unit output and a second intermediate unit output according to the initial unit operating state, the initial unit output, and the feasible unit operating state of the power unit in the first prediction period includes:

[0029] If the feasible operating state of the power unit in the first prediction period is the second preset state, a zero value is used as the first intermediate unit output and the second intermediate unit output respectively;

[0030] If the feasible operating state of the power unit in the first prediction period is the first preset state, and the initial unit operating state is the second preset state, the first intermediate unit output and the second intermediate unit output are determined according to the preset output lower limit value;

[0031] If the feasible operating state of the power unit in the first prediction period is the first preset state, and the initial unit operating state is the first preset state, the first intermediate unit output is determined according to the preset output lower limit value and the initial unit output, and the second intermediate unit output is determined according to the preset output upper limit value and the initial unit output.

[0032] In some embodiments, after updating the predicted unit operating state of the power unit in the first prediction period according to the unit continuous time state to obtain the unit feasible operating state of the power unit in the first prediction period, the method further includes:

[0033] If the unit continuous time state is the second preset state, updating the unit continuous time state according to the predicted unit operating state of the power unit in the first prediction period and the initial unit operating state;

[0034] If the continuous time state of the unit is a continuous online state or a continuous offline state, the continuous time state of the unit is updated according to a preset update step.

[0035] To achieve the above-mentioned purpose, a second aspect of an embodiment of the present application provides a unit prediction state verification device, the device comprising:

[0036] A first acquisition module is used to acquire a power unit model of the power unit combination; the power unit model includes a unit operation cost function and a unit operation constraint condition of the power unit combination;

[0037] A prediction module, used to predict the unit prediction state of each power unit in the power unit combination in each prediction period by using the unit operation cost function and the unit operation constraint condition through a machine learning model;

[0038] A second acquisition module is used to acquire the initial unit state of each power unit in the power unit combination in an initial period;

[0039] A first verification module is used to update the unit prediction state of each power unit in the first prediction period according to the corresponding initial unit state, so as to obtain the unit feasible state of the power unit in the first prediction period;

[0040] The second verification module is used to update the unit prediction state of the power unit in the current prediction period for each power unit, starting from the second prediction period, according to the unit feasible state of the power unit in the previous prediction period, to obtain the unit feasible state of the power unit in the current prediction period, until the unit feasible state of the power unit in the last prediction period is obtained.

[0041] To achieve the above-mentioned purpose, the third aspect of an embodiment of the present application proposes an electronic device, the electronic device includes a memory and a processor, the memory stores a computer program, and when the processor executes the computer program, the unit prediction state verification method of the above-mentioned first aspect is implemented.

[0042] To achieve the above-mentioned purpose, the fourth aspect of an embodiment of the present application proposes a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the unit prediction state verification method of the above-mentioned first aspect is implemented.

[0043] The unit prediction state verification method, unit prediction state verification device, electronic device, and computer-readable storage medium proposed in the present application obtain the power unit model of the power unit combination by modeling the unit combination problem. In order to accelerate the solution of the power unit model, a machine learning model is introduced. The power unit model is solved by the machine learning model using the unit operation cost function and the unit operation constraint conditions to obtain the unit prediction state of each power unit in the power unit combination in each prediction period. In order to improve the feasibility of the unit prediction state, for each power unit, the unit prediction state of the power unit in the first prediction period is updated according to the initial unit state to obtain the feasible unit state of the power unit in the first prediction period. The unit prediction states of the power unit in each prediction period are interdependent. In order to further improve the feasibility of the unit prediction state, taking the second prediction period as the starting point, the unit prediction state of the power unit in the current prediction period is updated according to the unit feasible state of the power unit in the previous prediction period, and the unit feasible state of the power unit in the current prediction period is obtained, until the unit feasible state of the power unit in the last prediction period is obtained, which ensures the feasibility of the unit prediction state of the power unit in all prediction periods, thereby improving the accuracy of solving the unit combination problem. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 is a flow chart of a unit prediction state verification method provided in an embodiment of the present application;

[0045] Figure 2 yes Figure 1 Flow chart of step S140 in FIG.

[0046] Figure 3 yes Figure 2 Flow chart of step S210 in FIG.

[0047] Figure 4 yes Figure 3 Flow chart of step S330 in FIG.

[0048] Figure 5 yes Figure 3 Another flow chart of step S330 in FIG.

[0049] Figure 6 yes Figure 2 Flow chart of step S220 in FIG.

[0050] Figure 7 is another flow chart of the unit prediction state verification method provided by an embodiment of the present application;

[0051] Figure 8 It is a structural schematic diagram of a unit prediction state verification device provided in an embodiment of the present application;

[0052] Fig. 9 It is a schematic diagram of the hardware structure of the electronic device provided in the embodiment of the present application. DETAILED DESCRIPTION

[0053] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0054] It should be noted that, although the functional modules are divided in the device schematic diagram and the logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first", "second", etc. in the specification, claims and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0055] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0056] The Unit Commitment Problem (UCP) is a key issue in power system dispatching. It is usually modeled as a mixed integer programming problem, and considers how to reasonably arrange the start and stop and output of power units, i.e., generators, in hours to optimally meet load demand and reduce unit operating costs. By solving the unit commitment problem, the operating efficiency of the power system can be optimized to ensure that the power system can operate in an efficient and economical manner.

[0057] With the increasing share of renewable energy such as wind power and solar energy, the growing uncertainty and load volatility have brought great challenges to the safe and stable operation of the power system. In order to enhance the flexibility and reliability of the power system, sub-hourly unit combination, that is, combined dispatching decisions with a time scale of 15 minutes or 30 minutes, is becoming the main dispatching decision model in the new power system to cope with the higher frequency of load and renewable energy supply fluctuations in the future. However, the decision frequency of sub-hourly unit combination problem is high, the time interval of decision is short, and the number of variables and constraints involved is large, resulting in a sharp increase in the scale of the problem. In addition, the relative reduction of the climbing ability of the power unit in each time period leads to the enhancement of the dependency between the coupled time periods, which further increases the complexity of the problem.

[0058] For the unit combination problem, the traditional method uses the operations optimization algorithm to solve it. This type of method has a relatively complete theoretical basis and proof reasoning, can ensure the convergence of the algorithm, and obtain the optimal solution or approximate optimal solution of the unit combination problem. However, due to the characteristics of combinatorial optimization problems, that is, the complexity of the problem increases exponentially with the increase of the problem scale, the use of operations optimization algorithms to solve large-scale sub-hourly unit combination problems will face defects such as long solution time and slow algorithm convergence, which cannot meet the actual operation and control requirements of the power system. Machine learning algorithms have brought new possibilities for the rapid solution of unit combination problems. After completing offline training, machine learning algorithms can provide prediction results in seconds or even milliseconds for online problems. Existing machine learning algorithms are usually used to deal with unconstrained classification or regression problems, and do not have the ability to solve problems with constraints. When machine learning algorithms are used to predict the results of problems with constraints, they often produce prediction values ​​that violate the problem constraints, causing the problem to fail to be solved, and it is necessary to use traditional operations optimization algorithms or heuristic methods to solve it again.

[0059] Machine learning algorithms often produce unit state predictions that violate the problem constraints. The existing commonly used method to improve the feasibility of the prediction value is to introduce a regularization term in the objective function, that is, to impose a certain penalty on the situation that violates the constraints of the unit combination problem. This method can improve the feasibility of the prediction value to a certain extent, but with the continuous increase in the scale of the unit combination problem and the increase in the dependence between the coupling cycles, this method cannot effectively improve the feasibility of the prediction value.

[0060] Based on this, the embodiments of the present application provide a unit prediction state verification method, a unit prediction state verification device, an electronic device and a computer-readable storage medium, which aim to improve the feasibility of the unit state prediction value when the scale of the unit combination problem is expanded and the dependence between coupling cycles is enhanced, and solve the problem that the machine learning algorithm cannot obtain a feasible solution or obtains a low proportion of feasible solutions when predicting the unit combination problem, thereby ensuring the rapid and efficient solution of the unit combination problem and supporting the safe and stable operation of the power system.

[0061] The unit prediction state verification method, unit prediction state verification device, electronic device and computer-readable storage medium provided in the embodiments of the present application are specifically explained through the following embodiments. First, the unit prediction state verification method in the embodiments of the present application is described.

[0062] The unit prediction state verification method provided in the embodiment of the present application relates to the technical field of power system. The unit prediction state verification method provided in the embodiment of the present application can be applied to the terminal, can also be applied to the server side, and can also be software running in the terminal or the server side. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, etc.; the server side can be configured as an independent physical server, or it can be configured as a server cluster or a distributed system composed of multiple physical servers, and can also be configured as a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application that implements the unit prediction state verification method, etc., but is not limited to the above forms.

[0063] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0064] Figure 1 is an optional flow chart of a unit prediction state verification method provided in an embodiment of the present application, Figure 1 The method may include but is not limited to steps S110 to S150.

[0065] Step S110, obtaining a power unit model of the power unit combination; the power unit model includes a unit operation cost function and unit operation constraint conditions of the power unit combination;

[0066] Step S120, predicting the unit prediction state of each power unit in the power unit combination in each prediction period by using the unit operation cost function and the unit operation constraint condition through the machine learning model;

[0067] Step S130, obtaining the initial unit state of each power unit in the power unit combination in the initial period;

[0068] Step S140, for each power unit, updating the unit prediction state of the power unit in the first prediction period according to the corresponding initial unit state, to obtain the unit feasible state of the power unit in the first prediction period;

[0069] Step S150, for each power unit, taking the second prediction period as the starting point, the unit prediction state of the power unit in the current prediction period is updated according to the unit feasible state of the power unit in the previous prediction period, and the unit feasible state of the power unit in the current prediction period is obtained, until the unit feasible state of the power unit in the last prediction period is obtained.

[0070] In step S110 of some embodiments, the unit combination problem is modeled as a mixed integer programming problem to obtain a power unit model of the power unit combination in the power system. The power unit combination is a combination of all power units in the power system, and the power unit combination includes multiple power units, and the power unit includes multiple power modules. The power unit model includes a unit operation cost function and unit operation constraints of the power unit combination. The unit operation cost function is the objective function of the unit combination problem, which can define the total operation cost of all power units, and solve the unit combination problem by minimizing the total operation cost. The unit operation cost function is expressed as:

[0071]

[0072] Wherein, i represents the i-th power unit; t represents the t-th prediction period, which is the period for which the state of the unit is to be solved; I represents the number of power units in the power unit combination; T represents the number of prediction periods; represents the initial startup cost of the i-th power unit in the t-th forecast period; represents the no-load cost of the i-th power unit in the t-th forecast period; b represents the b-th power module; B represents the number of power modules in the power unit; represents the power generation cost of the bth power module in the i-th power unit in the t-th forecast period; x i,t represents the operating status of the i-th power unit in the t-th prediction period, x i,t is a 0-1 variable, x i,t When it is 0, the power unit is in offline shutdown state; i,t When it is 1, the power unit is in online operation; i,b,t It represents the module output (power generation level) of the bth power module in the i-th power unit in the prediction period t.

[0073] The unit output of the i-th power unit in the forecast period t is the sum of the module outputs of all power modules in the i-th power unit in the forecast period t. The unit output is the power generation level of the power unit, that is, the output power. The unit output of the i-th power unit in the forecast period t is expressed as:

[0074]

[0075] Among them, p i,t It represents the unit output of the i-th power unit in the forecast period t.

[0076] Unit operation constraints are used to constrain the operation of the power system to ensure that the power system can operate safely, efficiently and reliably. Unit operation constraints include system-level constraints and unit-level constraints. System-level constraints include system demand constraints and transmission safety constraints. Unit-level constraints include unit capacity constraints, startup constraints, unit ramp rate constraints, minimum continuous startup constraints and minimum continuous shutdown constraints.

[0077] The system demand constraint is used to constrain the sum of the unit power generation levels to be equal to the sum of the node power demands, expressed as:

[0078]

[0079] Among them, D n,t represents the power demand of node n in the forecast period t; N is the number of nodes.

[0080] Transmission security constraints are used to constrain the power flow of each power grid transmission line l in each forecast period t to be within a certain safe transmission capacity F l and In , the transmission security constraint is expressed as:

[0081]

[0082] Among them, f t,l represents the DC flow of power grid transmission line l in the forecast period t; α n,l is the transfer factor, which indicates the influence of node n on the DC flow on the power grid transmission line l; F l and represents the transmission capacity constraint of the power grid transmission line l, F l Indicates the lower limit of transmission capacity, Represents the upper limit of transmission capacity; L is the number of grid transmission lines in the power system.

[0083] The unit capacity constraint is used to limit the power generation level of the generator set. The unit capacity constraint includes module output constraint, unit output and unit module processing constraint and unit output constraint. The module output constraint is the output constraint of each power module of the power unit, that is, the power generation level of each power module of the power unit cannot exceed the maximum power generation level of the power module. The module output constraint is expressed as:

[0084]

[0085] Among them, p i,b,t represents the power generation level (output) of the bth power module of the i-th power unit, Represents the maximum power generation level limit of the bth power module of the ith power unit.

[0086] The unit output and unit module processing constraints are used to constrain the relationship between the unit output of the power unit and the module output of the power module, that is, the power generation level of all power modules of the power unit is equal to the total power generation level of the power unit. The unit output and unit module processing constraints are expressed as:

[0087]

[0088] Among them, p i,t represents the output of the i-th power unit in the forecast period t; p i,b,t It represents the module output of the bth power module in the i-th power unit in the prediction period t.

[0089] The unit output constraint is used to constrain the unit output of the power unit. If the power unit has been started, that is, the power unit is in an online operation state, the power generation level of the power unit is subject to the minimum power generation level and the maximum power generation level. If the power unit is not started, that is, the power unit is in an offline shutdown state, the power generation level of the power unit is 0. The unit output constraint is expressed as:

[0090]

[0091] Among them, x i,t Indicates the running status of the unit. If the running status of the unit is 0, it means that the unit is not put into use. If the running status of the unit is 1, it means that the unit is put into use. i max Represents the maximum power generation level of power unit i.

[0092] The startup constraint is used to constrain the startup state of the power unit. The startup constraint is expressed as:

[0093] u i,t ≥x i,t -x i,t-1 ,

[0094] Among them, u i,t is a binary variable indicating the startup status of the unit, u i,t If it is 0, it means the power unit has not started. i,t is 1, indicating that the power unit has executed the startup action, if and only if the power unit is converted from the offline shutdown state to the startup state during the prediction period t i,t is 1.

[0095] The unit ramp rate constraint means that if the power unit is in online operation at time t and time t-1, the power generation level change of the power unit cannot exceed its ramp rate. The unit ramp rate constraint is expressed as:

[0096]

[0097] Among them, R i It represents the climbing rate of power unit i in normal operation; P i min Represents the minimum power generation level of power unit i.

[0098] When the power unit is in the switching state of starting or shutting down, according to its physical safety characteristics, the ramp rate of the power unit cannot exceed the minimum power generation level plus half of the normal ramp rate, that is,

[0099] The minimum continuous on / off constraint is used to limit the on / off time of the power unit. The power unit usually needs to remain online for a period of time after startup, or remain offline for a period of time after shutdown to avoid system stability degradation caused by frequent startup and shutdown. The minimum continuous on / off constraint specifies the minimum time that the power unit remains online / offline.

[0100] The minimum continuous startup constraints include the following three constraints. The first constraint is related to the initial state of the unit and is expressed as:

[0101]

[0102] Among them, T MOn Indicates the number of time periods during which the unit has been turned on and must remain on during the initial phase t.

[0103] The power unit needs to maintain a minimum continuous on-time in all possible continuous time periods, that is, if the power unit is turned on, the power unit must remain online for a continuous period of time. The second constraint is expressed as:

[0104]

[0105] Among them, T MU Indicates the minimum continuous online time of the power unit.

[0106] At a specific time period t, if the power unit is started, it will remain online until the last forecast period. The third constraint is expressed as:

[0107]

[0108] Similar to the minimum continuous on constraint, the minimum continuous off constraint includes:

[0109]

[0110] Among them, T MO ff represents the number of periods during which the power unit is in the off state at the start of phase t and must continue to remain in the off state.

[0111]

[0112] Among them, T MD Indicates the minimum continuous offline time of the power unit.

[0113]

[0114] In step S120 of some embodiments, traditional optimization methods such as branch and bound method, benders decomposition method and Lagrangian relaxation method are mainstream methods for solving unit combination problems. As the mainstream exact solution algorithm for mixed integer programming problems, traditional optimization methods can ensure algorithm convergence and obtain high-quality, feasible and accurate optimal solutions. However, as the scale of the problem increases, obtaining the optimal solution of large-scale unit combination problems within the specified time limit faces new challenges. In order to improve the efficiency of solving large-scale unit combination problems, various machine learning models such as k nearest neighbor, support vector machine, random forest, graph convolution network, reinforcement learning, time series prediction model, etc. are used. The time series prediction model can be a gated recurrent unit, a long short-term memory network, etc., based on the unit operation cost function and the unit operation constraint conditions, the optimal solution of the unit combination is predicted, and the unit prediction state of each power unit in the power unit combination in each prediction period is obtained. The unit prediction state includes the predicted unit operation state and the predicted unit output. The predicted unit operation state is the predicted unit operation state, and the predicted unit output is the predicted unit output. The predicted unit operating status is expressed as The predicted unit output is expressed as

[0115] In step S130 of some embodiments, the machine learning model is usually used to process unconstrained problems such as weather forecasting. However, in the constrained unit combination problem, the machine learning model cannot consider the various constraints in the optimization problem, and will produce prediction values ​​that violate the problem constraints. When the problem scale increases, it is often difficult to stably provide high-quality predictions, and it is necessary to re-call the traditional optimization method to solve, which reduces the solution efficiency. In order to improve the efficiency of unit state prediction and improve the feasibility of the unit state prediction value, the embodiment of the present application adds a rule-based feasibility verification layer after the machine learning model, and implements the feasibility verification layer in a vectorized manner to reduce processing time and ensure computational efficiency. The prediction solution is adjusted multiple times through the feasibility verification layer, so that the feasibility of the prediction solution is greatly improved. Specifically, the initial unit state of each power unit in the power unit combination in the initial period is obtained. The initial period is the period before the prediction period, and the initial period can be set to 0. The initial unit state includes the initial unit operating state and the initial unit output. The initial unit operating state is the unit operating state in the initial period, expressed as x i,0 , the initial unit output is the unit output in the initial period, expressed as p i,0.

[0116] See also Figure 2 In some embodiments, step S140 may include but is not limited to steps S210 to S240:

[0117] Step S210, updating the predicted unit operating state of the power unit in the first prediction period according to the initial unit output, and obtaining the unit feasible operating state of the power unit in the first prediction period;

[0118] Step S220, creating a first intermediate unit output and a second intermediate unit output according to the initial unit operating state, the initial unit output and the feasible unit operating state of the power unit in the first prediction period;

[0119] Step S230, updating the predicted unit output of the power unit in the first prediction period according to the first intermediate unit output and the second intermediate unit output, to obtain the unit feasible output of the power unit in the first prediction period;

[0120] Step S240, determining the feasible state of the power unit in the first prediction period according to the feasible operation state of the power unit in the first prediction period and the feasible output of the power unit in the first prediction period.

[0121] In step S210 of some embodiments, in order to improve the feasibility of the machine learning model to solve the unit combination problem and obtain the predicted value, the embodiment of the present application introduces a feasibility verification mechanism for the unit combination predicted value based on physical rules, verifies and corrects the unit level constraints, and obtains a predicted solution that meets the unit level constraints, aiming to increase the proportion of feasible solutions for the unit combination, reduce the possibility of infeasible solutions, improve the prediction quality in the machine learning process, and greatly reduce the trial and error cost to support the safe and stable operation of the power system. The feasibility verification mechanism updates the unit prediction state according to the unit level constraints. For each power unit, according to the initial unit output p of the power unit i,0 Predicted unit operating status of the power unit in the first prediction period Update and obtain the feasible operating state x of the power unit in the first prediction period i,1 The feasible operating state of the unit is the operating state of the unit after adjustment by the feasibility verification mechanism.

[0122] In step S220 of some embodiments, according to the initial unit operation state x i,0 , initial unit output p i,0 and the feasible operating state x of the power unit in the first prediction period i,1Two unit outputs are created to obtain a first intermediate unit output and a second intermediate unit output. The first intermediate unit output is less than or equal to the second intermediate unit output. The first intermediate unit output is the lower limit of the unit output expected by the power unit, and the second intermediate unit output is the upper limit of the unit output expected by the power unit.

[0123] In step S230 of some embodiments, the machine learning model estimates the unit state based on historical data, and it is impossible to ensure that the unit state meets the constraints during the prediction process. In order to ensure the feasibility of predicting the unit output, the predicted unit output of the power unit in the first prediction period is Perform rounding operation to obtain the rounded unit output of the power unit in the first forecast period Rounding operation is rounding operation, according to the output of the first intermediate unit and the second intermediate unit output The rounded unit output of the power unit in the first forecast period Update and obtain the feasible output p of the power unit in the first forecast period i,1 The feasible output of the unit is the unit output after adjustment by the feasibility verification mechanism. The calculation formula of the feasible output of the unit is expressed as:

[0124]

[0125] Among them, pi ,t represents the feasible output of power unit i in the forecast period t; represents the output of the first intermediate unit of power unit i; Represents the output of the second intermediate unit of power unit i.

[0126] In step S240 of some embodiments, the feasible operating state x of the power unit in the first prediction period is i,1 and the feasible output p of the power unit in the first forecast period i,1 As the feasible state of the power unit in the first forecast period, the feasible state of the unit is the forecast state of the unit after adjustment by the feasibility verification mechanism.

[0127] Through the above steps S210 to S240, the update of the unit state prediction value in the first prediction stage can be completed, which not only improves the efficiency of solving the unit combination problem, but also significantly improves the feasibility of the unit state prediction value and reduces the additional calculation caused by solution failure.

[0128] See also Figure 3 In some embodiments, step S210 may include but is not limited to steps S310 to S330:

[0129] Step S310, comparing the initial unit output with the preset unit output;

[0130] Step S320, if the initial unit output is greater than the preset unit output, the predicted unit operating state of the power unit in the first prediction period is updated to a first preset state, and the unit feasible operating state of the power unit in the first prediction period is obtained; the first preset state is used to indicate that the power unit is in an operating state;

[0131] Step S330, if the initial unit output is less than or equal to the preset unit output, determine the unit continuous time state according to the initial unit operating state and the predicted unit operating state of the power unit in the first prediction period; update the predicted unit operating state of the power unit in the first prediction period according to the unit continuous time state to obtain the feasible unit operating state of the power unit in the first prediction period.

[0132] In step S310 of some embodiments, the initial unit output and the preset unit output are compared, and the update method of the predicted unit operating state is determined according to the size relationship between the initial unit output and the preset unit output. The preset unit output is the weighted sum of the minimum power generation level and the ramp rate of the power unit. The preset unit output is defined as represents the minimum power generation level of power unit i, R i It represents the climbing rate of power unit i during normal operation.

[0133] In step S320 of some embodiments, if the initial unit output is greater than the preset unit output, that is, Then the predicted unit operating state of the power unit in the first prediction period is updated to the first preset state, and the feasible unit operating state of the power unit in the first prediction period is obtained. The first preset state is used to indicate that the power unit is in an operating state, that is, x i,1 =1.

[0134] In step S330 of some embodiments, if the initial unit output is less than or equal to the preset unit output, that is, The unit continuous time state is determined according to the initial unit operating state and the predicted unit operating state of the power unit in the first prediction period. The unit continuous time state can be used to determine whether the power unit is in the minimum continuous on / off state. The predicted unit operating state of the power unit in the first prediction period is updated according to the unit continuous time state so that the predicted unit operating state meets the minimum continuous on / off constraint, and the unit feasible operating state of the power unit in the first prediction period is obtained.

[0135] Through the above steps S310 to S330, by modifying the predicted unit operating state, the feasible operating state of the power unit in the first prediction period can meet the unit-level constraints, thereby improving the feasibility of the predicted value of the unit operating state.

[0136] See also Figure 4 In some embodiments, step S330 may include but is not limited to steps S410 to S440:

[0137] Step S410, if the initial unit output is less than or equal to the preset unit output, obtaining the difference between the predicted unit operating state of the power unit in the first prediction period and the initial unit operating state to obtain unit difference data;

[0138] Step S420, if the unit difference data is equal to the first preset difference value, then determine that the unit continuous time state is a second preset state; the second preset state is used to indicate that the power unit is in an offline state;

[0139] Step S430, if the unit difference data is equal to the second preset difference value, determining that the unit continuous time state is a continuous online state; wherein the second preset difference value is greater than the first preset difference value;

[0140] Step S440: If the unit difference data is equal to a third preset difference value, it is determined that the unit continuous time state is a continuous offline state; wherein the third preset difference value is smaller than the first preset difference value.

[0141] In step S410 of some embodiments, if the initial unit output is less than or equal to the preset unit output, the predicted unit operating state of the power unit in the first prediction period is Perform rounding operations to obtain the rounded unit operating status of the power unit in the first forecast period The running status of the whole unit With the initial unit operating state x i,0 Subtracting, we get the unit difference data as

[0142] It should be noted that the machine learning model can only obtain continuous value outputs, while the unit operating status is a 0-1 variable. The predicted unit operating status can be converted into a 0-1 variable through rounding operations.

[0143] In step S420 of some embodiments, the first preset difference value is set to 0, and the continuous time state of the power unit i is represented by m i If the unit difference data is equal to the first preset difference value, the unit continuous time state is determined to be the second preset state, that is, m i =0, the second preset state is used to indicate that the power unit is in an offline state.

[0144] In step S430 of some embodiments, the second preset difference value is set to 1. If the unit difference data is equal to the second preset difference value, it is determined that the unit continuous time state is a continuous online state (minimum continuous on state), that is, m i =T MU .

[0145] In step S440 of some embodiments, the third preset difference value is set to -1. If the unit difference data is equal to the third preset difference value, it is determined that the unit continuous time state is a continuous offline state (minimum continuous shutdown state), that is, m i =-T MD .

[0146] Through the above steps S410 to S440, it is possible to determine whether the power unit is in the minimum continuous on / off state, so as to adjust the predicted unit operation state according to the minimum continuous on / off state, thereby improving the feasibility of solving the unit combination problem.

[0147] See also Figure 5 In some embodiments, step S330 may include but is not limited to step S510, step S520 or step S530:

[0148] Step S510, if the continuous time state of the unit is the second preset state, rounding the predicted unit operation state of the power unit in the first prediction period to obtain the feasible unit operation state of the power unit in the first prediction period;

[0149] Step S520, if the continuous time state of the unit is a continuous online state, then the predicted unit operation state of the power unit in the first prediction period is updated to a first preset state, and the feasible unit operation state of the power unit in the first prediction period is obtained;

[0150] Step S530, if the continuous time state of the unit is a continuous offline state, the predicted unit operating state of the power unit in the first prediction period is updated to a second preset state to obtain the feasible unit operating state of the power unit in the first prediction period.

[0151] In step S510 of some embodiments, if the continuous time state of the unit is the second preset state, that is, m i = 0, then the rounded unit operating status of the power unit in the first forecast period Perform rounding operations to obtain the feasible operating status of the power unit in the first forecast period.

[0152] In step S520 of some embodiments, if the continuous time state of the unit is a continuous online state, that is, m i>0, the predicted unit operating state of the power unit in the first prediction period is updated to the first preset state, and the feasible unit operating state of the power unit in the first prediction period is obtained, that is, x i,1 =1.

[0153] In step S530 of some embodiments, if the unit is in a continuous offline state for a continuous period of time, that is, m i <0, the predicted unit operating state of the power unit in the first prediction period is updated to the second preset state, and the feasible unit operating state of the power unit in the first prediction period is obtained, that is, x i,1 =0.

[0154] Through the above steps S510 to S530, the predicted value of the unit operating state can meet various constraints, thereby improving the feasibility of the unit state prediction.

[0155] See also Figure 6 In some embodiments, step S220 may include but is not limited to step S610, step S620 or step S630:

[0156] Step S610, if the feasible operating state of the power unit in the first prediction period is the second preset state, a zero value is used as the first intermediate unit output and the second intermediate unit output respectively;

[0157] Step S620, if the feasible operating state of the power unit in the first prediction period is the first preset state, and the initial unit operating state is the second preset state, then determine the first intermediate unit output and the second intermediate unit output according to the preset output lower limit value;

[0158] Step S630, if the feasible operating state of the power unit in the first prediction period is the first preset state, and the initial unit operating state is the first preset state, then the first intermediate unit output is determined according to the preset output lower limit value and the initial unit output, and the second intermediate unit output is determined according to the preset output upper limit value and the initial unit output.

[0159] In step S610 of some embodiments, if the feasible operating state of the power unit in the first prediction period is the second preset state, that is, x i,1 =0, then the zero value is used as the output of the first intermediate unit and the output of the second intermediate unit, that is,

[0160] In step S620 of some embodiments, if the feasible operating state of the power unit in the first prediction period is the first preset state, and the initial unit operating state is the second preset state, that is, x i,1 =1 and x i,0=0, the preset lower limit of the power unit output (minimum power generation level) is used as the first intermediate unit output, and the weighted sum of the preset lower limit of the power unit output and the ramp rate (preset unit output) is used as the second intermediate unit output. That is, the output of the first intermediate unit is The output of the second intermediate unit is

[0161] In step S630 of some embodiments, if the feasible operating state of the power unit in the first prediction period is the first preset state, and the initial unit operating state is the first preset state, that is, x i,1 =1 and x i,0 =1, the output of the first intermediate unit is determined according to the preset output lower limit value, the initial unit output and the climbing rate of the power unit, and the output of the second intermediate unit is determined according to the preset output upper limit value, the initial unit output and the climbing rate of the power unit. The preset output lower limit value and the preset output upper limit value are the actual unit output lower limit value and unit output upper limit value of the power unit. The preset output upper limit value is the maximum power generation level of the power unit. The output of the first intermediate unit and the output of the second intermediate unit are respectively expressed as:

[0162]

[0163] Among them, p i,0 Indicates the initial unit output; Indicates the preset lower output limit; Indicates the preset output upper limit; R i Indicates the climbing rate.

[0164] Through the above steps S610 to S630, the first intermediate unit output and the second intermediate unit output can be determined, so as to adjust the unit output prediction value based on the first intermediate unit output and the second intermediate unit output, thereby improving the solution feasibility.

[0165] See also Figure 7 In some embodiments, after step S330, the unit prediction state verification method may also include but is not limited to step S710 or step S720:

[0166] Step S710, if the unit continuous time state is the second preset state, updating the unit continuous time state according to the predicted unit operation state of the power unit in the first prediction period and the initial unit operation state;

[0167] Step S720: If the continuous time state of the unit is a continuous online state or a continuous offline state, the continuous time state of the unit is updated according to a preset update step.

[0168] In step S710 of some embodiments, if the continuous time state of the unit is the second preset state, that is, mi =0, then for the initial unit operating state x i,0 Perform rounding operation to obtain The rounded unit operating status of the power unit in the first forecast period and Subtract, and update the unit continuous time state according to the subtraction result. The updating method is as follows:

[0169]

[0170] in, represents the rounded unit operating state of the power unit in the first forecast period, which is obtained by rounding the predicted unit operating state of the power unit in the first forecast period; Indicates the initial unit operating status after rounding operation.

[0171] In step S720 of some embodiments, the preset update step size may be 1. If the continuous time state of the unit is a continuous online state, that is, m i >0, the unit continuous time state is decremented according to the preset update step. If the unit continuous time state is continuous offline state, that is, m i <0, the unit continuous time state is incremented according to the preset update step. The update method is as follows:

[0172]

[0173] In the above steps S710 to S720, the continuous time state of the unit is updated to adjust the predicted value of the unit operation state for the next prediction period according to the updated continuous time state of the unit, thereby ensuring the feasibility of the predicted value of the unit operation state for the next prediction period.

[0174] In step S150 of some embodiments, for each power unit, starting from the second prediction period, the feasible output p of the power unit in the previous prediction period is compared. i,t-1 and the preset unit output, if the power unit has the feasible output p in the previous forecast period i,t-1 If the output of the power unit is greater than the preset unit output, the predicted unit operation state of the power unit in the current prediction period t is updated to the first preset state, that is, x i,t =1, the feasible operating state of the power unit in the current forecast period t is obtained. If the feasible output of the power unit in the previous forecast period p i,t-1If the output of the power unit is less than or equal to the preset unit output, the predicted unit operation state of the power unit in the current prediction period t is updated according to the unit continuous time state updated in the previous prediction period, and the unit feasible operation state of the power unit in the current prediction period t is obtained. The method of updating the predicted unit operation state using the unit continuous time state is as follows:

[0175]

[0176] Among them, m i The continuous time status of the unit after update in the previous forecast period; It is the predicted unit operating status of the power unit in the current prediction period t.

[0177] Referring to step S710 to step S720, the updated continuous time state of the unit in the previous prediction period is updated to obtain the updated continuous time state of the unit in the current prediction period.

[0178] According to the feasible operating status x of the power unit in the current forecast period i,t , the feasible operating status of the unit in the previous forecast period x i,t-1 , the feasible output of the unit in the previous forecast period p i,t-1 , climbing rate, preset lower output limit value and preset upper output limit value, create the first intermediate unit output and the second intermediate unit output of the current forecast period. The method for creating the first intermediate unit output and the second intermediate unit output is expressed as:

[0179]

[0180] in, Indicates the preset lower output limit; Indicates the preset output upper limit; R i Indicates the climbing rate.

[0181] Referring to step S230, the predicted unit output of the power unit in the current prediction period is updated according to the first intermediate unit output and the second intermediate unit output, and the unit feasible output of the power unit in the current prediction period is obtained. Repeat step S150 until the unit feasible operating state and unit feasible output of the power unit in the last prediction period are obtained, so that the prediction solutions of all power units in each prediction period meet all unit level constraints, improve the accuracy and feasibility of the prediction of the unit combination problem, solve the problem of no feasible solution or low feasible solution ratio in the prediction value, and reduce additional calculations.

[0182] See also Figure 8 The embodiment of the present application further provides a unit prediction state verification device, which can implement the above-mentioned unit prediction state verification method, and the unit prediction state verification device includes:

[0183] A first acquisition module 810 is used to acquire a power unit model of the power unit combination; the power unit model includes a unit operation cost function and a unit operation constraint condition of the power unit combination;

[0184] A prediction module 820, for predicting the unit prediction state of each power unit in the power unit combination in each prediction period according to the unit operation cost function and the unit operation constraint condition;

[0185] A second acquisition module 830 is used to acquire the initial unit state of each power unit in the power unit combination in the initial period;

[0186] A first verification module 840 is used to update the unit prediction state of the power unit in the first prediction period according to the corresponding initial unit state for each power unit, so as to obtain the unit feasible state of the power unit in the first prediction period;

[0187] The second verification module 850 is used to update the unit prediction state of the power unit in the current prediction period for each power unit, starting from the second prediction period, according to the unit feasible state of the power unit in the previous prediction period, to obtain the unit feasible state of the power unit in the current prediction period, until the unit feasible state of the power unit in the last prediction period is obtained.

[0188] The specific implementation of the unit prediction state verification device is basically the same as the specific implementation of the above-mentioned unit prediction state verification method, and will not be repeated here.

[0189] The embodiment of the present application also provides an electronic device, the electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the above-mentioned unit prediction state verification method when executing the computer program. The electronic device can be any intelligent terminal including a tablet computer, a car computer, etc.

[0190] See also Fig. 9 , Fig. 9 The hardware structure of an electronic device of another embodiment is illustrated, and the electronic device includes:

[0191] The processor 910 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application;

[0192] The memory 920 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 920 can store an operating system and other application programs. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 920, and the processor 910 calls and executes the unit prediction state verification method of the embodiment of this application;

[0193] Input / output interface 930, used to implement information input and output;

[0194] Communication interface 940, used to realize communication interaction between the device and other devices, which can be realized by wired mode (such as USB, network cable, etc.) or wireless mode (such as mobile network, WIFI, Bluetooth, etc.);

[0195] bus 950 , which transmits information between the various components of the device (e.g., processor 910 , memory 920 , input / output interface 930 , and communication interface 940 );

[0196] The processor 910 , the memory 920 , the input / output interface 930 , and the communication interface 940 are connected to each other in communication within the device via a bus 950 .

[0197] An embodiment of the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the above-mentioned unit prediction state verification method is implemented.

[0198] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0199] The embodiments described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0200] Those skilled in the art will appreciate that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.

[0201] The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0202] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0203] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0204] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0205] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the above units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. The mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0206] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0207] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0208] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including multiple instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store programs.

[0209] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but the scope of the rights of the present invention is not limited thereto. Any modification, equivalent substitution and improvement made by a person skilled in the art without departing from the scope and essence of the present invention should be within the scope of the rights of the present invention.

Claims

1. A method for verifying the predicted state of a unit, characterized in that: The method comprises: Acquire a power unit model of the power unit combination; the power unit model includes a unit operation cost function and unit operation constraint conditions of the power unit combination; Using the unit operation cost function and the unit operation constraint condition through a machine learning model, predicting the unit prediction state of each power unit in the power unit combination in each prediction period; Acquiring an initial unit state of each power unit in the power unit combination in an initial period; For each power unit, updating the unit prediction state of the power unit in the first prediction period according to the corresponding initial unit state to obtain the unit feasible state of the power unit in the first prediction period; For each power unit, taking the second prediction period as the starting point, the unit prediction state of the power unit in the current prediction period is updated according to the unit feasible state of the power unit in the previous prediction period, and the unit feasible state of the power unit in the current prediction period is obtained, until the unit feasible state of the power unit in the last prediction period is obtained.

2. The method according to claim 1, characterized in that The initial unit state includes an initial unit operating state and an initial unit output, the unit prediction state includes a predicted unit operating state and a predicted unit output, and the updating of the unit prediction state of the power unit in the first prediction period according to the corresponding initial unit state to obtain the unit feasible state of the power unit in the first prediction period includes: The predicted unit operating state of the power unit in the first prediction period is updated according to the initial unit output to obtain the feasible unit operating state of the power unit in the first prediction period; Creating a first intermediate unit output and a second intermediate unit output according to the initial unit operating state, the initial unit output, and the unit feasible operating state of the power unit in the first prediction period; updating the predicted unit output of the power unit in the first prediction period according to the first intermediate unit output and the second intermediate unit output to obtain the unit feasible output of the power unit in the first prediction period; The feasible state of the power unit in the first prediction period is determined according to the feasible operating state of the power unit in the first prediction period and the feasible output of the power unit in the first prediction period.

3. The method according to claim 2, characterized in that The updating of the predicted unit operating state of the power unit in the first prediction period according to the initial unit output to obtain the feasible unit operating state of the power unit in the first prediction period includes: comparing the initial unit output with the preset unit output; If the initial unit output is greater than the preset unit output, the predicted unit operating state of the power unit in the first prediction period is updated to a first preset state, and the feasible unit operating state of the power unit in the first prediction period is obtained; the first preset state is used to indicate that the power unit is in an operating state; If the initial unit output is less than or equal to the preset unit output, the unit continuous time state is determined according to the initial unit operating state and the predicted unit operating state of the power unit in the first prediction period; the predicted unit operating state of the power unit in the first prediction period is updated according to the unit continuous time state to obtain the feasible unit operating state of the power unit in the first prediction period.

4. The method according to claim 3, characterized in that If the initial unit output is less than or equal to the preset unit output, determining the unit continuous time state according to the initial unit operating state and the predicted unit operating state of the power unit in the first prediction period, including: If the initial unit output is less than or equal to the preset unit output, obtaining the difference between the predicted unit operating state of the power unit in the first prediction period and the initial unit operating state to obtain unit difference data; If the unit difference data is equal to the first preset difference value, it is determined that the unit continuous time state is a second preset state; the second preset state is used to indicate that the power unit is in an offline state; If the unit difference data is equal to a second preset difference value, it is determined that the unit continuous time state is a continuous online state; wherein the second preset difference value is greater than the first preset difference value; If the unit difference data is equal to a third preset difference value, it is determined that the unit continuous time state is a continuous offline state; wherein the third preset difference value is smaller than the first preset difference value.

5. The method according to claim 3, characterized in that: The updating of the predicted unit operation state of the power unit in the first prediction period according to the unit continuous time state to obtain the unit feasible operation state of the power unit in the first prediction period includes: If the continuous time state of the unit is the second preset state, rounding the predicted unit operation state of the power unit in the first prediction period is performed to obtain the feasible unit operation state of the power unit in the first prediction period; If the continuous time state of the unit is a continuous online state, the predicted unit operation state of the power unit in the first prediction period is updated to a first preset state, and the feasible unit operation state of the power unit in the first prediction period is obtained; If the continuous time state of the unit is a continuous offline state, the predicted unit operating state of the power unit in the first prediction period is updated to a second preset state to obtain the feasible unit operating state of the power unit in the first prediction period.

6. The method according to claim 2, characterized in that The step of creating a first intermediate unit output and a second intermediate unit output according to the initial unit operating state, the initial unit output, and the feasible unit operating state of the power unit in the first prediction period includes: If the feasible operating state of the power unit in the first prediction period is the second preset state, a zero value is used as the first intermediate unit output and the second intermediate unit output respectively; If the feasible operating state of the power unit in the first prediction period is the first preset state, and the initial unit operating state is the second preset state, the first intermediate unit output and the second intermediate unit output are determined according to the preset output lower limit value; If the feasible operating state of the power unit in the first prediction period is the first preset state, and the initial unit operating state is the first preset state, the first intermediate unit output is determined according to the preset output lower limit value and the initial unit output, and the second intermediate unit output is determined according to the preset output upper limit value and the initial unit output.

7. The method according to claim 3, characterized in that After updating the predicted unit operating state of the power unit in the first prediction period according to the unit continuous time state to obtain the unit feasible operating state of the power unit in the first prediction period, the method further includes: If the unit continuous time state is the second preset state, updating the unit continuous time state according to the predicted unit operating state of the power unit in the first prediction period and the initial unit operating state; If the continuous time state of the unit is a continuous online state or a continuous offline state, the continuous time state of the unit is updated according to a preset update step.

8. A unit prediction state verification device, characterized in that: The device comprises: A first acquisition module is used to acquire a power unit model of the power unit combination; the power unit model includes a unit operation cost function and a unit operation constraint condition of the power unit combination; A prediction module, used to predict the unit prediction state of each power unit in the power unit combination in each prediction period by using the unit operation cost function and the unit operation constraint condition through a machine learning model; A second acquisition module is used to acquire the initial unit state of each power unit in the power unit combination in an initial period; A first verification module is used to update the unit prediction state of each power unit in the first prediction period according to the corresponding initial unit state, so as to obtain the unit feasible state of the power unit in the first prediction period; The second verification module is used to update the unit prediction state of the power unit in the current prediction period for each power unit, starting from the second prediction period, according to the unit feasible state of the power unit in the previous prediction period, to obtain the unit feasible state of the power unit in the current prediction period, until the unit feasible state of the power unit in the last prediction period is obtained.

9. An electronic device, characterized in that: The electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the unit prediction state verification method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the unit prediction state verification method according to any one of claims 1 to 7 is implemented.

Images